item_type,number,title,url,primary_bucket,attention_score,attention_reason,maintainer_next_action,updated_at
issue,188812,[MPS] MetalShaderLibrary::getBundledLibrary singleton can crash at process exit,https://github.com/pytorch/pytorch/issues/188812,needs triage,151,"triage signal present; open PR references this issue; high-priority/blocker/regression label present; needs reproduction label present; component/oncall labels: module: crash, module: flaky-tests, module: macos, module: mps",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-07-04T18:31:40Z
issue,188900,`sparse.mul`: broadcasting a size-1 dimension of the sparse operand silently drops data,https://github.com/pytorch/pytorch/issues/188900,needs triage,139,"triage signal present; open PR references this issue; high-priority/blocker/regression label present; component/oncall labels: module: correctness (silent), module: sparse",review early; this can change release or regression risk,2026-07-06T17:28:34Z
issue,188323,[Inductor][CPU] dynamic=True convolution lowering crashes with ValueError: Exponent must be non-negative,https://github.com/pytorch/pytorch/issues/188323,needs triage,139,"triage signal present; open PR references this issue; high-priority/blocker/regression label present; component/oncall labels: module: convolution, module: cpu, module: crash, module: dynamic shapes",review early; this can change release or regression risk,2026-06-30T17:57:26Z
issue,186535,"Windows, gloo: Access violation (0xC0000005) in ProcessGroupGloo::enqueue when calling allreduce on CUDA tensors — GlooAllreduceRegistry has no kCUDA creator",https://github.com/pytorch/pytorch/issues/186535,needs triage,139,"triage signal present; open PR references this issue; high-priority/blocker/regression label present; component/oncall labels: module: crash, module: cuda, module: multi-gpu, module: windows",review early; this can change release or regression risk,2026-06-07T10:58:03Z
issue,187912,[CPU] Concurrent `cpublas::brgemm` calls can crash in the AMX path when the underlying oneDNN ukernel is shared,https://github.com/pytorch/pytorch/issues/187912,needs triage,127,"triage signal present; high-priority/blocker/regression label present; needs reproduction label present; component/oncall labels: module: cpu, module: crash, module: intel, module: linear algebra",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-07-06T17:07:04Z
issue,157668,NCCL error caused due to use of NVLS in torch 2.7.1-cu128 on aarch64 gb200 cluster,https://github.com/pytorch/pytorch/issues/157668,needs triage,127,"triage signal present; high-priority/blocker/regression label present; needs reproduction label present; component/oncall labels: module: arm, module: cuda, module: nccl, oncall: distributed",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-06-09T17:07:16Z
issue,150022,Dynamic Shapes with **kwargs,https://github.com/pytorch/pytorch/issues/150022,needs triage,121,"PyTorch issue lacks triaged label; open PR references this issue; high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: regression, oncall: export, oncall: pt2",review early; this can change release or regression risk,2025-12-04T07:28:20Z
issue,162422,Runtime failure when running torch.compile() & using GCC 11.5.0 on Neoverse V1,https://github.com/pytorch/pytorch/issues/162422,needs triage,121,"PyTorch issue lacks triaged label; open PR references this issue; high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: arm, module: crash, module: third_party, module: vectorization",review early; this can change release or regression risk,2025-11-30T04:27:08Z
issue,162476,heap-buffer-overflow in torch.quantized_max_pool2d via Python API,https://github.com/pytorch/pytorch/issues/162476,needs triage,121,"PyTorch issue lacks triaged label; open PR references this issue; high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, oncall: quantization",review early; this can change release or regression risk,2025-09-11T15:33:20Z
issue,189194,"[ROCm] torch 2.13 wheel: ""Can't detect vectorized ISA for CPU"" in torch.compile smoke test on non-ROCm image (regression vs 2.12.1)",https://github.com/pytorch/pytorch/issues/189194,needs triage,115,"triage signal present; high-priority/blocker/regression label present; component/oncall labels: module: binaries, module: inductor, module: regression, module: rocm",review early; this can change release or regression risk,2026-07-08T22:32:08Z
issue,189150,SIGSEGV / cudaErrorIllegalAddress when replaying a CUDA graph that captured multiple training iterations of two trainers over deeply unrolled recurrent modules,https://github.com/pytorch/pytorch/issues/189150,needs triage,115,"triage signal present; high-priority/blocker/regression label present; component/oncall labels: module: crash, module: cuda, module: cuda graphs",review early; this can change release or regression risk,2026-07-08T19:25:06Z
issue,189282,"[CPU] operator_benchmark: backward-pass (add, + batchnorm on aarch64) ~200-500x slower since ~Nov 2025 (x86 + aarch64; possible measurement artifact)",https://github.com/pytorch/pytorch/issues/189282,needs triage,115,"triage signal present; high-priority/blocker/regression label present; component/oncall labels: module: autograd, module: cpu, module: performance, module: regression",review early; this can change release or regression risk,2026-07-08T17:39:26Z
issue,189281,[CPU] operator_benchmark: embedding / embeddingbag ~70-130x slower since ~Nov 2025 (x86 + aarch64; possible measurement issue),https://github.com/pytorch/pytorch/issues/189281,needs triage,115,"triage signal present; high-priority/blocker/regression label present; component/oncall labels: module: cpu, module: embedding, module: performance, module: regression",review early; this can change release or regression risk,2026-07-08T17:39:24Z
issue,189239,[Inductor]Unable to apply layout optimization on convolution backward causes training performance regression.,https://github.com/pytorch/pytorch/issues/189239,needs triage,115,"triage signal present; high-priority/blocker/regression label present; component/oncall labels: module: convolution, module: inductor, module: memory format, module: performance",review early; this can change release or regression risk,2026-07-08T06:33:50Z
issue,189144,[XPU][B580] `flex-attn-causal` performance drop with `last_level_cache_size` cache clear between runs,https://github.com/pytorch/pytorch/issues/189144,needs triage,115,"triage signal present; high-priority/blocker/regression label present; component/oncall labels: module: flex attention, module: inductor, module: performance, module: regression",review early; this can change release or regression risk,2026-07-08T02:27:43Z
issue,144965,"RuntimeError ""global alloc not supported yet"" when using TorchScript optimization.",https://github.com/pytorch/pytorch/issues/144965,needs triage,115,triage signal present; high-priority/blocker/regression label present; component/oncall labels: oncall: jit,review early; this can change release or regression risk,2026-07-07T06:52:38Z
issue,189126,"Report a bug, as requested by the error message",https://github.com/pytorch/pytorch/issues/189126,needs triage,115,"triage signal present; high-priority/blocker/regression label present; component/oncall labels: module: crash, module: cuda, module: memory usage",review early; this can change release or regression risk,2026-07-07T05:17:55Z
issue,189065,"CI for distributed tests on B200 fully red since ~2026-07-01: rank 0 ""CUDA-capable device(s) busy or unavailable"" -> NCCL id timeout -> 1170-min job timeout",https://github.com/pytorch/pytorch/issues/189065,needs triage,115,"triage signal present; high-priority/blocker/regression label present; component/oncall labels: module: ci, oncall: distributed",review early; this can change release or regression risk,2026-07-06T19:18:35Z
issue,170003,batch_isend_irecv with nccl causes illegal memory access depending on P2P ordering,https://github.com/pytorch/pytorch/issues/170003,needs triage,115,"triage signal present; high-priority/blocker/regression label present; component/oncall labels: module: crash, module: nccl, oncall: distributed, oncall: distributed infra",review early; this can change release or regression risk,2026-06-15T04:21:25Z
issue,174288,[distributed] Batched isend/irecv with NCCL backend hangs on high load,https://github.com/pytorch/pytorch/issues/174288,needs triage,115,"triage signal present; high-priority/blocker/regression label present; component/oncall labels: module: vllm, oncall: distributed, oncall: distributed infra",review early; this can change release or regression risk,2026-06-12T05:49:56Z
issue,162731,DTensor cached op propagation results can be mutated when propagating other ops,https://github.com/pytorch/pytorch/issues/162731,needs triage,115,"triage signal present; high-priority/blocker/regression label present; component/oncall labels: module: dtensor, oncall: distributed, oncall: distributed parallelisms",review early; this can change release or regression risk,2026-06-11T17:25:13Z
issue,153517,[CI][CUDA][Distributed] test_non_blocking_with_eager_init timeout,https://github.com/pytorch/pytorch/issues/153517,needs triage,115,"PyTorch issue lacks triaged label; high-priority/blocker/regression label present; component/oncall labels: module: ci, module: nccl, module: regression, oncall: distributed",review early; this can change release or regression risk,2026-06-10T17:14:11Z
issue,167693,DISABLED test_side_stream_backward_overlap_cuda (__main__.TestAutogradStreamSynchronizationCUDA),https://github.com/pytorch/pytorch/issues/167693,needs triage,115,"triage signal present; high-priority/blocker/regression label present; component/oncall labels: module: autograd, module: flaky-tests, module: rocm",review early; this can change release or regression risk,2026-06-10T08:01:37Z
issue,116423,PyTorch Distributed Elastic Launch Segmentation Fault with Python 3.12,https://github.com/pytorch/pytorch/issues/116423,needs triage,115,"PyTorch issue lacks triaged label; high-priority/blocker/regression label present; component/oncall labels: module: crash, module: elastic, oncall: distributed, oncall: distributed infra",review early; this can change release or regression risk,2026-06-07T16:39:19Z
issue,106164,distributed.batch_isend_irecv() crash when send/recv refers to itself,https://github.com/pytorch/pytorch/issues/106164,needs triage,115,"PyTorch issue lacks triaged label; high-priority/blocker/regression label present; component/oncall labels: module: c10d, module: crash, oncall: distributed, oncall: distributed infra",review early; this can change release or regression risk,2026-06-07T16:36:35Z
issue,85088,reentrant torch.utils.checkpoint does not work with NamedTuple outputs,https://github.com/pytorch/pytorch/issues/85088,needs triage,115,"triage signal present; high-priority/blocker/regression label present; component/oncall labels: module: activation checkpointing, oncall: distributed, oncall: distributed parallelisms",review early; this can change release or regression risk,2026-06-07T16:35:52Z
issue,71303,[RFC] Cross-Process Performance Analysis: Straggler Detection,https://github.com/pytorch/pytorch/issues/71303,needs triage,115,"triage signal present; high-priority/blocker/regression label present; component/oncall labels: module: c10d, module: ddp, module: distributed_tool, oncall: distributed",review early; this can change release or regression risk,2026-06-06T16:30:33Z
issue,75725,Warning originating in C10 backend does not get translated to Python warning if run from subprocess,https://github.com/pytorch/pytorch/issues/75725,needs triage,115,"triage signal present; high-priority/blocker/regression label present; component/oncall labels: module: ddp, oncall: distributed, oncall: distributed parallelisms",review early; this can change release or regression risk,2026-06-06T16:28:36Z
issue,76449,Enhance _verify_param_shape_across_processes,https://github.com/pytorch/pytorch/issues/76449,needs triage,115,"triage signal present; high-priority/blocker/regression label present; component/oncall labels: module: ddp, module: fsdp, oncall: distributed, oncall: distributed parallelisms",review early; this can change release or regression risk,2026-06-06T16:28:02Z
issue,81684,Message exchange failure when perform alltoallv (cpus),https://github.com/pytorch/pytorch/issues/81684,needs triage,115,"triage signal present; high-priority/blocker/regression label present; component/oncall labels: module: c10d, oncall: distributed, oncall: distributed infra",review early; this can change release or regression risk,2026-06-05T17:11:38Z
issue,119845,Segmentation fault in dataloader worker sub-process,https://github.com/pytorch/pytorch/issues/119845,needs triage,115,"triage signal present; high-priority/blocker/regression label present; component/oncall labels: module: crash, module: dataloader, module: ddp, oncall: distributed",review early; this can change release or regression risk,2026-06-05T17:06:01Z
issue,183459,Regression in torch.distributed._functional_collectives.all_to_all_single in 2.9.1 -> 2.11,https://github.com/pytorch/pytorch/issues/183459,needs triage,115,"triage signal present; high-priority/blocker/regression label present; component/oncall labels: oncall: distributed, oncall: distributed infra",review early; this can change release or regression risk,2026-06-05T16:51:57Z
issue,180088,"[DTensor] RNG tracker does not advance state for CPU tensors on CUDA mesh, causing trunc_normal_ infinite loop",https://github.com/pytorch/pytorch/issues/180088,needs triage,115,"triage signal present; high-priority/blocker/regression label present; component/oncall labels: module: dtensor, oncall: distributed, oncall: distributed parallelisms",review early; this can change release or regression risk,2026-05-28T14:11:57Z
issue,173921,[Inductor] Significant numerical divergence (4.7% relative error) in Conv2d on CPU with torch.compile,https://github.com/pytorch/pytorch/issues/173921,needs triage,115,"PyTorch issue lacks triaged label; high-priority/blocker/regression label present; component/oncall labels: module: inductor, oncall: cpu inductor, oncall: pt2",review early; this can change release or regression risk,2026-05-21T18:17:56Z
issue,179502,[DTensor] sharded view incorrectly passes when redistribution is needed,https://github.com/pytorch/pytorch/issues/179502,needs triage,115,"triage signal present; high-priority/blocker/regression label present; component/oncall labels: module: dtensor, oncall: distributed",review early; this can change release or regression risk,2026-04-06T22:51:39Z
issue,145610,mmap fails on 64k page aarch64 systems for AOTI model loading,https://github.com/pytorch/pytorch/issues/145610,needs triage,115,"PyTorch issue lacks triaged label; high-priority/blocker/regression label present; component/oncall labels: module: aotinductor, module: arm, module: crash, oncall: export",review early; this can change release or regression risk,2026-02-17T08:26:10Z
issue,66504,BatchNorm runtimeError: one of the variables needed for gradient computation has been modified by an inplace operation,https://github.com/pytorch/pytorch/issues/66504,needs triage,115,"triage signal present; high-priority/blocker/regression label present; component/oncall labels: module: ddp, oncall: distributed",review early; this can change release or regression risk,2026-02-14T06:22:10Z
pr,188632,Fix Dynamo opaque object staticmethod guards,https://github.com/pytorch/pytorch/pull/188632,ready for maintainer decision,112,"needs-review label present; approved with no detected failing/pending checks; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",final maintainer merge/release decision,2026-07-09T01:33:58Z
pr,186429,Fix ONNX export of pad_sequence with symbolic split lengths,https://github.com/pytorch/pytorch/pull/186429,ready for maintainer decision,112,"needs-review label present; approved with no detected failing/pending checks; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/trunk",final maintainer merge/release decision,2026-07-08T18:59:33Z
pr,186461,Fix SymBool equality handling in symbolic shapes,https://github.com/pytorch/pytorch/pull/186461,ready for maintainer decision,112,needs-review label present; approved with no detected failing/pending checks; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,final maintainer merge/release decision,2026-07-08T18:17:51Z
pr,184716,Fix conv_transpose2d meta output padding validation,https://github.com/pytorch/pytorch/pull/184716,ready for maintainer decision,112,needs-review label present; approved with no detected failing/pending checks; PR closes or references open issue numbers,final maintainer merge/release decision,2026-07-08T07:16:54Z
pr,185691,Avoid UB in float to signed integer casts,https://github.com/pytorch/pytorch/pull/185691,ready for maintainer decision,112,needs-review label present; approved with no detected failing/pending checks; PR closes or references open issue numbers; CI lane labels: ciflow/trunk,final maintainer merge/release decision,2026-07-07T10:51:55Z
pr,184939,Support SymInt steps for linspace/logspace export,https://github.com/pytorch/pytorch/pull/184939,ready for maintainer decision,112,needs-review label present; approved with no detected failing/pending checks; PR closes or references open issue numbers; CI lane labels: ciflow/trunk,final maintainer merge/release decision,2026-06-18T10:35:58Z
pr,189307,[CI/CD] Copy newer CUPTI headers into manywheel binary-build images,https://github.com/pytorch/pytorch/pull/189307,PR blocked,108,failed checks or blocked/unstable merge state; approved with no detected failing/pending checks; CI lane labels: ciflow/docker,final maintainer merge/release decision,2026-07-09T01:53:06Z
pr,186927,[MPS] Gemv kernels,https://github.com/pytorch/pytorch/pull/186927,PR blocked,108,failed checks or blocked/unstable merge state; approved with no detected failing/pending checks; CI lane labels: ciflow/mps,final maintainer merge/release decision,2026-07-09T00:28:47Z
pr,189312,[CUDA][cuBLAS] Change cuBLAS default workspace size for SM 11.0 to 32 MiB,https://github.com/pytorch/pytorch/pull/189312,PR blocked,108,failed checks or blocked/unstable merge state; approved with no detected failing/pending checks,final maintainer merge/release decision,2026-07-08T23:49:19Z
pr,188301,[inductor] Fix inductor dropping ordering dep between effectful ops with different kernel types,https://github.com/pytorch/pytorch/pull/188301,PR blocked,108,"failed checks or blocked/unstable merge state; approved with no detected failing/pending checks; CI lane labels: ciflow/inductor, ciflow/torchtitan",final maintainer merge/release decision,2026-07-08T18:49:35Z
pr,184648,[Test] Add has_sufficient_memory() hook to DeviceTypeTestBase,https://github.com/pytorch/pytorch/pull/184648,PR blocked,108,failed checks or blocked/unstable merge state; approved with no detected failing/pending checks,final maintainer merge/release decision,2026-07-07T17:58:35Z
pr,188237,Optimize CSR SpMM CPU grain size and inner accumulation,https://github.com/pytorch/pytorch/pull/188237,PR blocked,108,failed checks or blocked/unstable merge state; approved with no detected failing/pending checks,final maintainer merge/release decision,2026-07-07T14:34:05Z
pr,187501,[Stable C Shim] Use error message retrieval shim if available at runtime,https://github.com/pytorch/pytorch/pull/187501,PR blocked,108,failed checks or blocked/unstable merge state; approved with no detected failing/pending checks,final maintainer merge/release decision,2026-06-25T16:14:35Z
pr,185882,[PrivateUse1] Return `False` instead of `None` when PU1 backend is not available,https://github.com/pytorch/pytorch/pull/185882,PR blocked,108,failed checks or blocked/unstable merge state; approved with no detected failing/pending checks,final maintainer merge/release decision,2026-06-23T15:23:08Z
issue,153410,[Feature request] `torch.export` .save/.load could support `safetensors` and/or `weights_only=True`,https://github.com/pytorch/pytorch/issues/153410,needs triage,107,"PyTorch issue lacks triaged label; open PR references this issue; label says issue is suitable for a scoped external contribution; component/oncall labels: oncall: export, oncall: pt2",review linked PR state before touching issue,2026-01-18T09:26:05Z
issue,188892,nvidia-cudnn-cu13==9.20.0.48 bundled by torch is incomplete → CUDNN_STATUS_SUBLIBRARY_VERSION_MISMATCH,https://github.com/pytorch/pytorch/issues/188892,has linked PR,106,"open PR references this issue; high-priority/blocker/regression label present; needs reproduction label present; component/oncall labels: module: binaries, module: convolution, module: crash, module: cudnn",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-07-07T09:16:01Z
issue,188970,[MPS] SIGSEGV in MPSStream::copy with triton.cudagraphs (reduce-overhead/max-autotune) on deep fp16 graphs,https://github.com/pytorch/pytorch/issues/188970,has linked PR,106,"open PR references this issue; high-priority/blocker/regression label present; needs reproduction label present; component/oncall labels: module: crash, module: inductor, module: macos, module: mps",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-07-07T05:43:13Z
pr,135631,[scan] Autograd,https://github.com/pytorch/pytorch/pull/135631,PR blocked,100,failed checks or blocked/unstable merge state; approved with no detected failing/pending checks; PR has not updated for at least 45 days; CI lane labels: ciflow/trunk,final maintainer merge/release decision,2025-02-03T00:37:41Z
pr,188115,[Inductor][TEST] Align `test_main_loop_scaling` with H100 support surface,https://github.com/pytorch/pytorch/pull/188115,PR blocked,98,"failed checks or blocked/unstable merge state; high-priority/blocker/regression label present; CI lane labels: ciflow/b200, ciflow/h100, ciflow/inductor, ciflow/rocm-mi300","identify whether block is CI, merge conflict, or review gate",2026-07-03T02:51:54Z
issue,149447,DTensor slicing on sharded dimension leads to replication,https://github.com/pytorch/pytorch/issues/149447,needs triage,97,"triage signal present; high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: dtensor, oncall: distributed",review early; this can change release or regression risk,2025-12-20T20:52:34Z
issue,164929,get_optimizer_state_dict modifies optimizer state,https://github.com/pytorch/pytorch/issues/164929,needs triage,97,"triage signal present; high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: oncall: distributed, oncall: distributed checkpointing",review early; this can change release or regression risk,2025-12-10T02:42:02Z
issue,156563,Segmentation fault (core dumped) in `torch.profiler.profile`,https://github.com/pytorch/pytorch/issues/156563,needs triage,97,"PyTorch issue lacks triaged label; high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: error checking, oncall: profiler",review early; this can change release or regression risk,2025-11-19T18:16:25Z
issue,122019,`F.max_pool3d`: Segfault due to incorrect `dilation` check in kernel,https://github.com/pytorch/pytorch/issues/122019,needs triage,97,"triage signal present; high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, oncall: quantization",review early; this can change release or regression risk,2025-09-26T23:19:29Z
issue,159238,Performance regression: torch.jit.trace() significantly slower on RTX 5090 than RTX 4060 (cu128 nightly),https://github.com/pytorch/pytorch/issues/159238,needs triage,97,"PyTorch issue lacks triaged label; high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: regression, oncall: jit",review early; this can change release or regression risk,2025-08-09T16:07:58Z
issue,121219,Torch profiler corrupted names with Python 3.11,https://github.com/pytorch/pytorch/issues/121219,needs triage,97,triage signal present; high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: oncall: profiler,review early; this can change release or regression risk,2025-08-07T04:30:15Z
issue,91165,[FSDP] FSDP with CPU offload consumes `1.65X` more GPU memory when training models with most of the params frozen,https://github.com/pytorch/pytorch/issues/91165,needs triage,97,"triage signal present; high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: fsdp, oncall: distributed",review early; this can change release or regression risk,2025-08-04T00:11:21Z
issue,47260,DDP doesn't work with retain_graph = True,https://github.com/pytorch/pytorch/issues/47260,needs triage,97,triage signal present; high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,review early; this can change release or regression risk,2025-06-09T11:34:58Z
issue,100253,profiler.export_stacks doesn't return stack trace unless experimental_config is provided,https://github.com/pytorch/pytorch/issues/100253,needs triage,97,"PyTorch issue lacks triaged label; high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: regression, oncall: profiler",review early; this can change release or regression risk,2025-06-08T23:12:36Z
issue,155147,torch.profiler raises Aborted (core dumped) failurer related with GIL (gilstate_tss_set),https://github.com/pytorch/pytorch/issues/155147,needs triage,97,"PyTorch issue lacks triaged label; high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: wsl, oncall: profiler",review early; this can change release or regression risk,2025-06-06T07:45:23Z
issue,155146,Segmentation fault when using torch.profiler,https://github.com/pytorch/pytorch/issues/155146,needs triage,97,"PyTorch issue lacks triaged label; high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: wsl, oncall: profiler",review early; this can change release or regression risk,2025-06-05T22:41:29Z
issue,153941,TorchScript Fails Type Inference When Using .item()-Derived Scalar in Tensor Construction,https://github.com/pytorch/pytorch/issues/153941,needs triage,97,"PyTorch issue lacks triaged label; high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: regression, oncall: jit",review early; this can change release or regression risk,2025-05-20T14:07:31Z
issue,153260,Export shouldn't run TS under the hood.,https://github.com/pytorch/pytorch/issues/153260,needs triage,97,"PyTorch issue lacks triaged label; high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, oncall: export, oncall: jit, oncall: pt2",review early; this can change release or regression risk,2025-05-13T17:23:02Z
issue,153145,Segmentation faults with CPU FlexAttention,https://github.com/pytorch/pytorch/issues/153145,needs triage,97,"PyTorch issue lacks triaged label; high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: crash, module: flex attention, module: higher order operators",review early; this can change release or regression risk,2025-05-09T06:36:24Z
issue,152108,[inductor][cpu] AMP static shape default wrapper AOTInductor performance regression in 2025_04_20 nightly release,https://github.com/pytorch/pytorch/issues/152108,needs triage,97,"PyTorch issue lacks triaged label; high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: regression, oncall: cpu inductor, oncall: pt2",review early; this can change release or regression risk,2025-04-24T23:33:51Z
issue,145088,[torchbench] torch._dynamo.exc.Unsupported: Graph break due to unsupported builtin None.morphologyEx,https://github.com/pytorch/pytorch/issues/145088,needs triage,97,"PyTorch issue lacks triaged label; high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: export, oncall: pt2",review early; this can change release or regression risk,2025-03-05T18:26:38Z
issue,146788,Segmentation Fault in `torch.choose_qparams_optimized` with Invalid Parameters,https://github.com/pytorch/pytorch/issues/146788,needs triage,97,"PyTorch issue lacks triaged label; high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, oncall: quantization",review early; this can change release or regression risk,2025-02-09T20:17:13Z
issue,124950,[distributed] First NCCL barrier does not respect timeout,https://github.com/pytorch/pytorch/issues/124950,needs triage,97,triage signal present; high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,review early; this can change release or regression risk,2025-01-30T00:43:40Z
issue,71855,Regression in multi-node training speed with Transformers + PyTorch,https://github.com/pytorch/pytorch/issues/71855,needs triage,97,"triage signal present; high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: ddp, module: regression, oncall: distributed",review early; this can change release or regression risk,2024-11-27T19:48:40Z
issue,136712,Segmentation fault (core dumped) in `torch.ao.nn.quantized.dynamic.LSTMCell/GRUCell`,https://github.com/pytorch/pytorch/issues/136712,needs triage,97,"PyTorch issue lacks triaged label; high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, oncall: quantization",review early; this can change release or regression risk,2024-09-27T01:12:24Z
issue,136728,Aborted (core dumped) in `torch.package.package_exporter.PackageExporter`/`torch.package.PackageExporter`,https://github.com/pytorch/pytorch/issues/136728,needs triage,97,"PyTorch issue lacks triaged label; high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, oncall: package/deploy",review early; this can change release or regression risk,2024-09-26T03:56:33Z
issue,135835,A segmentation fault will be raised when using the `torch._C._jit_to_backend`,https://github.com/pytorch/pytorch/issues/135835,needs triage,97,"PyTorch issue lacks triaged label; high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, oncall: jit",review early; this can change release or regression risk,2024-09-12T21:00:13Z
issue,81085,RuntimeError: required keyword attribute 'value' is undefined,https://github.com/pytorch/pytorch/issues/81085,needs triage,97,triage signal present; high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,review early; this can change release or regression risk,2024-09-09T23:14:38Z
issue,128693,Segmentation fault (core dumped) in `torch.fused_moving_avg_obs_fake_quant`,https://github.com/pytorch/pytorch/issues/128693,needs triage,97,"PyTorch issue lacks triaged label; high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: edge cases, oncall: quantization",review early; this can change release or regression risk,2024-07-22T17:28:53Z
issue,83606,"test_profiler_experimental_tree_cuda_detailed is too unstable, and as its CUDA only difficult to regen",https://github.com/pytorch/pytorch/issues/83606,needs triage,97,"triage signal present; high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: tests, oncall: profiler",review early; this can change release or regression risk,2024-07-16T13:34:20Z
issue,103947,"[torch.fx] Deserialization Error - TypeError: ones() received an invalid combination of arguments - got (tuple, device=Attribute)",https://github.com/pytorch/pytorch/issues/103947,needs triage,97,"triage signal present; high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: fx, oncall: fx",review early; this can change release or regression risk,2024-02-05T12:28:08Z
issue,68407,[Runtime Error in ddp_pipeline.py] RuntimeError: unsupported operation: some elements of the input tensor and the written-to tensor refer to a single memory location. Please clone() the tensor before performing the operation.,https://github.com/pytorch/pytorch/issues/68407,needs triage,97,"triage signal present; high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: ddp, module: docs, oncall: distributed",review early; this can change release or regression risk,2024-01-08T22:50:47Z
issue,116351,`torch.quantize_per_channel`: FPE,https://github.com/pytorch/pytorch/issues/116351,needs triage,97,"PyTorch issue lacks triaged label; high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, oncall: quantization",review early; this can change release or regression risk,2024-01-05T21:55:02Z
issue,34329,[jit] Hook support tracking,https://github.com/pytorch/pytorch/issues/34329,needs triage,97,triage signal present; high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,review early; this can change release or regression risk,2024-01-05T18:39:51Z
issue,113572,Deepcopying an exported model changes numerics for MobileNetV2,https://github.com/pytorch/pytorch/issues/113572,needs triage,97,triage signal present; high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: oncall: export,review early; this can change release or regression risk,2024-01-03T22:07:07Z
issue,115818,Fx graph recompilation along with partitioner produces different node order even though graph operations are same,https://github.com/pytorch/pytorch/issues/115818,needs triage,97,triage signal present; high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: oncall: fx,review early; this can change release or regression risk,2024-01-03T07:58:33Z
issue,115640,"Quantisation results in a ""Illegal instruction (core dumped)""",https://github.com/pytorch/pytorch/issues/115640,needs triage,97,"triage signal present; high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, oncall: quantization",review early; this can change release or regression risk,2023-12-21T15:18:12Z
issue,73602,Report DDP efficiency metric to users and guide them for setting up DDP correctly,https://github.com/pytorch/pytorch/issues/73602,needs triage,97,triage signal present; high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,review early; this can change release or regression risk,2023-11-07T16:38:55Z
issue,83289,Improvements to ProcessGroupGloo monitored_barrier,https://github.com/pytorch/pytorch/issues/83289,needs triage,97,"triage signal present; high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: c10d, oncall: distributed",review early; this can change release or regression risk,2023-10-19T20:27:58Z
issue,80151,activation checkpointing with non_reentrant implementation memory leaks,https://github.com/pytorch/pytorch/issues/80151,needs triage,97,triage signal present; high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,review early; this can change release or regression risk,2023-05-23T04:52:08Z
issue,64345,Profiler UTF-8 decode issue,https://github.com/pytorch/pytorch/issues/64345,needs triage,97,triage signal present; high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: oncall: profiler,review early; this can change release or regression risk,2023-02-03T14:07:37Z
issue,86594,JIT model with mean will crash when computing the gradients on cuda,https://github.com/pytorch/pytorch/issues/86594,needs triage,97,"triage signal present; high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, oncall: jit",review early; this can change release or regression risk,2022-10-10T18:33:46Z
issue,86595,JIT model with `relu+div+sgn` will crash when computing the gradient,https://github.com/pytorch/pytorch/issues/86595,needs triage,97,"PyTorch issue lacks triaged label; high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: functorch, oncall: jit",review early; this can change release or regression risk,2022-10-10T15:28:27Z
issue,86532,Conv2d will crash by using `jit.trace`,https://github.com/pytorch/pytorch/issues/86532,needs triage,97,"PyTorch issue lacks triaged label; high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, oncall: jit",review early; this can change release or regression risk,2022-10-10T15:25:01Z
issue,67646,test_jit_fuser_legacy and test_jit_fuser_te are failing with SIGIOT,https://github.com/pytorch/pytorch/issues/67646,needs triage,97,"triage signal present; high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: tests, oncall: jit",review early; this can change release or regression risk,2022-10-08T01:26:04Z
issue,65357,Improve DDP tutorial,https://github.com/pytorch/pytorch/issues/65357,needs triage,97,"triage signal present; high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: ddp, module: docs, oncall: distributed",review early; this can change release or regression risk,2022-07-27T19:47:25Z
issue,81115,torch.package can not be used to serialize `resnet18` from TorchVision-0.12,https://github.com/pytorch/pytorch/issues/81115,needs triage,97,"PyTorch issue lacks triaged label; high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: regression, module: vision, oncall: package/deploy",review early; this can change release or regression risk,2022-07-11T17:20:29Z
issue,80349,Distributed Store `get` doesn't work well with `add`,https://github.com/pytorch/pytorch/issues/80349,needs triage,97,"triage signal present; high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: c10d, module: docs, oncall: distributed",review early; this can change release or regression risk,2022-06-29T14:26:43Z
issue,74115,DistributedDataParallel high peak memory usage with find_unused_parameters=True,https://github.com/pytorch/pytorch/issues/74115,needs triage,97,"triage signal present; high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: ddp, oncall: distributed",review early; this can change release or regression risk,2022-06-08T23:47:29Z
issue,72859,Profiler crashes with ProfilerActivity.CUDA on AWS p4d.24xlarge with A100-SXM4-40GB,https://github.com/pytorch/pytorch/issues/72859,needs triage,97,"triage signal present; high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda, oncall: profiler",review early; this can change release or regression risk,2022-04-13T21:42:32Z
issue,69179,DISABLED test_builtin_remote_message_dropped_timeout_to_self (__main__.FaultyFaultyAgentRpcTest),https://github.com/pytorch/pytorch/issues/69179,needs triage,97,"triage signal present; high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: flaky-tests, module: rpc, oncall: distributed",review early; this can change release or regression risk,2022-03-25T22:16:25Z
issue,73885,operation not supported crash when initializing RPC tensorpipe,https://github.com/pytorch/pytorch/issues/73885,needs triage,97,"triage signal present; high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, oncall: distributed",review early; this can change release or regression risk,2022-03-17T13:18:23Z
issue,58111,Static graph training fails if forward is called multiple times before backward,https://github.com/pytorch/pytorch/issues/58111,needs triage,97,"triage signal present; high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: ddp, oncall: distributed",review early; this can change release or regression risk,2022-03-04T04:50:33Z
issue,67389,Missing doc for torch.distributed functions,https://github.com/pytorch/pytorch/issues/67389,needs triage,97,"triage signal present; high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: docs, oncall: distributed",review early; this can change release or regression risk,2022-03-04T04:36:47Z
issue,73352,Segfault on unloading a model,https://github.com/pytorch/pytorch/issues/73352,needs triage,97,"PyTorch issue lacks triaged label; high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: crash, oncall: jit",review early; this can change release or regression risk,2022-02-25T04:38:55Z
issue,63917,Add tests for DDP broadcast_buffers=True,https://github.com/pytorch/pytorch/issues/63917,needs triage,97,"triage signal present; high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, module: ddp, oncall: distributed",review early; this can change release or regression risk,2021-12-21T19:49:21Z
issue,67541,TestBenchNetwork::test_forward[resnet50_jit-profiling-te] is failing intermittently,https://github.com/pytorch/pytorch/issues/67541,needs triage,97,triage signal present; high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,review early; this can change release or regression risk,2021-10-29T15:56:37Z
issue,47218,[JIT] Alias Annotations for Native Schema List Ops Is Subtly Wrong,https://github.com/pytorch/pytorch/issues/47218,needs triage,97,PyTorch issue lacks triaged label; high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,review early; this can change release or regression risk,2021-10-22T20:44:46Z
issue,66894,test_nccl_barrier_timeout_new_group_non_member fails intermittently,https://github.com/pytorch/pytorch/issues/66894,needs triage,97,"triage signal present; high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: tests, oncall: distributed",review early; this can change release or regression risk,2021-10-19T20:21:54Z
issue,54524,Store created by dist.new_group doesn't appear to respect timeout,https://github.com/pytorch/pytorch/issues/54524,needs triage,97,triage signal present; high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,review early; this can change release or regression risk,2021-10-14T00:42:36Z
issue,39351,Clean up RPC request callback implementation,https://github.com/pytorch/pytorch/issues/39351,needs triage,97,"triage signal present; high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: rpc, oncall: distributed",review early; this can change release or regression risk,2021-10-13T21:33:23Z
issue,64518,TensorPipeDistAutogradTest is frequently failing,https://github.com/pytorch/pytorch/issues/64518,needs triage,97,"triage signal present; high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: tests, oncall: distributed",review early; this can change release or regression risk,2021-10-12T20:43:42Z
issue,62780,Seeing unit test failures for ProcessGroupShareTensorTest,https://github.com/pytorch/pytorch/issues/62780,needs triage,97,"triage signal present; high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: c10d, module: nccl, oncall: distributed",review early; this can change release or regression risk,2021-08-17T14:25:11Z
issue,58856,NCCL multi-gpu test intermittently failing after NCCL version upgrade,https://github.com/pytorch/pytorch/issues/58856,needs triage,97,"triage signal present; high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: flaky-tests, module: nccl, oncall: distributed",review early; this can change release or regression risk,2021-07-21T19:25:48Z
issue,55905,Attempt to use jited `torch.isnan` hit internal assert,https://github.com/pytorch/pytorch/issues/55905,needs triage,97,"PyTorch issue lacks triaged label; high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: ci, module: crash, oncall: jit",review early; this can change release or regression risk,2021-04-15T05:29:50Z
issue,40867,[JIT] Infer type of argument foo = None to be Optional[Tensor],https://github.com/pytorch/pytorch/issues/40867,needs triage,97,triage signal present; high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,review early; this can change release or regression risk,2021-03-26T07:31:24Z
issue,47999,Torch DDP seems not prefer IPv6 over IPv4?,https://github.com/pytorch/pytorch/issues/47999,needs triage,97,triage signal present; high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,review early; this can change release or regression risk,2021-03-15T17:45:33Z
issue,45549,Can't initialize NCCL/GLOO process group if default process group is MPI,https://github.com/pytorch/pytorch/issues/45549,needs triage,97,"triage signal present; high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, module: nccl, oncall: distributed",review early; this can change release or regression risk,2021-02-11T02:14:42Z
issue,28786,RPC couldn't match torch.ones with requires_grad=True,https://github.com/pytorch/pytorch/issues/28786,needs triage,97,"triage signal present; high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: internals, module: rpc, oncall: distributed",review early; this can change release or regression risk,2020-12-07T18:10:50Z
issue,46166,Error with DistributedDataParallel with specific model,https://github.com/pytorch/pytorch/issues/46166,needs triage,97,triage signal present; high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,review early; this can change release or regression risk,2020-12-07T02:00:04Z
issue,34116,autograd with TorchScript does not match finite differences,https://github.com/pytorch/pytorch/issues/34116,needs triage,97,triage signal present; high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,review early; this can change release or regression risk,2020-10-27T17:14:58Z
issue,41289,List[Any] support in TorchScript,https://github.com/pytorch/pytorch/issues/41289,needs triage,97,triage signal present; high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,review early; this can change release or regression risk,2020-10-11T18:45:01Z
issue,43548,JIT tracing incorrectly records some slice bounds,https://github.com/pytorch/pytorch/issues/43548,needs triage,97,triage signal present; high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,review early; this can change release or regression risk,2020-09-17T17:17:29Z
issue,42258,Cannot load certain function from dumped Torchscript file,https://github.com/pytorch/pytorch/issues/42258,needs triage,97,triage signal present; high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,review early; this can change release or regression risk,2020-08-17T21:51:35Z
issue,42317,Instance norm annotation is incorrect,https://github.com/pytorch/pytorch/issues/42317,needs triage,97,triage signal present; high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,review early; this can change release or regression risk,2020-07-30T16:34:37Z
issue,41695,Improve overload resolution order rules for `invokeOperatorFromPython`,https://github.com/pytorch/pytorch/issues/41695,needs triage,97,PyTorch issue lacks triaged label; high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,review early; this can change release or regression risk,2020-07-20T21:58:55Z
issue,35998,JITed GRU too slow,https://github.com/pytorch/pytorch/issues/35998,needs triage,97,triage signal present; high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,review early; this can change release or regression risk,2020-06-10T22:11:44Z
issue,37015,Converting to Torch Script: cpp_module does not match nn_module,https://github.com/pytorch/pytorch/issues/37015,needs triage,97,triage signal present; high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,review early; this can change release or regression risk,2020-04-29T17:35:29Z
issue,7036,[Caffe2] compilation linker error `libtbb.so.2: undefined reference to std::__exception_ptr`,https://github.com/pytorch/pytorch/issues/7036,needs triage,97,"PyTorch issue lacks triaged label; high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: build, module: crash",review early; this can change release or regression risk,2018-04-27T17:12:50Z
issue,189267,nn.functional.threshold OpInfo reference raises OverflowError on unsigned dtypes with NumPy >= 2.5,https://github.com/pytorch/pytorch/issues/189267,has linked PR,94,"open PR references this issue; high-priority/blocker/regression label present; component/oncall labels: module: nn, module: numpy, module: regression, module: tests",review early; this can change release or regression risk,2026-07-08T16:58:04Z
issue,189034,Limited CI on H100 (smoke) red since 2026-07-03: test_foreach import crash from NumPy 2.x / SciPy ABI mismatch,https://github.com/pytorch/pytorch/issues/189034,has linked PR,94,"open PR references this issue; high-priority/blocker/regression label present; component/oncall labels: module: ci, module: regression",review early; this can change release or regression risk,2026-07-06T17:33:27Z
issue,188792,CPU float unary ops can silently return low-precision results: MKL VML first-call kernel dispatch race,https://github.com/pytorch/pytorch/issues/188792,has linked PR,94,"open PR references this issue; high-priority/blocker/regression label present; component/oncall labels: module: correctness (silent), module: cpu, module: intel, module: mkl",review early; this can change release or regression risk,2026-07-06T17:25:49Z
issue,188492,"swin_base_patch4_window7_224 fails inductor accuracy (bf16 inference, H100) after Triton 3.8 pin bump",https://github.com/pytorch/pytorch/issues/188492,has linked PR,94,"open PR references this issue; high-priority/blocker/regression label present; component/oncall labels: module: bfloat16, module: correctness (silent), module: cuda, module: dependency bug",review early; this can change release or regression risk,2026-07-06T17:23:27Z
issue,188358,[Dynamo] Eager and torch.compile behavior mismatch in F.batch_norm with multidimensional bias,https://github.com/pytorch/pytorch/issues/188358,has linked PR,94,"open PR references this issue; high-priority/blocker/regression label present; component/oncall labels: module: pt2-dispatcher, oncall: pt2",review early; this can change release or regression risk,2026-07-06T17:20:16Z
issue,188545,[PT2] Output mismatch between eager and torch.compile for torch.special.bessel_y0 with +inf input on CUDA,https://github.com/pytorch/pytorch/issues/188545,has linked PR,94,"open PR references this issue; high-priority/blocker/regression label present; component/oncall labels: module: correctness (silent), module: cuda, module: edge cases, module: nans and infs",review early; this can change release or regression risk,2026-07-06T17:12:08Z
issue,188541,"[Inductor][CUDA] F.logsigmoid flushes negative float32 subnormal outputs to +0.0, changing sign-sensitive behavior",https://github.com/pytorch/pytorch/issues/188541,has linked PR,94,"open PR references this issue; high-priority/blocker/regression label present; component/oncall labels: module: correctness (silent), module: cuda, module: edge cases, module: inductor",review early; this can change release or regression risk,2026-07-06T17:09:07Z
issue,188033,[BUG] THPStream_richcompare casts arbitrary objects to THPStream* without type check,https://github.com/pytorch/pytorch/issues/188033,has linked PR,94,"open PR references this issue; high-priority/blocker/regression label present; component/oncall labels: module: crash, module: cuda, module: error checking, module: pybind",review early; this can change release or regression risk,2026-07-06T17:08:05Z
issue,188008,expandable_segments: partial mapAndSetAccess failure → bulk cuMemUnmap of unmapped range → terminate in destructor,https://github.com/pytorch/pytorch/issues/188008,has linked PR,94,"open PR references this issue; high-priority/blocker/regression label present; component/oncall labels: module: crash, module: cuda, module: cudacachingallocator",review early; this can change release or regression risk,2026-07-06T17:07:36Z
issue,188547,[dynamo] Infinite recursion when returning a Pybind11 enum from a compiled function,https://github.com/pytorch/pytorch/issues/188547,has linked PR,94,"open PR references this issue; high-priority/blocker/regression label present; component/oncall labels: module: crash, module: dynamo, module: pybind, oncall: pt2",review early; this can change release or regression risk,2026-07-06T17:02:39Z
issue,188890,Non-deterministic gradients when using torch.compile and PyTorch >= 2.12,https://github.com/pytorch/pytorch/issues/188890,has linked PR,94,"open PR references this issue; high-priority/blocker/regression label present; component/oncall labels: module: correctness (silent), module: determinism, module: regression, oncall: pt2",review early; this can change release or regression risk,2026-07-06T15:19:14Z
issue,148317,compile SageAttention faing error C2872: “std” for latest torch nightly,https://github.com/pytorch/pytorch/issues/148317,has linked PR,94,"open PR references this issue; high-priority/blocker/regression label present; component/oncall labels: module: regression, module: windows, oncall: pt2",review early; this can change release or regression risk,2026-07-05T04:05:19Z
issue,177116,MPS: catastrophically wrong gradients in backward pass (>32K elements),https://github.com/pytorch/pytorch/issues/177116,has linked PR,94,"open PR references this issue; high-priority/blocker/regression label present; component/oncall labels: module: autograd, module: mps",review early; this can change release or regression risk,2026-07-04T08:42:57Z
issue,188578,[Inductor] Optimize split reduction heuristics on GB200,https://github.com/pytorch/pytorch/issues/188578,has linked PR,94,"open PR references this issue; high-priority/blocker/regression label present; component/oncall labels: module: inductor, module: performance, module: regression, oncall: pt2",review early; this can change release or regression risk,2026-06-30T18:40:02Z
issue,188231,[inductor] `TypeError: A Boolean argument can only be used in Eq and Ne` during CPP codegen,https://github.com/pytorch/pytorch/issues/188231,has linked PR,94,"open PR references this issue; high-priority/blocker/regression label present; component/oncall labels: module: crash, module: inductor, oncall: pt2",review early; this can change release or regression risk,2026-06-29T19:32:45Z
issue,188198,[MPS] cat op dropped performance in torch==2.10 release,https://github.com/pytorch/pytorch/issues/188198,has linked PR,94,"open PR references this issue; high-priority/blocker/regression label present; component/oncall labels: module: mps, module: performance, module: regression",review early; this can change release or regression risk,2026-06-29T18:21:12Z
issue,187988,[MPS] linear backward crashes (SIGABRT) with a 1-D weight: 'mps.matmul' op contracting dimensions differ,https://github.com/pytorch/pytorch/issues/187988,has linked PR,94,"open PR references this issue; high-priority/blocker/regression label present; component/oncall labels: module: arm, module: autograd, module: crash, module: macos",review early; this can change release or regression risk,2026-06-29T18:01:10Z
issue,188034,[BUG] Reference leak in autograd Function.apply — setup_context return value never released,https://github.com/pytorch/pytorch/issues/188034,has linked PR,94,"open PR references this issue; high-priority/blocker/regression label present; component/oncall labels: module: autograd, module: memory usage, module: pybind",review early; this can change release or regression risk,2026-06-29T17:41:30Z
issue,188023,torch.from_dlpack aborts the Python process for NumPy arrays with negative strides,https://github.com/pytorch/pytorch/issues/188023,has linked PR,94,"open PR references this issue; high-priority/blocker/regression label present; component/oncall labels: module: crash, module: dlpack, module: error checking, module: numpy",review early; this can change release or regression risk,2026-06-29T17:40:54Z
issue,188261,[inductor][XPU] PR #184366 causes AMP fp16 training accuracy regression on XPU,https://github.com/pytorch/pytorch/issues/188261,has linked PR,94,"open PR references this issue; high-priority/blocker/regression label present; component/oncall labels: module: amp (automated mixed precision), module: correctness (silent), module: half, module: inductor",review early; this can change release or regression risk,2026-06-29T17:29:49Z
issue,188113,CK SDPA flash attention kernel asserts on gfx1200/gfx1201 (RDNA4),https://github.com/pytorch/pytorch/issues/188113,has linked PR,94,"open PR references this issue; high-priority/blocker/regression label present; component/oncall labels: module: crash, module: rocm, module: sdpa",review early; this can change release or regression risk,2026-06-26T23:57:15Z
issue,188069,"torch.compile(mode=""max-autotune"") crashes with too many positional arguments during bmm autotuning when template inputs are deduplicated",https://github.com/pytorch/pytorch/issues/188069,has linked PR,94,"open PR references this issue; high-priority/blocker/regression label present; component/oncall labels: module: crash, module: inductor, oncall: pt2",review early; this can change release or regression risk,2026-06-26T12:55:56Z
issue,185886,[torch.compile] Clarification/regression: Dynamo cache lookup guard eval is much slower in 2.12 vs 2.7 after `inline_inbuilt_nn_modules` became always True,https://github.com/pytorch/pytorch/issues/185886,has linked PR,94,"open PR references this issue; high-priority/blocker/regression label present; component/oncall labels: module: dynamo, module: guards, module: performance, module: regression",review early; this can change release or regression risk,2026-06-24T08:24:58Z
issue,185484,[Inductor] Forward output mismatch between Eager and Inductor for F.softshrink with bfloat16 and float scalar,https://github.com/pytorch/pytorch/issues/185484,has linked PR,94,"open PR references this issue; high-priority/blocker/regression label present; component/oncall labels: module: bfloat16, module: correctness (silent), module: inductor, module: type promotion",review early; this can change release or regression risk,2026-06-23T17:29:18Z
issue,186084,"[inductor] insert_overlap_deps wraps non-overlap-critical compute with control_deps, blocking fusion and causing silent bf16 accuracy regression",https://github.com/pytorch/pytorch/issues/186084,has linked PR,94,"open PR references this issue; high-priority/blocker/regression label present; component/oncall labels: module: inductor, oncall: pt2",review early; this can change release or regression risk,2026-06-23T17:23:02Z
issue,187415,[Inductor] torch.compile fails on CUDA with bool mask indexing after torch.eye constructor movement: AssertionError in _rename_unbacked_to,https://github.com/pytorch/pytorch/issues/187415,has linked PR,94,"open PR references this issue; high-priority/blocker/regression label present; component/oncall labels: module: dynamic shapes, module: inductor, oncall: pt2",review early; this can change release or regression risk,2026-06-22T15:15:37Z
issue,187340,[torch.compile][Inductor][CUDA] torch.nextafter is incorrectly compiled as no-op for non-contiguous float16 tensors,https://github.com/pytorch/pytorch/issues/187340,has linked PR,94,"open PR references this issue; high-priority/blocker/regression label present; component/oncall labels: module: correctness (silent), module: inductor, oncall: pt2",review early; this can change release or regression risk,2026-06-22T15:15:28Z
issue,187567,[MPS] torch.linalg.qr is 25-250× slower than CPU on tall-skinny matrices,https://github.com/pytorch/pytorch/issues/187567,has linked PR,94,"open PR references this issue; high-priority/blocker/regression label present; component/oncall labels: module: linear algebra, module: macos, module: mps, module: performance",review early; this can change release or regression risk,2026-06-22T15:14:47Z
issue,187399,[MPS] torch.baddbmm — INTERNAL ASSERT for size-0 batch dimensions,https://github.com/pytorch/pytorch/issues/187399,has linked PR,94,"open PR references this issue; high-priority/blocker/regression label present; component/oncall labels: module: crash, module: edge cases, module: linear algebra, module: mps",review early; this can change release or regression risk,2026-06-22T15:14:30Z
issue,187429,"torch.where silently casts out-of-range Python float to +/-Inf in fp16 (inconsistent with full_like / full / fill_, which raise)",https://github.com/pytorch/pytorch/issues/187429,has linked PR,94,"open PR references this issue; high-priority/blocker/regression label present; component/oncall labels: module: correctness (silent), module: edge cases, module: error checking, module: half",review early; this can change release or regression risk,2026-06-22T15:13:08Z
issue,187027,flip after repeat_interleave generates negative (out-of-bounds) index under dynamic shapes,https://github.com/pytorch/pytorch/issues/187027,has linked PR,94,"open PR references this issue; high-priority/blocker/regression label present; component/oncall labels: module: correctness (silent), module: dynamic shapes, module: inductor, oncall: pt2",review early; this can change release or regression risk,2026-06-18T20:46:56Z
issue,187284,`torch.compile` with `inductor` backend silently drops forward-mode AD tangents (returns `None`),https://github.com/pytorch/pytorch/issues/187284,has linked PR,94,"open PR references this issue; high-priority/blocker/regression label present; component/oncall labels: module: aotdispatch, module: correctness (silent), module: dynamo, module: forward ad",review early; this can change release or regression risk,2026-06-15T22:11:36Z
issue,186537,torch 2.12 regression: torch.compile thread-local config - even backward thread has separate config,https://github.com/pytorch/pytorch/issues/186537,has linked PR,94,"open PR references this issue; high-priority/blocker/regression label present; component/oncall labels: module: bc-breaking, module: dynamo, module: multithreading, module: regression",review early; this can change release or regression risk,2026-06-14T18:03:12Z
issue,162358,heap-buffer-overflow in torch.linalg.eig,https://github.com/pytorch/pytorch/issues/162358,has linked PR,94,"open PR references this issue; high-priority/blocker/regression label present; component/oncall labels: module: crash, module: error checking, module: linear algebra, module: sanitizers",review early; this can change release or regression risk,2026-06-14T16:50:12Z
issue,160018,Triton kernel generated for Flex Attention throws InductorError: RuntimeError: No valid triton configs when seq length < 128,https://github.com/pytorch/pytorch/issues/160018,has linked PR,94,"open PR references this issue; high-priority/blocker/regression label present; component/oncall labels: module: crash, module: flex attention, module: higher order operators, module: pt2-dispatcher",review early; this can change release or regression risk,2026-06-13T23:19:37Z
issue,175370,`F.embedding_bag` segfaults with float64 weight and empty offsets,https://github.com/pytorch/pytorch/issues/175370,has linked PR,94,"open PR references this issue; high-priority/blocker/regression label present; component/oncall labels: module: cpu, module: crash, module: embedding, module: empty tensor",review early; this can change release or regression risk,2026-06-12T00:34:42Z
issue,183806,`torch.linalg.eig` and `torch.linalg.eigh` 30-130x slower on 2.11.0+cu130 vs 2.5.1+cu124,https://github.com/pytorch/pytorch/issues/183806,has linked PR,94,"open PR references this issue; high-priority/blocker/regression label present; component/oncall labels: module: cuda, module: linear algebra, module: magma, module: performance",review early; this can change release or regression risk,2026-06-11T10:06:06Z
issue,41508,nn.MultiheadAttention causes gradients to become NaN under some use cases,https://github.com/pytorch/pytorch/issues/41508,has linked PR,94,"open PR references this issue; high-priority/blocker/regression label present; component/oncall labels: module: nans and infs, module: nn",review early; this can change release or regression risk,2026-06-09T11:38:12Z
issue,186256,`native_batch_norm` silently produces incorrect 3rd+ order derivatives in training mode with `create_graph=True`,https://github.com/pytorch/pytorch/issues/186256,has linked PR,94,"open PR references this issue; high-priority/blocker/regression label present; component/oncall labels: module: autograd, module: correctness (silent), module: derivatives, module: double backwards",review early; this can change release or regression risk,2026-06-09T11:32:06Z
issue,185715,INTERNAL ASSERT FAILED: NYI SymInt equality in c10/core/Scalar.h when using torch.compile(dynamic=True) and torch.autograd.grad with torch.pow,https://github.com/pytorch/pytorch/issues/185715,has linked PR,94,"open PR references this issue; high-priority/blocker/regression label present; component/oncall labels: module: autograd, module: crash, module: dynamic shapes, oncall: pt2",review early; this can change release or regression risk,2026-06-08T17:11:30Z
issue,186241,TMA descriptor loads use stale `xoffset` in mix-order reduction loop,https://github.com/pytorch/pytorch/issues/186241,has linked PR,94,"open PR references this issue; high-priority/blocker/regression label present; component/oncall labels: module: correctness (silent), module: crash, module: inductor, module: reductions",review early; this can change release or regression risk,2026-06-08T15:57:02Z
issue,186453,"INTERNAL ASSERT FAILED at ""/pytorch/c10/core/AllocatorConfig.h"":124 on python torch import",https://github.com/pytorch/pytorch/issues/186453,has linked PR,94,"open PR references this issue; high-priority/blocker/regression label present; component/oncall labels: module: crash, module: cudacachingallocator",review early; this can change release or regression risk,2026-06-08T14:44:46Z
issue,186233,C++ API: at::range crashes due to division-by-zero in RangeFactories.cpp,https://github.com/pytorch/pytorch/issues/186233,has linked PR,94,"open PR references this issue; high-priority/blocker/regression label present; component/oncall labels: module: cpp, module: crash, module: edge cases, module: tensor creation",review early; this can change release or regression risk,2026-06-08T14:25:45Z
issue,185482,[inductor]`torch.cond` Branch with sliced view and scatter_add crashes with non-tensor input,https://github.com/pytorch/pytorch/issues/185482,has linked PR,94,"open PR references this issue; high-priority/blocker/regression label present; component/oncall labels: module: crash, module: inductor, oncall: pt2",review early; this can change release or regression risk,2026-06-02T21:02:48Z
issue,185247,Segmentation fault in `torch.linalg.eigvals` triggered by invalid `avg_pool3d` divisor,https://github.com/pytorch/pytorch/issues/185247,has linked PR,94,"open PR references this issue; high-priority/blocker/regression label present; component/oncall labels: module: crash, module: error checking, module: linear algebra, module: pooling",review early; this can change release or regression risk,2026-06-01T17:18:02Z
issue,185240,CUDACachingAllocator.cpp:995 NVML internal assert fires instead of clean OOMError when cudaMalloc fails on Tegra (Jetson),https://github.com/pytorch/pytorch/issues/185240,has linked PR,94,"open PR references this issue; high-priority/blocker/regression label present; component/oncall labels: module: arm, module: crash, module: cuda, module: cudacachingallocator",review early; this can change release or regression risk,2026-06-01T17:16:46Z
issue,184408,`torch.div(rounding_mode='trunc')` produces off-by-one results under `torch.compile` on CUDA,https://github.com/pytorch/pytorch/issues/184408,has linked PR,94,"open PR references this issue; high-priority/blocker/regression label present; component/oncall labels: module: inductor, oncall: pt2",review early; this can change release or regression risk,2026-05-26T18:20:10Z
issue,175154,[inductor/aot_eager] interpolate under torch.compile produces incorrect result when mode is `nearest`,https://github.com/pytorch/pytorch/issues/175154,has linked PR,94,"open PR references this issue; high-priority/blocker/regression label present; component/oncall labels: module: decompositions, module: dispatch, oncall: pt2",review early; this can change release or regression risk,2026-05-26T17:55:02Z
issue,181696,Eval-mode `BatchNorm`/`GroupNorm` produces wrong output under `torch.compile(backend='inductor')`,https://github.com/pytorch/pytorch/issues/181696,has linked PR,94,"open PR references this issue; high-priority/blocker/regression label present; component/oncall labels: module: aotdispatch, module: correctness (silent), module: decompositions, module: norms and normalization",review early; this can change release or regression risk,2026-05-26T17:54:20Z
issue,182649,torch.compile Inductor crashes on onnx2torch-converted Add-Sub-Reshape model with FakeTensor shape input,https://github.com/pytorch/pytorch/issues/182649,has linked PR,94,"open PR references this issue; high-priority/blocker/regression label present; component/oncall labels: module: crash, module: dynamo, module: faketensor, module: pt2-dispatcher",review early; this can change release or regression risk,2026-05-26T17:54:09Z
issue,182651,torch.compile Inductor crashes on onnx2torch-converted ASPP-style Conv-Add-Reshape model with FakeTensor shape input,https://github.com/pytorch/pytorch/issues/182651,has linked PR,94,"open PR references this issue; high-priority/blocker/regression label present; component/oncall labels: module: crash, module: dynamo, oncall: pt2",review early; this can change release or regression risk,2026-05-26T17:54:03Z
issue,182652,torch.compile Inductor crashes on onnx2torch-converted Conv-Reshape model with FakeTensor shape input,https://github.com/pytorch/pytorch/issues/182652,has linked PR,94,"open PR references this issue; high-priority/blocker/regression label present; component/oncall labels: module: crash, module: faketensor, module: pt2-dispatcher, module: viewing and reshaping",review early; this can change release or regression risk,2026-05-26T17:54:00Z
issue,184876,SIGSEGV when assigning to nn.Parameter.data inside a torch.func transform,https://github.com/pytorch/pytorch/issues/184876,has linked PR,94,"open PR references this issue; high-priority/blocker/regression label present; component/oncall labels: module: crash, module: functorch",review early; this can change release or regression risk,2026-05-26T17:52:30Z
issue,168126,Large numeric divergence for torch compile vs eager in bf16,https://github.com/pytorch/pytorch/issues/168126,has linked PR,94,"open PR references this issue; high-priority/blocker/regression label present; component/oncall labels: module: aotdispatch, module: decompositions, module: inductor, module: pt2-dispatcher",review early; this can change release or regression risk,2026-05-21T18:19:54Z
issue,171894,`aot_eager` backend does not support cpu pinned memory and device copy in cudagraph,https://github.com/pytorch/pytorch/issues/171894,has linked PR,94,"open PR references this issue; high-priority/blocker/regression label present; component/oncall labels: module: cuda graphs, module: dynamo, oncall: pt2",review early; this can change release or regression risk,2026-05-21T18:18:39Z
issue,184101,`torch.compile` backward crashes for `abs()`/`angle()` on transposed complex tensors,https://github.com/pytorch/pytorch/issues/184101,has linked PR,94,"open PR references this issue; high-priority/blocker/regression label present; component/oncall labels: module: complex, module: crash, module: inductor, oncall: pt2",review early; this can change release or regression risk,2026-05-19T18:10:09Z
issue,177118,[Inductor]T5 models performance dropped 10% on inductor mode after pattern match changes,https://github.com/pytorch/pytorch/issues/177118,has linked PR,94,"open PR references this issue; high-priority/blocker/regression label present; component/oncall labels: module: cuda, module: inductor, module: performance, module: regression",review early; this can change release or regression risk,2026-05-19T15:23:39Z
issue,146569,torch.compile with fullgraph=True causes overwritten variable error in versions later than torch==2.5.1,https://github.com/pytorch/pytorch/issues/146569,has linked PR,94,"open PR references this issue; high-priority/blocker/regression label present; component/oncall labels: module: inductor, module: regression, oncall: pt2",review early; this can change release or regression risk,2026-05-18T21:43:34Z
issue,184032,torch.UntypedStorage.<from_file/torch.load(weights_only=True)/_new_shared_cuda>.new() segfaults via null Allocator dereference at c10/core/StorageImpl.h:84,https://github.com/pytorch/pytorch/issues/184032,has linked PR,94,"open PR references this issue; high-priority/blocker/regression label present; component/oncall labels: module: crash, module: internals, module: serialization",review early; this can change release or regression risk,2026-05-18T17:20:12Z
issue,161783,torch.compile has runtime overhead on small graphs,https://github.com/pytorch/pytorch/issues/161783,has linked PR,94,"open PR references this issue; high-priority/blocker/regression label present; component/oncall labels: module: dynamo, module: pt2-dispatcher, oncall: pt2",review early; this can change release or regression risk,2026-05-18T00:03:17Z
issue,181690,Heap corruption when compiling models with `nn.LSTM` or `nn.GRU` using `torch.compile(backend='inductor')`,https://github.com/pytorch/pytorch/issues/181690,has linked PR,94,"open PR references this issue; high-priority/blocker/regression label present; component/oncall labels: module: crash, module: inductor, module: rnn, oncall: pt2",review early; this can change release or regression risk,2026-05-17T02:50:15Z
issue,180740,[Inductor] functorch_dp_cifar10 accuracy regression in dynamic_inductor_torchbench,https://github.com/pytorch/pytorch/issues/180740,has linked PR,94,"open PR references this issue; high-priority/blocker/regression label present; component/oncall labels: module: dynamic shapes, module: inductor, module: norms and normalization, module: regression",review early; this can change release or regression risk,2026-05-17T02:47:59Z
issue,183369,torch.compile crashes on nested_tensor_from_jagged + layer_norm with InternalTorchDynamoError,https://github.com/pytorch/pytorch/issues/183369,has linked PR,94,"open PR references this issue; high-priority/blocker/regression label present; component/oncall labels: module: crash, module: dynamic shapes, module: nestedtensor, oncall: pt2",review early; this can change release or regression risk,2026-05-13T23:05:25Z
issue,182283,INTERNAL ASSERT FAILED at /pytorch/c10/core/Scalar.cpp:42,https://github.com/pytorch/pytorch/issues/182283,has linked PR,94,open PR references this issue; high-priority/blocker/regression label present; component/oncall labels: module: autograd,review early; this can change release or regression risk,2026-05-11T17:25:44Z
issue,182200,`torch.compile` crashes on `stft`/`fftn`/`rfftn`/`ifftn` backward — meta kernel stride mismatch,https://github.com/pytorch/pytorch/issues/182200,has linked PR,94,"open PR references this issue; high-priority/blocker/regression label present; component/oncall labels: module: aotdispatch, module: crash, module: fft, module: meta tensors",review early; this can change release or regression risk,2026-05-04T16:13:06Z
issue,182217,`torch.compile` crashes on `F.layer_norm(NestedTensor)` — `InternalTorchDynamoError` in guard building,https://github.com/pytorch/pytorch/issues/182217,has linked PR,94,"open PR references this issue; high-priority/blocker/regression label present; component/oncall labels: module: crash, module: dynamic shapes, module: guards, module: nestedtensor",review early; this can change release or regression risk,2026-05-04T16:12:21Z
issue,181510,"torch.nn.utils.weight_norm triggers a UBSan null pointer error for nn.Linear(0, 1) on CPU",https://github.com/pytorch/pytorch/issues/181510,has linked PR,94,"open PR references this issue; high-priority/blocker/regression label present; component/oncall labels: module: cpu, module: crash, module: edge cases, module: empty tensor",review early; this can change release or regression risk,2026-04-28T15:48:29Z
issue,178680,"torch.compile (inductor/aot_eager) produces nondeterministic / incorrect results on model with `add_`, `clamp_`, and `squeeze`",https://github.com/pytorch/pytorch/issues/178680,has linked PR,94,"open PR references this issue; high-priority/blocker/regression label present; component/oncall labels: module: aotdispatch, module: pt2-dispatcher, oncall: pt2",review early; this can change release or regression risk,2026-03-31T18:28:00Z
issue,178089,bad-free in torch.sparse.spdiags due to missing bounds check on diagonal offsets,https://github.com/pytorch/pytorch/issues/178089,has linked PR,94,"open PR references this issue; high-priority/blocker/regression label present; component/oncall labels: module: crash, module: error checking, module: sparse",review early; this can change release or regression risk,2026-03-23T17:19:02Z
issue,139500,"Custom operators registered via decorator slower than ops registered via `torch.Library.{define, impl}`",https://github.com/pytorch/pytorch/issues/139500,has linked PR,94,"open PR references this issue; high-priority/blocker/regression label present; component/oncall labels: module: custom-operators, module: pt2-dispatcher, module: vllm, oncall: pt2",review early; this can change release or regression risk,2026-03-20T14:56:46Z
issue,153044,Unexpected float32 overflow for amp training with torch.compile,https://github.com/pytorch/pytorch/issues/153044,has linked PR,94,"open PR references this issue; high-priority/blocker/regression label present; component/oncall labels: module: pt2-dispatcher, oncall: pt2",review early; this can change release or regression risk,2026-03-11T17:33:14Z
issue,155044,"Execution of an `ExportProgram` of a model with `torch.autograd.grad(torch.sqrt(x), x, torch.ones_like(torch.sqrt(x)))` returns a `FakeTensor` instead of a real tensor",https://github.com/pytorch/pytorch/issues/155044,has linked PR,94,"open PR references this issue; high-priority/blocker/regression label present; component/oncall labels: oncall: export, oncall: pt2",review early; this can change release or regression risk,2026-03-09T23:26:34Z
issue,124863,Deprecate unsupported types in operator registration,https://github.com/pytorch/pytorch/issues/124863,has linked PR,94,"open PR references this issue; high-priority/blocker/regression label present; component/oncall labels: module: custom-operators, module: pt2-dispatcher, oncall: pt2",review early; this can change release or regression risk,2026-03-09T23:17:25Z
issue,145064,`_pdist_forward` causes segmentation fault for 3D tensor with last dimension of size 0,https://github.com/pytorch/pytorch/issues/145064,has linked PR,94,"open PR references this issue; high-priority/blocker/regression label present; component/oncall labels: module: crash, module: empty tensor, module: error checking",review early; this can change release or regression risk,2026-03-03T02:33:56Z
issue,175978,The following code snippet causes a segmentation fault when running torch.nn.lstm_cell:,https://github.com/pytorch/pytorch/issues/175978,has linked PR,94,"open PR references this issue; high-priority/blocker/regression label present; component/oncall labels: module: crash, module: error checking, module: rnn",review early; this can change release or regression risk,2026-03-02T18:10:38Z
issue,175190,[MPS] AvgPool2d/AdaptiveAvgPool2d backward crashes (SIGABRT) on channels_last inputs,https://github.com/pytorch/pytorch/issues/175190,has linked PR,94,"open PR references this issue; high-priority/blocker/regression label present; component/oncall labels: module: autograd, module: crash, module: memory format, module: mps",review early; this can change release or regression risk,2026-02-23T18:23:16Z
issue,142454,Segmentation fault (core dumped) in `torch.max_pool1d`,https://github.com/pytorch/pytorch/issues/142454,has linked PR,94,"open PR references this issue; high-priority/blocker/regression label present; component/oncall labels: module: crash, module: edge cases, module: error checking, module: nn",review early; this can change release or regression risk,2026-02-18T14:31:53Z
issue,173946,Segmentation fault in aten.lstm.data results if hx shapes are invalid (when running on CPU),https://github.com/pytorch/pytorch/issues/173946,has linked PR,94,"open PR references this issue; high-priority/blocker/regression label present; component/oncall labels: module: cpu, module: crash, module: error checking, module: rnn",review early; this can change release or regression risk,2026-02-10T00:57:48Z
issue,121343,Floating point exception on torch.lcm,https://github.com/pytorch/pytorch/issues/121343,has linked PR,94,open PR references this issue; high-priority/blocker/regression label present; component/oncall labels: module: crash,review early; this can change release or regression risk,2026-01-27T08:35:51Z
issue,173318,[docs] `torch.cuda.Stream` link broken on 2.10+,https://github.com/pytorch/pytorch/issues/173318,has linked PR,94,"open PR references this issue; high-priority/blocker/regression label present; component/oncall labels: module: docs, module: regression",review early; this can change release or regression risk,2026-01-26T22:01:55Z
issue,171838,`torch.nn.functional.pixel_shuffle` throws SIGFPE (Floating Point Exception) with large upscale_factor,https://github.com/pytorch/pytorch/issues/171838,has linked PR,94,"open PR references this issue; high-priority/blocker/regression label present; component/oncall labels: module: crash, module: edge cases, module: error checking, module: intel",review early; this can change release or regression risk,2026-01-16T15:27:23Z
issue,65761,torch.utils._pytree -> stable,https://github.com/pytorch/pytorch/issues/65761,has linked PR,94,open PR references this issue; high-priority/blocker/regression label present; needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: pytree,route to design/research decision before patch planning,2025-11-07T05:31:33Z
issue,147326,torch.export.export fails when one input is a class inheriting from torch.nn.Module,https://github.com/pytorch/pytorch/issues/147326,needs triage,89,"PyTorch issue lacks triaged label; open PR references this issue; component/oncall labels: oncall: export, oncall: pt2",review linked PR state before touching issue,2026-07-08T02:46:50Z
issue,189121,"[inductor] Single-config reductions extended by _dynamic_scale_rblock are permanently uncacheable — re-benchmark {R0_BLOCK, R0_BLOCK/2} every process; near-tied winners flip perf and numerics",https://github.com/pytorch/pytorch/issues/189121,needs triage,89,"triage signal present; open PR references this issue; component/oncall labels: module: compile-time, module: determinism, module: inductor, module: performance",review linked PR state before touching issue,2026-07-07T05:10:40Z
issue,160399,`torch.compile` causes segmentation fault with simple `ModuleList` model after multiple compile calls,https://github.com/pytorch/pytorch/issues/160399,needs triage,89,"PyTorch issue lacks triaged label; open PR references this issue; component/oncall labels: oncall: cpu inductor, oncall: pt2",review linked PR state before touching issue,2026-07-05T20:06:55Z
issue,188772,Make Fsspec filesystem checkpointing public,https://github.com/pytorch/pytorch/issues/188772,needs triage,89,PyTorch issue lacks triaged label; open PR references this issue; component/oncall labels: oncall: distributed,review linked PR state before touching issue,2026-07-02T04:37:17Z
issue,184229,torch.export with prefer_deferred_runtime_asserts_over_guards=True crashes on DTensor-parallelized models (IndexError in set_missing_meta_vals),https://github.com/pytorch/pytorch/issues/184229,needs triage,89,"PyTorch issue lacks triaged label; open PR references this issue; component/oncall labels: oncall: export, oncall: pt2",review linked PR state before touching issue,2026-06-26T20:15:58Z
issue,188193,torch::cuda::nccl::gather/scatter currently implemented as O(world_size) p2p connections on root,https://github.com/pytorch/pytorch/issues/188193,needs triage,89,"PyTorch issue lacks triaged label; open PR references this issue; component/oncall labels: oncall: distributed, oncall: distributed infra",review linked PR state before touching issue,2026-06-26T16:46:04Z
issue,185441,[JIT] Null pointer write in SourceRangeDeserializer::deserialize_source() via unchecked text_table indices,https://github.com/pytorch/pytorch/issues/185441,needs triage,89,PyTorch issue lacks triaged label; open PR references this issue; component/oncall labels: oncall: jit,review linked PR state before touching issue,2026-06-23T13:12:38Z
issue,172930,torch.export failed while exporting alpamayo model (Qwen3_vl) with symbolic error,https://github.com/pytorch/pytorch/issues/172930,needs triage,89,"PyTorch issue lacks triaged label; open PR references this issue; component/oncall labels: oncall: export, oncall: pt2",review linked PR state before touching issue,2026-06-17T17:12:59Z
issue,187480,`TestQuantizedTensor.test_qtensor_fill_per_channel_nhwc` fails on GB200,https://github.com/pytorch/pytorch/issues/187480,needs triage,89,PyTorch issue lacks triaged label; open PR references this issue; component/oncall labels: oncall: quantization,review linked PR state before touching issue,2026-06-16T18:57:47Z
issue,187138,"[DCP] load silently casts checkpoint tensors to the template dtype, truncating precision without any warning",https://github.com/pytorch/pytorch/issues/187138,needs triage,89,"triage signal present; open PR references this issue; component/oncall labels: oncall: distributed, oncall: distributed checkpointing",review linked PR state before touching issue,2026-06-15T16:14:43Z
issue,187048,[Bug][Quantization] Inconsistent numerical behavior in _choose_qparams_per_tensor and redundant code in test_quantized_tensor.py,https://github.com/pytorch/pytorch/issues/187048,needs triage,89,PyTorch issue lacks triaged label; open PR references this issue; component/oncall labels: oncall: quantization,review linked PR state before touching issue,2026-06-11T14:08:02Z
issue,186938,"ShardedTensor.device hardcodes torch.cuda.current_device(), breaks FSDP checkpoint resume on XPU/non-CUDA builds",https://github.com/pytorch/pytorch/issues/186938,needs triage,89,PyTorch issue lacks triaged label; open PR references this issue; component/oncall labels: oncall: distributed,review linked PR state before touching issue,2026-06-10T15:41:19Z
issue,178618,torch.export in PyTorch 2.10 emits INT64_MAX instead of None for aten.slice.Tensor end bounds,https://github.com/pytorch/pytorch/issues/178618,needs triage,89,"PyTorch issue lacks triaged label; open PR references this issue; component/oncall labels: oncall: export, oncall: pt2",review linked PR state before touching issue,2026-06-09T17:13:18Z
issue,143580,"RuntimeError: expect_autograd_hooks_ INTERNAL ASSERT FAILED at ""../torch/csrc/distributed/c10d/reducer.cpp""",https://github.com/pytorch/pytorch/issues/143580,needs triage,89,"triage signal present; open PR references this issue; component/oncall labels: module: ddp, oncall: distributed, oncall: distributed parallelisms",review linked PR state before touching issue,2026-06-09T16:58:39Z
issue,184841,`torch.export.export` raises `RuntimeError: Unhandled FakeTensor Device Propagation` for CUDA/MPS example inputs instead of a clear user-facing error,https://github.com/pytorch/pytorch/issues/184841,needs triage,89,"PyTorch issue lacks triaged label; open PR references this issue; component/oncall labels: oncall: export, oncall: pt2",review linked PR state before touching issue,2026-06-02T22:55:05Z
issue,185167,"DTensor: `F.nll_loss(weight=..., reduction='mean')` over a sharded batch returns wrong result",https://github.com/pytorch/pytorch/issues/185167,needs triage,89,"triage signal present; open PR references this issue; component/oncall labels: module: dtensor, module: nn, oncall: distributed, oncall: distributed parallelisms",review linked PR state before touching issue,2026-05-28T14:09:00Z
issue,163189,[export] Silent fake tensor leakage when doing buffer mutation,https://github.com/pytorch/pytorch/issues/163189,needs triage,89,"PyTorch issue lacks triaged label; open PR references this issue; component/oncall labels: oncall: export, oncall: pt2",review linked PR state before touching issue,2026-05-27T00:13:19Z
issue,184996,"torch.export.export accepts a parameter whose storage is too small, then save/runtime fails with low-level setStorage error",https://github.com/pytorch/pytorch/issues/184996,needs triage,89,"PyTorch issue lacks triaged label; open PR references this issue; component/oncall labels: oncall: export, oncall: pt2",review linked PR state before touching issue,2026-05-25T08:27:05Z
issue,181061,[RFC] M-to-N cast primitive (for RL weight rollout & pipeline parallel),https://github.com/pytorch/pytorch/issues/181061,needs triage,89,"PyTorch issue lacks triaged label; open PR references this issue; component/oncall labels: oncall: distributed, oncall: distributed infra",review linked PR state before touching issue,2026-05-23T16:27:41Z
issue,178473,ProcessGroupNCCL scalable communicator init may reuse stale UniqueNCCLID store keys,https://github.com/pytorch/pytorch/issues/178473,needs triage,89,"PyTorch issue lacks triaged label; open PR references this issue; component/oncall labels: oncall: distributed, oncall: distributed infra",review linked PR state before touching issue,2026-05-22T16:52:24Z
issue,178798,all_gather_into_tensor in gloo backend does not support stacking,https://github.com/pytorch/pytorch/issues/178798,needs triage,89,"PyTorch issue lacks triaged label; open PR references this issue; component/oncall labels: oncall: distributed, oncall: distributed infra",review linked PR state before touching issue,2026-05-22T16:51:38Z
issue,151510,"[eager] [cpu] [edge case] When processing `torch.nan_to_num-.long()`, inductor outputs the `reciprocal` of eager",https://github.com/pytorch/pytorch/issues/151510,needs triage,89,"PyTorch issue lacks triaged label; open PR references this issue; component/oncall labels: oncall: cpu inductor, oncall: pt2",review linked PR state before touching issue,2026-05-19T04:12:11Z
issue,182915,Add NVFP4-aware fused scaled_mm_v2 + reduce_scatter support in symmetric memory,https://github.com/pytorch/pytorch/issues/182915,needs triage,89,"PyTorch issue lacks triaged label; open PR references this issue; component/oncall labels: oncall: distributed, oncall: distributed infra",review linked PR state before touching issue,2026-05-17T07:58:27Z
issue,183520,Cannot JIT a deprecated class,https://github.com/pytorch/pytorch/issues/183520,needs triage,89,PyTorch issue lacks triaged label; open PR references this issue; component/oncall labels: oncall: jit,review linked PR state before touching issue,2026-05-13T14:20:25Z
issue,183207,FSDP2 with ignore_params does not work with foreach optimizers,https://github.com/pytorch/pytorch/issues/183207,needs triage,89,"PyTorch issue lacks triaged label; open PR references this issue; component/oncall labels: module: dtensor, module: fsdp, oncall: distributed, oncall: distributed parallelisms",review linked PR state before touching issue,2026-05-11T19:50:06Z
issue,181238,"torch.cuda.synchronize() syncs on device 0 when current device is not explicitly set, causing incomplete profiling results",https://github.com/pytorch/pytorch/issues/181238,needs triage,89,PyTorch issue lacks triaged label; open PR references this issue; component/oncall labels: oncall: profiler,review linked PR state before touching issue,2026-05-08T13:11:50Z
issue,180354,`torch.cond` with tensors created in `[true/false]_fn` fails during `run_decompositions`,https://github.com/pytorch/pytorch/issues/180354,needs triage,89,"PyTorch issue lacks triaged label; open PR references this issue; component/oncall labels: oncall: export, oncall: pt2",review linked PR state before touching issue,2026-05-04T21:32:24Z
issue,182102,DeviceMesh pickling warning spam in DCP process,https://github.com/pytorch/pytorch/issues/182102,needs triage,89,"PyTorch issue lacks triaged label; open PR references this issue; component/oncall labels: oncall: distributed, oncall: distributed checkpointing",review linked PR state before touching issue,2026-05-04T08:39:53Z
issue,182031,DTensor.local_map inconsistently handles pytree on input versus placements,https://github.com/pytorch/pytorch/issues/182031,needs triage,89,"PyTorch issue lacks triaged label; open PR references this issue; component/oncall labels: module: dtensor, oncall: distributed, oncall: distributed parallelisms",review linked PR state before touching issue,2026-04-30T14:46:45Z
issue,153614,Add opentelemetry traces,https://github.com/pytorch/pytorch/issues/153614,needs triage,89,PyTorch issue lacks triaged label; open PR references this issue; component/oncall labels: oncall: profiler,review linked PR state before touching issue,2026-04-28T21:19:40Z
issue,178522,AArch64 static runtimes failures,https://github.com/pytorch/pytorch/issues/178522,needs triage,89,"PyTorch issue lacks triaged label; open PR references this issue; component/oncall labels: module: arm, oncall: quantization",review linked PR state before touching issue,2026-04-20T15:40:13Z
issue,143157,can we make not(x) --> sym_not(x) automatically during tracing.,https://github.com/pytorch/pytorch/issues/143157,needs triage,89,"PyTorch issue lacks triaged label; open PR references this issue; component/oncall labels: module: dynamic shapes, oncall: export, oncall: pt2",review linked PR state before touching issue,2026-04-01T15:45:37Z
issue,177510,[export] re-raising augmented exceptions resets traceback to except block,https://github.com/pytorch/pytorch/issues/177510,needs triage,89,"PyTorch issue lacks triaged label; open PR references this issue; component/oncall labels: oncall: export, oncall: pt2",review linked PR state before touching issue,2026-03-16T09:01:48Z
issue,176428,torch.export.load() slow due to Python-side schema deserialization,https://github.com/pytorch/pytorch/issues/176428,needs triage,89,"PyTorch issue lacks triaged label; open PR references this issue; component/oncall labels: oncall: export, oncall: pt2",review linked PR state before touching issue,2026-03-09T23:28:26Z
issue,172839,_dynamo_graph_capture_for_export doesn't work on pytree output with tensor closures,https://github.com/pytorch/pytorch/issues/172839,needs triage,89,"PyTorch issue lacks triaged label; open PR references this issue; component/oncall labels: oncall: export, oncall: pt2",review linked PR state before touching issue,2026-03-09T23:27:54Z
issue,171978,`torch.export` fails on models using `pack_padded_sequence` with LSTM due to FakeTensor in dimension,https://github.com/pytorch/pytorch/issues/171978,needs triage,89,"PyTorch issue lacks triaged label; open PR references this issue; component/oncall labels: oncall: export, oncall: pt2",review linked PR state before touching issue,2026-03-09T23:27:43Z
issue,166093,aoti cross compile for windows failed with undefined reference to WinMain,https://github.com/pytorch/pytorch/issues/166093,needs triage,89,"PyTorch issue lacks triaged label; open PR references this issue; component/oncall labels: oncall: export, oncall: pt2",review linked PR state before touching issue,2026-02-27T08:57:01Z
issue,173643,"Typo ""emitDefautlArgsWithOutArgs"" in torch/csrc/jit/serialization/export_module.cpp",https://github.com/pytorch/pytorch/issues/173643,needs triage,89,PyTorch issue lacks triaged label; open PR references this issue; component/oncall labels: oncall: jit,review linked PR state before touching issue,2026-01-28T16:20:46Z
issue,173495,`torch.fbgemm_linear_quantize_weight` crashes with segmentation fault (SIGSEGV) with a cuda tensor,https://github.com/pytorch/pytorch/issues/173495,needs triage,89,"PyTorch issue lacks triaged label; open PR references this issue; component/oncall labels: module: error checking, oncall: quantization",review linked PR state before touching issue,2026-01-27T15:44:41Z
issue,170127,Could not gaurd on data-dependent expression for SDPA,https://github.com/pytorch/pytorch/issues/170127,needs triage,89,"PyTorch issue lacks triaged label; open PR references this issue; component/oncall labels: module: dynamic shapes, oncall: export, oncall: pt2",review linked PR state before touching issue,2026-01-22T06:27:13Z
issue,171964,`torch.export.export` on unflattened module fails for dynamic shapes,https://github.com/pytorch/pytorch/issues/171964,needs triage,89,"PyTorch issue lacks triaged label; open PR references this issue; component/oncall labels: oncall: export, oncall: pt2",review linked PR state before touching issue,2026-01-21T18:51:08Z
issue,172711,torch.compile produces wrong output channels for mkldnn._convolution_transpose_pointwise (FakeTensor shape inference bug),https://github.com/pytorch/pytorch/issues/172711,needs triage,89,"PyTorch issue lacks triaged label; open PR references this issue; component/oncall labels: module: correctness (silent), oncall: cpu inductor, oncall: pt2",review linked PR state before touching issue,2026-01-21T17:22:19Z
issue,171897,Torch decomposition has no effect on torch.ops.aten._fused_rms_norm,https://github.com/pytorch/pytorch/issues/171897,needs triage,89,"PyTorch issue lacks triaged label; open PR references this issue; component/oncall labels: oncall: export, oncall: pt2",review linked PR state before touching issue,2026-01-16T02:06:20Z
issue,171977,`torch.export` triggers internal state corruption (`AssertionError: cpu`) via `copy.copy` on FakeTensor,https://github.com/pytorch/pytorch/issues/171977,needs triage,89,"PyTorch issue lacks triaged label; open PR references this issue; component/oncall labels: oncall: export, oncall: pt2",review linked PR state before touching issue,2026-01-13T17:20:20Z
issue,171974,`torch.export` fails on models with KV-cache-like stateful updates using data-dependent slice assignment,https://github.com/pytorch/pytorch/issues/171974,needs triage,89,"PyTorch issue lacks triaged label; open PR references this issue; component/oncall labels: oncall: export, oncall: pt2",review linked PR state before touching issue,2026-01-13T17:18:59Z
issue,171870,[s390x/ZVECTOR] C++ compilation failure in argmax/argmin operations with torch.compile inductor backend,https://github.com/pytorch/pytorch/issues/171870,needs triage,89,"PyTorch issue lacks triaged label; open PR references this issue; component/oncall labels: module: power, oncall: cpu inductor, oncall: pt2",review linked PR state before touching issue,2026-01-13T07:14:37Z
issue,147170,[BUG][PyTorch 2.0 Export][quant]:get_source_partitions() may return different matches with same input graph,https://github.com/pytorch/pytorch/issues/147170,needs triage,89,"PyTorch issue lacks triaged label; open PR references this issue; label says issue is suitable for a scoped external contribution; old update age or low-urgency/stale label; component/oncall labels: oncall: export, oncall: pt2, oncall: quantization",review linked PR state before touching issue,2025-11-15T21:57:13Z
issue,130861,torch.fx.Tracer.record_stack_traces is broken in torch 2.4.0,https://github.com/pytorch/pytorch/issues/130861,needs triage,89,PyTorch issue lacks triaged label; open PR references this issue; label says issue is suitable for a scoped external contribution; old update age or low-urgency/stale label; component/oncall labels: oncall: export,review linked PR state before touching issue,2025-09-23T12:08:48Z
issue,170039,PyTorch crashes with std::__1::system_error: mutex lock failed: Invalid argument,https://github.com/pytorch/pytorch/issues/170039,high-priority/blocker,88,"high-priority/blocker/regression label present; needs design/research label present; component/oncall labels: module: convolution, module: crash, module: macos",route to design/research decision before patch planning,2026-02-08T13:11:27Z
issue,177730,`nanosleep()` call in `socket.cpp` delay method cannot handle signal interrupts (e.g. from sampling-based profilers),https://github.com/pytorch/pytorch/issues/177730,needs triage,83,"PyTorch issue lacks triaged label; needs design/research label present; component/oncall labels: oncall: distributed, oncall: distributed infra",route to design/research decision before patch planning,2026-05-21T17:01:27Z
issue,51856,Assignment target is transposed when using jit.script and avanced indexing,https://github.com/pytorch/pytorch/issues/51856,needs triage,83,PyTorch issue lacks triaged label; label says issue is suitable for a scoped external contribution; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2026-02-18T19:46:05Z
issue,131941,[export] Schematize nn_module_stack serialization,https://github.com/pytorch/pytorch/issues/131941,needs triage,83,PyTorch issue lacks triaged label; label says issue is suitable for a scoped external contribution; component/oncall labels: oncall: export,categorize and decide next maintainer owner/action,2026-02-02T09:14:46Z
issue,167872,exported program cannot be loaded with load_pt2: GraphModuleDeserializer' object has no attribute 'symbol_name_to_range',https://github.com/pytorch/pytorch/issues/167872,needs triage,83,"PyTorch issue lacks triaged label; open PR references this issue; needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: oncall: export, oncall: pt2",confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-12-03T22:00:26Z
issue,161053,[export] Non-strict export of `torch.cond` fails with `UnspecializedNNModuleVariable is already tracked for mutation`,https://github.com/pytorch/pytorch/issues/161053,needs triage,83,"PyTorch issue lacks triaged label; open PR references this issue; needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: onnx, oncall: export, oncall: pt2",confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-08-26T15:56:28Z
issue,177906,CPU: 2x CPU utilization regression in 2.9+ with no proportional wall-clock improvement,https://github.com/pytorch/pytorch/issues/177906,high-priority/blocker,82,"high-priority/blocker/regression label present; needs reproduction label present; component/oncall labels: module: cpu, module: multithreading, module: openmp, module: performance",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-07-03T09:21:32Z
issue,169823,Abnormal CPU usage with ComfyUI,https://github.com/pytorch/pytorch/issues/169823,high-priority/blocker,82,"high-priority/blocker/regression label present; needs reproduction label present; component/oncall labels: module: performance, module: regression",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-07-03T09:18:57Z
issue,188255,"Crash while import torch, in some special virtualization situation.",https://github.com/pytorch/pytorch/issues/188255,high-priority/blocker,82,"high-priority/blocker/regression label present; needs reproduction label present; component/oncall labels: module: arm, module: binaries, module: crash",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-06-30T06:56:13Z
issue,179214,"cudaErrorLaunchOutOfResources in CTC loss backward on RTX 5090 (sm_120, Blackwell) with CUDA 13.0",https://github.com/pytorch/pytorch/issues/179214,high-priority/blocker,82,high-priority/blocker/regression label present; needs reproduction label present; component/oncall labels: module: cuda,confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-06-23T08:05:36Z
issue,182279,Segmentation fault on RTX 5090 during repeated PyTorch CUDA forward passes,https://github.com/pytorch/pytorch/issues/182279,high-priority/blocker,82,"high-priority/blocker/regression label present; needs reproduction label present; component/oncall labels: module: crash, module: cuda",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-05-20T19:01:30Z
issue,183812,`torch.scatter_reduce` CPU performance regression (~15–58×) introduced in 2.13.0.,https://github.com/pytorch/pytorch/issues/183812,high-priority/blocker,82,"high-priority/blocker/regression label present; needs reproduction label present; component/oncall labels: module: cpu, module: performance, module: regression, module: scatter & gather ops",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-05-18T22:17:24Z
issue,183815,`F.conv2d` CPU performance regression (~4–16×) for bfloat16 inputs introduced in 2.13.0.,https://github.com/pytorch/pytorch/issues/183815,high-priority/blocker,82,"high-priority/blocker/regression label present; needs reproduction label present; component/oncall labels: module: bfloat16, module: convolution, module: cpu, module: intel",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-05-18T17:06:20Z
issue,176426,Triton kernels with 2 or more tl.load() calls segfault at runtime on sm_120 (NVIDIA RTX PRO 6000 Blackwell). The kernels compile without errors but produce invalid code that crashes on execution.,https://github.com/pytorch/pytorch/issues/176426,high-priority/blocker,82,"high-priority/blocker/regression label present; needs reproduction label present; component/oncall labels: module: crash, module: cuda, module: inductor, oncall: pt2",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-05-16T14:26:06Z
issue,175614,"Segfault during device_map=""auto"" weight materialization on RTX 5090 D v2 (GB202-240, sm_120)",https://github.com/pytorch/pytorch/issues/175614,high-priority/blocker,82,"high-priority/blocker/regression label present; needs reproduction label present; component/oncall labels: module: crash, module: cuda, module: windows",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-05-13T00:06:32Z
issue,183251,PyTorch Segmentation Fault on M2 Mac - Bug Report,https://github.com/pytorch/pytorch/issues/183251,high-priority/blocker,82,"high-priority/blocker/regression label present; needs reproduction label present; component/oncall labels: module: crash, module: macos",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-05-11T17:41:40Z
issue,178538,[BUG] [CRASH] Torch cause notebook crash on any exception (import torch),https://github.com/pytorch/pytorch/issues/178538,high-priority/blocker,82,"high-priority/blocker/regression label present; needs reproduction label present; component/oncall labels: module: crash, module: windows",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-05-08T18:55:28Z
issue,181853,[Bug] DataLoader worker segmentation fault in multi-task training with long sequences and numerical instability,https://github.com/pytorch/pytorch/issues/181853,high-priority/blocker,82,"high-priority/blocker/regression label present; needs reproduction label present; component/oncall labels: module: crash, module: dataloader",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-05-06T04:44:12Z
issue,181534,CTCLoss backward is inconsistent with its forward derivative with respect to the documented log_probs input.,https://github.com/pytorch/pytorch/issues/181534,high-priority/blocker,82,"high-priority/blocker/regression label present; needs reproduction label present; component/oncall labels: module: autograd, module: correctness (silent), module: loss, module: nn",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-04-29T22:16:15Z
issue,180458,[Bug] ValueError: I/O operation on closed file during torch.save() on Windows (PyTorch Nightly + CUDA 13.2),https://github.com/pytorch/pytorch/issues/180458,high-priority/blocker,82,"high-priority/blocker/regression label present; needs reproduction label present; component/oncall labels: module: multiprocessing, module: regression, module: serialization, module: windows",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-04-20T17:29:39Z
issue,178038,"torch.dot crashes with SIGFPE (exit code 136) on RTX 5060 Ti (Blackwell, sm_120)",https://github.com/pytorch/pytorch/issues/178038,high-priority/blocker,82,"high-priority/blocker/regression label present; needs reproduction label present; component/oncall labels: module: crash, module: cublas, module: cuda, module: linear algebra",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-04-17T12:04:36Z
issue,177561,CUDA driver error: invalid argument in fft backward when output used in subtraction,https://github.com/pytorch/pytorch/issues/177561,high-priority/blocker,82,"high-priority/blocker/regression label present; needs reproduction label present; component/oncall labels: module: autograd, module: cuda, module: fft",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-04-13T17:22:55Z
issue,178841,Importing torch before QApplication on macOS prevents Qt from discovering cocoa platform plugin (inverse of #11326),https://github.com/pytorch/pytorch/issues/178841,high-priority/blocker,82,"high-priority/blocker/regression label present; needs reproduction label present; component/oncall labels: module: binaries, module: crash, module: macos",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-04-13T17:10:27Z
issue,178892,[Windows] torch.save triggers 0xC0000005 Access Violation on RTX 4090 Laptop (WDDM Driver Conflict),https://github.com/pytorch/pytorch/issues/178892,high-priority/blocker,82,"high-priority/blocker/regression label present; needs reproduction label present; component/oncall labels: module: crash, module: cuda, module: pickle, module: serialization",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-04-13T16:39:33Z
issue,177227,Floating point exception (core dumped) with AMD processor,https://github.com/pytorch/pytorch/issues/177227,high-priority/blocker,82,"high-priority/blocker/regression label present; needs reproduction label present; component/oncall labels: module: convolution, module: cpu, module: crash, module: mkldnn",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-03-16T17:19:33Z
issue,174468,Memory leak by torch.compile,https://github.com/pytorch/pytorch/issues/174468,high-priority/blocker,82,"high-priority/blocker/regression label present; needs reproduction label present; component/oncall labels: module: memory usage, module: regression, oncall: pt2",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-03-12T02:14:09Z
issue,121101,Segmentation error for torch==2.2.1 on MacOs,https://github.com/pytorch/pytorch/issues/121101,high-priority/blocker,82,"high-priority/blocker/regression label present; needs reproduction label present; component/oncall labels: module: crash, module: intel, module: macos, module: openmp",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-02-10T04:09:37Z
issue,171606,cos CPU ops deterministic not ensured,https://github.com/pytorch/pytorch/issues/171606,high-priority/blocker,82,"high-priority/blocker/regression label present; needs reproduction label present; component/oncall labels: module: cpu, module: determinism",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-02-05T14:06:49Z
issue,113180,Higher train loss and worse evaluation metrics when using `torch.compile()`,https://github.com/pytorch/pytorch/issues/113180,high-priority/blocker,82,"high-priority/blocker/regression label present; needs reproduction label present; component/oncall labels: module: pt2 accuracy, oncall: pt2",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-01-24T21:28:00Z
pr,189286,Fix DeviceContext mode leaks from checkpoint recompute and set_default_device restore,https://github.com/pytorch/pytorch/pull/189286,ready for maintainer decision,80,"approved with no detected failing/pending checks; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",final maintainer merge/release decision,2026-07-09T00:20:54Z
pr,188012,Add sparse triangular_solve CPU fallback,https://github.com/pytorch/pytorch/pull/188012,ready for maintainer decision,80,approved with no detected failing/pending checks; PR closes or references open issue numbers,final maintainer merge/release decision,2026-07-08T20:49:23Z
pr,186969,Bump onnx to 1.22.0 (stable) and unify typing-extensions to 4.15.0,https://github.com/pytorch/pytorch/pull/186969,ready for maintainer decision,80,"approved with no detected failing/pending checks; PR closes or references open issue numbers; CI lane labels: ciflow/docker, ciflow/trunk",final maintainer merge/release decision,2026-07-08T18:44:13Z
pr,187171,[inductor] Fix test_comprehensive_addbmm_cpu_float16 precision mismatch (#187170),https://github.com/pytorch/pytorch/pull/187171,ready for maintainer decision,80,approved with no detected failing/pending checks; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,final maintainer merge/release decision,2026-07-08T18:44:12Z
pr,189265,test/inductor: relax FlexAttention invalid-block-size regex,https://github.com/pytorch/pytorch/pull/189265,ready for maintainer decision,80,approved with no detected failing/pending checks; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,final maintainer merge/release decision,2026-07-08T18:09:53Z
pr,189142,fx: Fix incorrect return_annotation for tuple types in operator schemas,https://github.com/pytorch/pytorch/pull/189142,ready for maintainer decision,80,approved with no detected failing/pending checks; PR closes or references open issue numbers,final maintainer merge/release decision,2026-07-08T17:00:08Z
pr,186502,Make context decorator frames identify the context manager,https://github.com/pytorch/pytorch/pull/186502,ready for maintainer decision,80,approved with no detected failing/pending checks; PR closes or references open issue numbers,final maintainer merge/release decision,2026-07-08T15:34:38Z
pr,187354,Match Inductor i0/i1 infinities to eager CUDA,https://github.com/pytorch/pytorch/pull/187354,ready for maintainer decision,80,"approved with no detected failing/pending checks; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk",final maintainer merge/release decision,2026-07-08T13:52:08Z
pr,188990,[FSDP] Fix to_cpu when flat_param is already on CPU,https://github.com/pytorch/pytorch/pull/188990,ready for maintainer decision,80,"approved with no detected failing/pending checks; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",final maintainer merge/release decision,2026-07-08T08:40:41Z
pr,188967,Update logging docs to match current API,https://github.com/pytorch/pytorch/pull/188967,ready for maintainer decision,80,approved with no detected failing/pending checks; PR closes or references open issue numbers,final maintainer merge/release decision,2026-07-08T07:05:30Z
pr,187969,Enable traceable DTensor RNG ops on XPU,https://github.com/pytorch/pytorch/pull/187969,ready for maintainer decision,80,"approved with no detected failing/pending checks; PR closes or references open issue numbers; CI lane labels: ciflow/dtensor, ciflow/inductor, ciflow/torchtitan, ciflow/trunk",final maintainer merge/release decision,2026-07-08T01:30:36Z
pr,189122,Fix sparse-dense mul dropping data when broadcasting a size-1 sparse dim,https://github.com/pytorch/pytorch/pull/189122,ready for maintainer decision,80,approved with no detected failing/pending checks; PR closes or references open issue numbers,final maintainer merge/release decision,2026-07-08T00:36:19Z
pr,181036,Clamp spdiags diagonal length to zero for out-of-range offsets,https://github.com/pytorch/pytorch/pull/181036,ready for maintainer decision,80,approved with no detected failing/pending checks; PR closes or references open issue numbers,final maintainer merge/release decision,2026-07-07T22:08:00Z
pr,187481,Add overflow check to FBGEMM version of quantize_val_arm,https://github.com/pytorch/pytorch/pull/187481,ready for maintainer decision,80,approved with no detected failing/pending checks; PR closes or references open issue numbers,final maintainer merge/release decision,2026-07-07T20:09:25Z
pr,188698,Make Fsspec filesystem checkpointing public and exposing cache_staged_state_dict to enable async save,https://github.com/pytorch/pytorch/pull/188698,ready for maintainer decision,80,approved with no detected failing/pending checks; PR closes or references open issue numbers; CI lane labels: ciflow/torchtitan,final maintainer merge/release decision,2026-07-07T18:23:02Z
pr,188114,Fix CK SDPA flash attention on gfx1200/gfx1201 (RDNA4),https://github.com/pytorch/pytorch/pull/188114,ready for maintainer decision,80,"approved with no detected failing/pending checks; PR closes or references open issue numbers; CI lane labels: ciflow/b200, ciflow/docker, ciflow/h100, ciflow/inductor",final maintainer merge/release decision,2026-07-07T14:35:08Z
pr,186583,Propagate contextvars to Python autograd functions,https://github.com/pytorch/pytorch/pull/186583,ready for maintainer decision,80,approved with no detected failing/pending checks; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,final maintainer merge/release decision,2026-07-07T10:42:16Z
pr,178768,Pass tensor data pointer to dlDeviceToTorchDevice from maybeCopyTensor in DLConvertor.cpp,https://github.com/pytorch/pytorch/pull/178768,ready for maintainer decision,80,approved with no detected failing/pending checks; PR closes or references open issue numbers,final maintainer merge/release decision,2026-07-07T08:21:28Z
pr,184745,distributions: fix Kumaraswamy mode formula,https://github.com/pytorch/pytorch/pull/184745,ready for maintainer decision,80,approved with no detected failing/pending checks; PR closes or references open issue numbers; CI lane labels: ciflow/trunk,final maintainer merge/release decision,2026-07-05T17:56:11Z
pr,188799,Fix callable value-opaque objects causing graph break on method access,https://github.com/pytorch/pytorch/pull/188799,ready for maintainer decision,80,"approved with no detected failing/pending checks; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",final maintainer merge/release decision,2026-07-05T09:18:17Z
pr,188378,"Fix linspace/logspace docstrings: start/end are Number, not float",https://github.com/pytorch/pytorch/pull/188378,ready for maintainer decision,80,approved with no detected failing/pending checks; PR closes or references open issue numbers,final maintainer merge/release decision,2026-07-04T07:01:07Z
pr,186492,[inductor] Fix stale xoffset for TMA in mix-order reduction loop,https://github.com/pytorch/pytorch/pull/186492,ready for maintainer decision,80,"approved with no detected failing/pending checks; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",final maintainer merge/release decision,2026-07-02T19:15:00Z
pr,188321,[c10d] Integrate ncclScatter into PT,https://github.com/pytorch/pytorch/pull/188321,ready for maintainer decision,80,approved with no detected failing/pending checks; PR closes or references open issue numbers; CI lane labels: ciflow/trunk,final maintainer merge/release decision,2026-06-29T22:05:01Z
pr,185913,[BugFix] Fix torch.kron crash on non-contiguous input (Issue #185650),https://github.com/pytorch/pytorch/pull/185913,ready for maintainer decision,80,approved with no detected failing/pending checks; PR closes or references open issue numbers; CI lane labels: ciflow/trunk,final maintainer merge/release decision,2026-06-29T09:47:27Z
pr,188287,Improve FX CodeGen process_inputs typing,https://github.com/pytorch/pytorch/pull/188287,ready for maintainer decision,80,approved with no detected failing/pending checks; PR closes or references open issue numbers,final maintainer merge/release decision,2026-06-29T07:46:43Z
pr,154279,Add Support for transposed convolution with Padding Mode 'same' and 'valid',https://github.com/pytorch/pytorch/pull/154279,ready for maintainer decision,80,approved with no detected failing/pending checks; PR closes or references open issue numbers,final maintainer merge/release decision,2026-06-27T18:40:27Z
pr,188036,Fix THPStream_richcompare reading off the end of non-Stream objects,https://github.com/pytorch/pytorch/pull/188036,ready for maintainer decision,80,approved with no detected failing/pending checks; PR closes or references open issue numbers,final maintainer merge/release decision,2026-06-26T08:44:36Z
pr,182937,Align XPU unit tests,https://github.com/pytorch/pytorch/pull/182937,ready for maintainer decision,80,approved with no detected failing/pending checks; PR closes or references open issue numbers; CI lane labels: ciflow/xpu,final maintainer merge/release decision,2026-06-26T06:58:48Z
pr,185812,Fix arange int64 fractional truncation,https://github.com/pytorch/pytorch/pull/185812,ready for maintainer decision,80,"approved with no detected failing/pending checks; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/mps",final maintainer merge/release decision,2026-06-25T05:49:49Z
pr,186489,Fix division by zero in range when step becomes zero after conversion,https://github.com/pytorch/pytorch/pull/186489,ready for maintainer decision,80,approved with no detected failing/pending checks; PR closes or references open issue numbers; CI lane labels: ciflow/mps,final maintainer merge/release decision,2026-06-25T03:17:03Z
pr,187928,test: generate native function type hint checks,https://github.com/pytorch/pytorch/pull/187928,ready for maintainer decision,80,approved with no detected failing/pending checks; PR closes or references open issue numbers,final maintainer merge/release decision,2026-06-24T12:29:19Z
pr,183697,Fix SVE blendv mask handling after bool fill,https://github.com/pytorch/pytorch/pull/183697,ready for maintainer decision,80,"approved with no detected failing/pending checks; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/trunk",final maintainer merge/release decision,2026-06-22T16:19:01Z
pr,184881,Fix SIGSEGV from functorch key contamination via Tensor.set_data,https://github.com/pytorch/pytorch/pull/184881,ready for maintainer decision,80,approved with no detected failing/pending checks; PR closes or references open issue numbers,final maintainer merge/release decision,2026-06-21T15:57:14Z
pr,169249,[ARM] Disable SVE for GCC 11.5.0 to workaround compiler bug,https://github.com/pytorch/pytorch/pull/169249,ready for maintainer decision,80,approved with no detected failing/pending checks; PR closes or references open issue numbers,final maintainer merge/release decision,2026-06-19T14:20:18Z
pr,186346,Handle invalid CUDA JITerator cache entries,https://github.com/pytorch/pytorch/pull/186346,ready for maintainer decision,80,approved with no detected failing/pending checks; PR closes or references open issue numbers; CI lane labels: ciflow/trunk,final maintainer merge/release decision,2026-06-19T05:59:07Z
pr,187396,Fix integer linspace decomposition to match eager,https://github.com/pytorch/pytorch/pull/187396,ready for maintainer decision,80,approved with no detected failing/pending checks; PR closes or references open issue numbers; CI lane labels: ciflow/trunk,final maintainer merge/release decision,2026-06-19T05:12:07Z
pr,178781,[MPS] Remove stale @skipIfMPS from test_nn.py,https://github.com/pytorch/pytorch/pull/178781,ready for maintainer decision,80,"approved with no detected failing/pending checks; PR closes or references open issue numbers; CI lane labels: ciflow/mps, ciflow/trunk",final maintainer merge/release decision,2026-06-19T01:01:59Z
pr,182323,[Mergerules][Temp] Restrict docker changes to just maintainers,https://github.com/pytorch/pytorch/pull/182323,ready for maintainer decision,80,approved with no detected failing/pending checks; PR closes or references open issue numbers,final maintainer merge/release decision,2026-06-19T01:01:50Z
pr,177069,[xpu] Increase tolerance for tests that fail due to non-deterministic operators behavior.,https://github.com/pytorch/pytorch/pull/177069,ready for maintainer decision,80,approved with no detected failing/pending checks; PR closes or references open issue numbers; CI lane labels: ciflow/trunk,final maintainer merge/release decision,2026-06-18T20:52:27Z
pr,186566,Fix ldexp gradient for negative integer exponents,https://github.com/pytorch/pytorch/pull/186566,ready for maintainer decision,80,approved with no detected failing/pending checks; PR closes or references open issue numbers; CI lane labels: ciflow/trunk,final maintainer merge/release decision,2026-06-18T14:09:56Z
pr,186570,[distributed] Fix Windows gloo allreduce crash by moving ProcessGroupGlooCuda.cpp to base_sources,https://github.com/pytorch/pytorch/pull/186570,ready for maintainer decision,80,approved with no detected failing/pending checks; PR closes or references open issue numbers,final maintainer merge/release decision,2026-06-17T17:28:56Z
pr,186836,Fix Cauchy/Gumbel icdf boundary values at q=0/q=1 (#186824),https://github.com/pytorch/pytorch/pull/186836,ready for maintainer decision,80,approved with no detected failing/pending checks; PR closes or references open issue numbers,final maintainer merge/release decision,2026-06-17T14:56:14Z
pr,187326,Fix torch.special.zeta returning NaN for q=+inf,https://github.com/pytorch/pytorch/pull/187326,ready for maintainer decision,80,approved with no detected failing/pending checks; PR closes or references open issue numbers; CI lane labels: ciflow/trunk,final maintainer merge/release decision,2026-06-16T06:06:42Z
pr,179528,[Codemod][PSS3] Upgrade fbcode/mobile-vision to Python Scientific Stack 3 (#179528),https://github.com/pytorch/pytorch/pull/179528,ready for maintainer decision,80,approved with no detected failing/pending checks; PR closes or references open issue numbers; CI lane labels: ciflow/trunk,final maintainer merge/release decision,2026-06-15T18:54:22Z
pr,182034,Add no-tree-loop-optimize pragma for apply_grad_input on GCC/AArch64,https://github.com/pytorch/pytorch/pull/182034,ready for maintainer decision,80,approved with no detected failing/pending checks; PR closes or references open issue numbers; CI lane labels: ciflow/trunk,final maintainer merge/release decision,2026-06-15T08:47:20Z
pr,178100,docs: mention torch.cuda.memory.mem_get_info in CUDA memory management notes,https://github.com/pytorch/pytorch/pull/178100,ready for maintainer decision,80,approved with no detected failing/pending checks; PR closes or references open issue numbers,final maintainer merge/release decision,2026-06-10T08:17:44Z
pr,182546,Fix sparse compressed set_data device metadata,https://github.com/pytorch/pytorch/pull/182546,ready for maintainer decision,80,approved with no detected failing/pending checks; PR closes or references open issue numbers,final maintainer merge/release decision,2026-06-09T22:15:55Z
pr,185190,Make constant SymInts hashable,https://github.com/pytorch/pytorch/pull/185190,ready for maintainer decision,80,"approved with no detected failing/pending checks; PR closes or references open issue numbers; CI lane labels: ciflow/dtensor, ciflow/dynamo, ciflow/inductor, ciflow/torchtitan",final maintainer merge/release decision,2026-06-05T08:06:05Z
pr,182344,Inductor: Fix Searchsorted/Bucketize Sliced View Boundaries,https://github.com/pytorch/pytorch/pull/182344,ready for maintainer decision,80,"approved with no detected failing/pending checks; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",final maintainer merge/release decision,2026-06-04T15:34:22Z
pr,186133,[ROCm] remove redundant @skipIfRocm from test_cudnn_attention_broken_166211,https://github.com/pytorch/pytorch/pull/186133,ready for maintainer decision,80,approved with no detected failing/pending checks; PR closes or references open issue numbers,final maintainer merge/release decision,2026-06-03T22:56:03Z
pr,179732,Fix torch.compile istft failing when length exceeds reconstructed sig…,https://github.com/pytorch/pytorch/pull/179732,ready for maintainer decision,80,approved with no detected failing/pending checks; PR closes or references open issue numbers,final maintainer merge/release decision,2026-06-02T20:34:32Z
pr,174198,Adjust test_linalg_lstsq_input_check to be compatible with other devices,https://github.com/pytorch/pytorch/pull/174198,ready for maintainer decision,80,approved with no detected failing/pending checks; PR closes or references open issue numbers; CI lane labels: ciflow/xpu,final maintainer merge/release decision,2026-06-01T07:16:27Z
pr,180742,[autograd] lazily start worker threads in the autograd engine,https://github.com/pytorch/pytorch/pull/180742,ready for maintainer decision,80,approved with no detected failing/pending checks; PR closes or references open issue numbers,final maintainer merge/release decision,2026-05-29T09:36:24Z
issue,162466,Launch one kernel take long time.,https://github.com/pytorch/pytorch/issues/162466,needs triage,77,triage signal present; needs reproduction label present; component/oncall labels: oncall: distributed,confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-06-13T16:37:21Z
issue,124007,gpu memory normal but one of gpus GPU-Util is 0 so cant train,https://github.com/pytorch/pytorch/issues/124007,needs triage,77,"triage signal present; needs reproduction label present; component/oncall labels: oncall: distributed, oncall: distributed parallelisms",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-06-05T17:04:45Z
issue,178085,DDP mode: CUDA error: an illegal memory access was encountered,https://github.com/pytorch/pytorch/issues/178085,needs triage,77,"triage signal present; needs reproduction label present; component/oncall labels: module: ddp, oncall: distributed, oncall: distributed parallelisms",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-05-28T14:13:15Z
issue,178766,torch.distributed._functional_collectives.reduce_scatter_tensor gets a wrong result,https://github.com/pytorch/pytorch/issues/178766,needs triage,77,"triage signal present; needs reproduction label present; component/oncall labels: oncall: distributed, oncall: distributed infra",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-05-23T16:24:27Z
issue,150011,[Windows][CPU] 51 UT of inductor/test_torchinductor_opinfo.py failed with CPP wrapper,https://github.com/pytorch/pytorch/issues/150011,needs triage,77,"PyTorch issue lacks triaged label; needs reproduction label present; component/oncall labels: module: windows, oncall: cpu inductor, oncall: pt2",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-02-22T22:32:45Z
issue,173907,[ARM][QNNPACK] quantized::conv2d.new not supported for TorchScript VisionTransformer models,https://github.com/pytorch/pytorch/issues/173907,needs triage,77,"PyTorch issue lacks triaged label; needs reproduction label present; component/oncall labels: module: arm, module: nnpack, oncall: quantization",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-01-31T18:08:43Z
issue,172500,`RuntimeError: device >= 0 && device < num_gpus INTERNAL ASSERT FAILED` caused by early CUDA init,https://github.com/pytorch/pytorch/issues/172500,needs triage,77,"PyTorch issue lacks triaged label; needs reproduction label present; component/oncall labels: module: cuda, oncall: profiler",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-01-22T18:39:17Z
issue,172254,Python 3.14 test error: `AttributeError: 'MobileNetV2' object has no attribute '__annotations__'.`,https://github.com/pytorch/pytorch/issues/172254,needs triage,77,"PyTorch issue lacks triaged label; needs reproduction label present; component/oncall labels: module: python version, oncall: jit",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-01-21T16:38:13Z
issue,170877,AArch64 unit test failure test_transpose_mxn_16_16_bf16_fp16 on Neoverse-V1,https://github.com/pytorch/pytorch/issues/170877,has linked PR,76,"open PR references this issue; high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: arm, module: regression",review early; this can change release or regression risk,2025-12-22T15:42:47Z
issue,168288,ICE in gcc < 13.5 for AArch64 on Neoverse-V1,https://github.com/pytorch/pytorch/issues/168288,has linked PR,76,"open PR references this issue; high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: arm, module: build, module: crash, module: third_party",review early; this can change release or regression risk,2025-11-21T10:49:47Z
issue,167086,Unchecked uint32 overflow in CUDA scan (cumsum/cumprod/cummin/etc),https://github.com/pytorch/pytorch/issues/167086,has linked PR,76,"open PR references this issue; high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: 64-bit, module: cuda",review early; this can change release or regression risk,2025-11-10T18:07:36Z
issue,123800,add `FlopCounterMode` documentation,https://github.com/pytorch/pytorch/issues/123800,has linked PR,76,open PR references this issue; high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: docs,review early; this can change release or regression risk,2025-10-17T03:14:51Z
issue,139789,"compiling ""input mutation"" followed by ""input mutation + aliasing"" does not properly recompile",https://github.com/pytorch/pytorch/issues/139789,has linked PR,76,"open PR references this issue; high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: guards, oncall: pt2",review early; this can change release or regression risk,2025-10-16T18:25:25Z
issue,27614,scatter_add allows index tensor that doesn't match input size in forward pass but fails on backward pass,https://github.com/pytorch/pytorch/issues/27614,has linked PR,76,open PR references this issue; high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash,review early; this can change release or regression risk,2025-09-29T23:18:14Z
issue,150779,Decorator `skipIfXpu` disables tests when used on class,https://github.com/pytorch/pytorch/issues/150779,has linked PR,76,"open PR references this issue; high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: ci, module: regression, module: testing, module: tests",review early; this can change release or regression risk,2025-04-15T13:13:10Z
issue,147077,[inductor] SIGSEGV when using `torch.compile` with `torch.as_strided_copy`,https://github.com/pytorch/pytorch/issues/147077,has linked PR,76,"open PR references this issue; high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: error checking, oncall: pt2",review early; this can change release or regression risk,2025-03-18T13:56:07Z
issue,146790,Segmentation Fault in `torch.ops.aten.multi_margin_loss_backward` with Empty `grad_output`,https://github.com/pytorch/pytorch/issues/146790,has linked PR,76,"open PR references this issue; high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: error checking",review early; this can change release or regression risk,2025-02-09T20:16:38Z
issue,145220,Regression in the compilation of the torch.all operation in PyTorch version 2.6.0 compared to 2.5.1,https://github.com/pytorch/pytorch/issues/145220,has linked PR,76,"open PR references this issue; high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, module: empty tensor, module: regression, oncall: pt2",review early; this can change release or regression risk,2025-01-21T19:11:06Z
issue,143649,Floating Point Exception (core dumped) when running floordiv/remainder/fmod under torch.compile,https://github.com/pytorch/pytorch/issues/143649,has linked PR,76,"open PR references this issue; high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, oncall: cpu inductor, oncall: pt2",review early; this can change release or regression risk,2024-12-25T06:58:11Z
pr,188980,Fix inconsistent accelerator guard in checkpoint staging test,https://github.com/pytorch/pytorch/pull/188980,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-07-09T02:09:58Z
pr,189193,[XCCL] Allow XCCL to be set as default in `_new_process_group_helper`,https://github.com/pytorch/pytorch/pull/189193,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-09T01:53:22Z
pr,181559,[XPU] Enable _flash_attention_forward/_flash_attention_backward,https://github.com/pytorch/pytorch/pull/181559,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/trunk, ciflow/xpu","identify whether block is CI, merge conflict, or review gate",2026-07-09T01:52:50Z
pr,189317,[doc] Add accelerator test reuse guide (#181474),https://github.com/pytorch/pytorch/pull/189317,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers,"identify whether block is CI, merge conflict, or review gate",2026-07-09T01:41:05Z
pr,188521,Fix OrderingBarrier additional_buffer_deps key mismatch in control_deps lowering,https://github.com/pytorch/pytorch/pull/188521,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-09T01:03:51Z
pr,186252,[inductor] Fix truncdiv off-by-one on CUDA with _div_rn,https://github.com/pytorch/pytorch/pull/186252,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-09T00:21:46Z
pr,188978,"swap isinstance(t, FakeTensor) to is_fake",https://github.com/pytorch/pytorch/pull/188978,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/docker, ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-08T23:23:41Z
pr,186358,Fix output mismtach for F.softshrink with bfloat16 and float scalar,https://github.com/pytorch/pytorch/pull/186358,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-08T22:48:00Z
pr,189129,inductor: Fix non-deterministic gradients in slice_scatter backward fusion,https://github.com/pytorch/pytorch/pull/189129,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T21:51:00Z
pr,189294,[CI] Realign scipy after nvmath install to fix H100 smoke NumPy ABI crash,https://github.com/pytorch/pytorch/pull/189294,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/h100,"identify whether block is CI, merge conflict, or review gate",2026-07-08T21:23:06Z
pr,188931,Fix matrix-vector linear fallback crashes with 1D weight and bias,https://github.com/pytorch/pytorch/pull/188931,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers,"identify whether block is CI, merge conflict, or review gate",2026-07-08T21:16:55Z
pr,188913,[inductor] Add triton.cudagraph_max_recorded_sizes to cap dynamic shape re-recording,https://github.com/pytorch/pytorch/pull/188913,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T21:14:32Z
pr,188998,[inductor] Fix bf16 matmul accumulation precision in Triton templates (#188492),https://github.com/pytorch/pytorch/pull/188998,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T21:14:23Z
pr,188996,[inductor] Fix FP8 blockwise scaling shape validation in Triton (#188711),https://github.com/pytorch/pytorch/pull/188996,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/b200, ciflow/h100, ciflow/inductor, ciflow/rocm-mi300","identify whether block is CI, merge conflict, or review gate",2026-07-08T21:13:10Z
pr,189141,"[test] Add opt-in sign-sensitive special-value parity check (copysign, atan2, reciprocal)",https://github.com/pytorch/pytorch/pull/189141,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers,"identify whether block is CI, merge conflict, or review gate",2026-07-08T20:56:01Z
pr,187989,[MPS] Fix linear backward SIGABRT with a 1-D weight,https://github.com/pytorch/pytorch/pull/187989,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/b200, ciflow/docker, ciflow/dtensor, ciflow/h100","identify whether block is CI, merge conflict, or review gate",2026-07-08T20:52:32Z
pr,181074,[inductor][coordesc] extract get_neighbor_configs and get_all_direction_configs (#181074),https://github.com/pytorch/pytorch/pull/181074,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T20:49:30Z
pr,188784,Fix `torch.asarray` for buffer-protocol objects with no dtype,https://github.com/pytorch/pytorch/pull/188784,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers,"identify whether block is CI, merge conflict, or review gate",2026-07-08T20:48:55Z
pr,189149,"[inductor] Emit pointwise kernel for small mm (K<8, N<8) instead of cuBLAS",https://github.com/pytorch/pytorch/pull/189149,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/b200, ciflow/h100, ciflow/inductor, ciflow/rocm-mi300","identify whether block is CI, merge conflict, or review gate",2026-07-08T20:37:31Z
pr,188965,Fix non-deterministic node ordering in get_source_partitions(),https://github.com/pytorch/pytorch/pull/188965,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers,"identify whether block is CI, merge conflict, or review gate",2026-07-08T20:33:53Z
pr,184279,Avoid MultiOutput for single-output fallback kernels,https://github.com/pytorch/pytorch/pull/184279,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-08T19:50:16Z
pr,186349,Improve Dynamo logger method graph-break guidance,https://github.com/pytorch/pytorch/pull/186349,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T19:47:13Z
pr,186355,Fix AOTAutograd unused output backward pruning,https://github.com/pytorch/pytorch/pull/186355,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T19:43:09Z
pr,183884,Fix compile worker quiesce draining,https://github.com/pytorch/pytorch/pull/183884,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-08T19:35:57Z
pr,186364,Prevent guard validation from mutating ShapeEnv,https://github.com/pytorch/pytorch/pull/186364,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T19:32:00Z
pr,186372,Fix torch.Tensor construction from traced NumPy arrays,https://github.com/pytorch/pytorch/pull/186372,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T19:28:15Z
pr,186406,Fix invalid source in AOT repro generation,https://github.com/pytorch/pytorch/pull/186406,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T19:24:07Z
pr,186415,Fix Dynamo Parameter construction in new modules,https://github.com/pytorch/pytorch/pull/186415,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T19:21:06Z
pr,186403,Avoid CUDA init during CPU torch.compile,https://github.com/pytorch/pytorch/pull/186403,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-08T19:13:22Z
pr,186849,Fix CuTeDSL grouped_mm descriptor init shape,https://github.com/pytorch/pytorch/pull/186849,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T19:12:36Z
pr,186425,Skip Dynamo inside non-traceable subclass dispatch hooks,https://github.com/pytorch/pytorch/pull/186425,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T19:08:07Z
pr,184124,Fix Inductor multi-output replacement insertion order,https://github.com/pytorch/pytorch/pull/184124,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T18:48:16Z
pr,184343,Prune slow Hopper WGMMA autotune configs,https://github.com/pytorch/pytorch/pull/184343,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-08T18:45:11Z
pr,186455,Fix AOTAutograd Dynamo memory leak repro,https://github.com/pytorch/pytorch/pull/186455,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T18:34:47Z
pr,186457,Fix torch.ops float parsing for scalar tensor args,https://github.com/pytorch/pytorch/pull/186457,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T18:30:28Z
pr,186460,Support dynamic mode in Dynamo explain,https://github.com/pytorch/pytorch/pull/186460,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T18:21:50Z
pr,185776,Propagate fake RuntimeErrors to Dynamo handlers,https://github.com/pytorch/pytorch/pull/185776,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T18:10:20Z
pr,186466,Fix scalar Tensor args for custom op float schemas in Dynamo,https://github.com/pytorch/pytorch/pull/186466,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T17:57:44Z
pr,186467,Preserve non-persistent buffers in compiled modules,https://github.com/pytorch/pytorch/pull/186467,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T17:54:05Z
pr,186557,Resolve scalar overloads before AOT compilation,https://github.com/pytorch/pytorch/pull/186557,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T17:50:03Z
pr,189233,[MPS] Fix avg_pool2d/adaptive_avg_pool2d backward on channels_last inputs,https://github.com/pytorch/pytorch/pull/189233,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/mps,"identify whether block is CI, merge conflict, or review gate",2026-07-08T17:32:04Z
pr,185947,Fix Dynamo logit with tensor eps,https://github.com/pytorch/pytorch/pull/185947,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T17:24:18Z
pr,185952,Support compiled async all_reduce wait,https://github.com/pytorch/pytorch/pull/185952,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T17:20:37Z
pr,185971,Fix Dynamo random.Random replay on cache hits,https://github.com/pytorch/pytorch/pull/185971,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T17:07:28Z
pr,185972,Handle malformed MSVC help output,https://github.com/pytorch/pytorch/pull/185972,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T17:03:22Z
pr,186011,[dynamo] skip empty graph fallback recursively,https://github.com/pytorch/pytorch/pull/186011,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T16:52:11Z
pr,186016,Reject invalid bitwise dtypes in Inductor lowering,https://github.com/pytorch/pytorch/pull/186016,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T16:48:18Z
pr,186021,[Inductor][CPU] Guard compiled integer div by zero,https://github.com/pytorch/pytorch/pull/186021,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T16:39:16Z
pr,186472,Graph break on dynamic unspecialized ModuleList indexing,https://github.com/pytorch/pytorch/pull/186472,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T16:31:06Z
pr,186487,Run Dynamo compilation with empty functorch stack,https://github.com/pytorch/pytorch/pull/186487,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T16:24:08Z
pr,186248,Simplify Min/Max using ShapeEnv value ranges,https://github.com/pytorch/pytorch/pull/186248,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-08T16:17:54Z
pr,184048,Enable Inductor OpInfo CPU coverage,https://github.com/pytorch/pytorch/pull/184048,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-08T16:17:39Z
pr,186493,Optimize AOTAutograd compiled forward wrapper,https://github.com/pytorch/pytorch/pull/186493,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2026-07-08T16:17:07Z
pr,186494,[Dynamo] Constant-fold CUDA Tensor.is_pinned(),https://github.com/pytorch/pytorch/pull/186494,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T16:12:01Z
pr,188047,fix(dlpack): raise Python RuntimeError instead of aborting for negative DLPack strides,https://github.com/pytorch/pytorch/pull/188047,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-07-08T16:06:27Z
pr,186498,Add dynamic shape test assertions for input dims,https://github.com/pytorch/pytorch/pull/186498,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2026-07-08T15:58:12Z
pr,186500,Fix Dynamo resume after f-string logging graph breaks,https://github.com/pytorch/pytorch/pull/186500,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T15:45:31Z
pr,184392,Fix ModularIndexing printer semantics,https://github.com/pytorch/pytorch/pull/184392,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/mps, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T15:41:30Z
pr,186504,Allow HOP speculation through torch autograd.Function skipfiles,https://github.com/pytorch/pytorch/pull/186504,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T15:28:59Z
pr,186507,Invalidate Dynamo caches when module hooks change,https://github.com/pytorch/pytorch/pull/186507,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T15:20:55Z
pr,186512,[dynamo] Graph break on shape-changing Parameter.data assignment,https://github.com/pytorch/pytorch/pull/186512,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T15:09:42Z
pr,186520,Preserve SymInt bindings in dynamic-shape minifier repros,https://github.com/pytorch/pytorch/pull/186520,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T15:01:48Z
pr,186529,Preserve grad state in Dynamo minifier repros,https://github.com/pytorch/pytorch/pull/186529,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T14:45:26Z
pr,186530,[dynamo] Restore autocast dtype after trace-time updates,https://github.com/pytorch/pytorch/pull/186530,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T14:41:41Z
pr,186534,Account for fanout in Inductor reuse realization,https://github.com/pytorch/pytorch/pull/186534,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T14:28:35Z
pr,186538,Fix CUDA driver linking for Inductor CUTLASS builds,https://github.com/pytorch/pytorch/pull/186538,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T14:17:49Z
pr,189224,[xpu][skip] Skip test_combo_kernel_dynamic_scale_rblock on XPU,https://github.com/pytorch/pytorch/pull/189224,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/trunk, ciflow/xpu","identify whether block is CI, merge conflict, or review gate",2026-07-08T13:52:22Z
pr,186539,[Dynamo] Limit FX node growth from loop unrolling,https://github.com/pytorch/pytorch/pull/186539,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T13:38:21Z
pr,186361,Improve autograd.Function backward graph grad names,https://github.com/pytorch/pytorch/pull/186361,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T13:36:08Z
pr,185304,Fix while_loop stride metadata preservation,https://github.com/pytorch/pytorch/pull/185304,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/dtensor, ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T13:12:21Z
pr,186091,Preserve CUDA LSTM in Inductor,https://github.com/pytorch/pytorch/pull/186091,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T12:47:20Z
pr,185280,Fix Dynamo automatic dynamic int specialization,https://github.com/pytorch/pytorch/pull/185280,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-08T12:27:31Z
pr,188049,Fix Inductor div floor nonfinite CUDA results,https://github.com/pytorch/pytorch/pull/188049,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-08T12:27:26Z
pr,186546,Preserve original Dynamo minifier backend failures,https://github.com/pytorch/pytorch/pull/186546,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T12:09:12Z
pr,186549,Fix fake deepcopy for custom module mappings,https://github.com/pytorch/pytorch/pull/186549,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T12:06:30Z
pr,186553,[dynamo] Graph break on packed sequence lower ops,https://github.com/pytorch/pytorch/pull/186553,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T12:03:34Z
pr,186551,Preserve symbolic tensor metadata in after-AOT repros,https://github.com/pytorch/pytorch/pull/186551,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T12:01:01Z
pr,187057,Fix value-opaque graph output replay,https://github.com/pytorch/pytorch/pull/187057,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T11:53:46Z
pr,186395,Preserve overlap deps without blocking fusion,https://github.com/pytorch/pytorch/pull/186395,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T11:46:42Z
pr,185871,Fix symbolic stride-order sorting without explicit ShapeEnv,https://github.com/pytorch/pytorch/pull/185871,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-08T11:27:18Z
pr,185873,Preserve mutation renames after scheduler recompute,https://github.com/pytorch/pytorch/pull/185873,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T11:24:49Z
pr,185324,Reject mixed-device fake convolution inputs,https://github.com/pytorch/pytorch/pull/185324,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-08T11:15:02Z
pr,186476,Fix Dynamo resume tensor lifetimes across graph breaks,https://github.com/pytorch/pytorch/pull/186476,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T11:14:32Z
pr,185876,Prevent saving mutated grad inputs in AOT min-cut,https://github.com/pytorch/pytorch/pull/185876,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T11:05:12Z
pr,186246,Fix stale scalar value in addcdiv graph-break decomposition,https://github.com/pytorch/pytorch/pull/186246,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T11:00:22Z
pr,184508,Fix export with unused DTensor param attrs,https://github.com/pytorch/pytorch/pull/184508,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/dtensor,"identify whether block is CI, merge conflict, or review gate",2026-07-08T10:33:24Z
pr,186564,Skip uninitialized outputs during accuracy minification,https://github.com/pytorch/pytorch/pull/186564,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T10:26:53Z
pr,185939,Fix int8 WoQ GEMM residual epilogue fusion,https://github.com/pytorch/pytorch/pull/185939,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T10:22:03Z
pr,186380,Run python decompositions under functionalization,https://github.com/pytorch/pytorch/pull/186380,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2026-07-08T10:01:47Z
pr,186446,Fix fake overlap handling for Dynamo try/except,https://github.com/pytorch/pytorch/pull/186446,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T09:56:39Z
pr,186505,Simplify mod-one symbolic shape guards,https://github.com/pytorch/pytorch/pull/186505,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2026-07-08T09:54:20Z
pr,186462,Allow cond branch mutation during export,https://github.com/pytorch/pytorch/pull/186462,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T09:50:39Z
pr,186923,Graph break on custom op fake tensor failures,https://github.com/pytorch/pytorch/pull/186923,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T09:45:51Z
pr,186485,Fix export dynamic dims with size-one examples,https://github.com/pytorch/pytorch/pull/186485,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T09:42:45Z
pr,187493,Fix Inductor nextafter for low precision CUDA tensors,https://github.com/pytorch/pytorch/pull/187493,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T09:39:13Z
pr,187311,Preserve forward AD tangents for Inductor dual inputs,https://github.com/pytorch/pytorch/pull/187311,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T09:33:04Z
pr,184323,Share foreach combo kernel bodies,https://github.com/pytorch/pytorch/pull/184323,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-08T09:32:51Z
pr,183654,Fix Inductor CPU group norm affine precision,https://github.com/pytorch/pytorch/pull/183654,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T09:26:14Z
pr,184595,Allow functorch grad fake propagation in Dynamo,https://github.com/pytorch/pytorch/pull/184595,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T09:20:54Z
pr,187486,Fix unbacked memo reuse during fake replay,https://github.com/pytorch/pytorch/pull/187486,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T09:18:17Z
pr,185469,Fix stable TMA descriptor reconstruction in Dynamo,https://github.com/pytorch/pytorch/pull/185469,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T08:54:06Z
pr,185804,Fix non-strict export tensor len handling,https://github.com/pytorch/pytorch/pull/185804,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-08T08:41:16Z
pr,183511,Fix Inductor addmm beta-zero input handling,https://github.com/pytorch/pytorch/pull/183511,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/b200, ciflow/h100, ciflow/inductor, ciflow/rocm-mi300","identify whether block is CI, merge conflict, or review gate",2026-07-08T08:41:01Z
pr,186388,Avoid native eager router graph breaks under Dynamo,https://github.com/pytorch/pytorch/pull/186388,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T08:41:00Z
pr,186656,Autotune XBLOCK for mix-order reductions,https://github.com/pytorch/pytorch/pull/186656,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T08:40:37Z
pr,184656,Fix CPU SDPA higher-order grad compile,https://github.com/pytorch/pytorch/pull/184656,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2026-07-08T08:38:02Z
pr,186857,Use real flex block counts for autotune inputs,https://github.com/pytorch/pytorch/pull/186857,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T08:33:36Z
pr,185878,Fix Inductor expm1 on float subnormal inputs,https://github.com/pytorch/pytorch/pull/185878,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T08:28:20Z
pr,183744,Update ACL dependencies for BF16 inner product,https://github.com/pytorch/pytorch/pull/183744,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/docker, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-08T08:23:33Z
pr,183854,Filter C++-wrapper-incompatible extern autotune choices,https://github.com/pytorch/pytorch/pull/183854,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/b200, ciflow/h100, ciflow/inductor, ciflow/rocm-mi300","identify whether block is CI, merge conflict, or review gate",2026-07-08T08:17:45Z
pr,185501,Fix Issue #185396 inductor complex64 triton signature,https://github.com/pytorch/pytorch/pull/185501,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-08T07:45:41Z
pr,186150,Fix compile storage semantics for as_strided scatter,https://github.com/pytorch/pytorch/pull/186150,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T07:32:08Z
pr,183866,Preserve native matmul lowp output precision,https://github.com/pytorch/pytorch/pull/183866,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/b200, ciflow/h100, ciflow/inductor, ciflow/rocm-mi300","identify whether block is CI, merge conflict, or review gate",2026-07-08T07:22:01Z
pr,184713,Support Dynamo class setattr for hook handles,https://github.com/pytorch/pytorch/pull/184713,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T07:19:25Z
pr,184793,Fix no-tensor-aliasing recompile reason for stale sources,https://github.com/pytorch/pytorch/pull/184793,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T07:05:29Z
pr,183862,[Inductor] Preserve view_as_complex input layout before fake prop,https://github.com/pytorch/pytorch/pull/183862,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T06:38:21Z
pr,185063,Handle scalar carries in Inductor while_loop,https://github.com/pytorch/pytorch/pull/185063,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T06:19:11Z
pr,185800,Fix TORCH_LOGS stderr handling at process exit,https://github.com/pytorch/pytorch/pull/185800,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-08T06:14:19Z
pr,185908,[Dynamo] Preserve live dict view semantics around mappingproxy mutation,https://github.com/pytorch/pytorch/pull/185908,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T06:11:11Z
pr,185085,[dynamo] Support datetime.now attrs as runtime inputs,https://github.com/pytorch/pytorch/pull/185085,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T06:07:41Z
pr,160585,Add dim=None support to torch.logsumexp,https://github.com/pytorch/pytorch/pull/160585,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/dtensor, ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T05:58:24Z
pr,183994,Specialize numpy scalar module attrs,https://github.com/pytorch/pytorch/pull/183994,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T05:50:01Z
pr,186913,Fix int64 value_expr arithmetic in Triton codegen,https://github.com/pytorch/pytorch/pull/186913,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T05:47:06Z
pr,188327,[test] Add error_inputs for nn.Conv2d module,https://github.com/pytorch/pytorch/pull/188327,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers,"identify whether block is CI, merge conflict, or review gate",2026-07-08T05:39:05Z
pr,187614,Use sym_numel in linalg_cond string overload,https://github.com/pytorch/pytorch/pull/187614,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2026-07-08T04:57:51Z
pr,183652,Fix dynamic slice lowering for generic views,https://github.com/pytorch/pytorch/pull/183652,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-08T04:50:00Z
pr,184934,Validate packed RNN batch_sizes before fallback,https://github.com/pytorch/pytorch/pull/184934,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers,"identify whether block is CI, merge conflict, or review gate",2026-07-08T04:29:01Z
pr,184950,Fix mkldnn transposed conv pointwise shape inference,https://github.com/pytorch/pytorch/pull/184950,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-08T04:19:40Z
pr,184803,[dynamo] Avoid key-set guard for dict setitem,https://github.com/pytorch/pytorch/pull/184803,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T03:53:31Z
pr,183681,Fix Triton compile-time autotune fallback,https://github.com/pytorch/pytorch/pull/183681,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/b200, ciflow/h100, ciflow/inductor, ciflow/rocm-mi300","identify whether block is CI, merge conflict, or review gate",2026-07-08T03:40:48Z
pr,186845,[make_fx] Decompose detach by default for higher-order grads,https://github.com/pytorch/pytorch/pull/186845,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2026-07-08T03:40:38Z
pr,184838,Fix Dynamo raw descriptor building,https://github.com/pytorch/pytorch/pull/184838,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T03:35:58Z
pr,184848,Fix nearest upsample scale shortcuts in compile,https://github.com/pytorch/pytorch/pull/184848,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T03:32:45Z
pr,184864,Error on overlapping as_strided mutations in compile,https://github.com/pytorch/pytorch/pull/184864,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers,"identify whether block is CI, merge conflict, or review gate",2026-07-08T02:34:02Z
pr,185108,Prevent non-strict export from leaking fake buffers,https://github.com/pytorch/pytorch/pull/185108,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2026-07-08T02:29:12Z
pr,183731,[inductor] Validate fallback output dtype metadata,https://github.com/pytorch/pytorch/pull/183731,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-08T02:23:00Z
pr,179376,fix(export): add SDPA decomposition for contiguous output,https://github.com/pytorch/pytorch/pull/179376,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers,"identify whether block is CI, merge conflict, or review gate",2026-07-08T02:18:50Z
pr,184964,Handle tensor scalar slice bounds in export setitem,https://github.com/pytorch/pytorch/pull/184964,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2026-07-08T02:13:43Z
pr,184967,Fix export CIA decomps with backend kernels,https://github.com/pytorch/pytorch/pull/184967,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers,"identify whether block is CI, merge conflict, or review gate",2026-07-08T02:08:48Z
pr,185815,Fix Dynamo NumPy scalar identity handling,https://github.com/pytorch/pytorch/pull/185815,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T02:02:30Z
pr,184977,Fix Dynamo wrapper method sources for nested compile,https://github.com/pytorch/pytorch/pull/184977,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T01:48:44Z
pr,188641,Mark linalg.eigh as cudagraph-unsafe in Inductor,https://github.com/pytorch/pytorch/pull/188641,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T01:43:28Z
pr,185953,Clear Dynamo weakrefs before graph-break resumes,https://github.com/pytorch/pytorch/pull/185953,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T01:37:41Z
pr,185288,Support symbolic list lengths in Triton grids,https://github.com/pytorch/pytorch/pull/185288,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T01:35:05Z
pr,186027,Fix symbolic broadcast inference for unbacked sizes,https://github.com/pytorch/pytorch/pull/186027,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/dtensor, ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T01:14:58Z
pr,188636,Handle reciprocal SymPy powers in shape range eval,https://github.com/pytorch/pytorch/pull/188636,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2026-07-08T01:11:14Z
pr,186893,Fix dynamic size-one tensor recompiles,https://github.com/pytorch/pytorch/pull/186893,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/inductor-pallas, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T01:08:57Z
pr,184988,Allow HOPs to pass through unchanged input tensors,https://github.com/pytorch/pytorch/pull/184988,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T01:06:12Z
pr,184990,[dynamo] Route fake RuntimeError through user exception handlers,https://github.com/pytorch/pytorch/pull/184990,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T01:03:39Z
pr,186032,Avoid reshape metadata guards on symbolic size-one checks,https://github.com/pytorch/pytorch/pull/186032,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2026-07-08T00:58:50Z
pr,184628,Fix SingletonInt static guard evaluation,https://github.com/pytorch/pytorch/pull/184628,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2026-07-08T00:41:33Z
pr,185033,Support Dynamo Tensor.backward by default,https://github.com/pytorch/pytorch/pull/185033,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T00:29:13Z
pr,185034,Enable Dynamo capture of torch.autograd.grad,https://github.com/pytorch/pytorch/pull/185034,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T00:25:51Z
pr,185042,Avoid value guards for residual lazy constants,https://github.com/pytorch/pytorch/pull/185042,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T00:20:19Z
pr,184590,Fix uncached NestedTensor seqlens under compile,https://github.com/pytorch/pytorch/pull/184590,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T00:13:03Z
pr,184920,[dynamo] Store FunctionIdSet entries by identity weakrefs,https://github.com/pytorch/pytorch/pull/184920,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T23:53:56Z
pr,185775,Avoid storage access for no-storage tensors in MetaConverter,https://github.com/pytorch/pytorch/pull/185775,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2026-07-07T23:50:14Z
pr,186886,Fix keyed user Triton autotune in Inductor,https://github.com/pytorch/pytorch/pull/186886,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T23:43:59Z
pr,185100,Keep fake tensor cache synthesis out of proxy traces,https://github.com/pytorch/pytorch/pull/185100,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2026-07-07T23:35:34Z
pr,185103,Fix unbacked scalar equality reasoning through div/mod,https://github.com/pytorch/pytorch/pull/185103,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2026-07-07T23:31:06Z
pr,185116,Thread effect tokens through cond branches,https://github.com/pytorch/pytorch/pull/185116,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T23:21:55Z
pr,184007,[inductor] Clear stale subprocess cache env vars,https://github.com/pytorch/pytorch/pull/184007,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/b200, ciflow/h100, ciflow/inductor, ciflow/rocm-mi300","identify whether block is CI, merge conflict, or review gate",2026-07-07T23:19:03Z
pr,182224,Enable skipped static_runtime tests on AArch64,https://github.com/pytorch/pytorch/pull/182224,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-07-07T22:49:48Z
pr,189191,Cuda Graph Default Enable (#189191),https://github.com/pytorch/pytorch/pull/189191,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T22:49:03Z
pr,185138,[dynamo] Enforce one compiled frame for fullgraph skipped frames,https://github.com/pytorch/pytorch/pull/185138,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T22:37:36Z
pr,185149,Reject partial-overlap copy aliases,https://github.com/pytorch/pytorch/pull/185149,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2026-07-07T22:28:51Z
pr,185159,Inductor: gate max-autotune 1x1 conv GEMM lowering,https://github.com/pytorch/pytorch/pull/185159,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/b200, ciflow/h100, ciflow/inductor, ciflow/rocm-mi300","identify whether block is CI, merge conflict, or review gate",2026-07-07T22:21:35Z
pr,184047,Enable cooperative reductions by default,https://github.com/pytorch/pytorch/pull/184047,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/inductor-perf-compare, ciflow/inductor-periodic, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T21:46:44Z
pr,175566,Fix test_extension_backend race condition,https://github.com/pytorch/pytorch/pull/175566,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2026-07-07T20:11:47Z
pr,184276,[inductor] fix torch.randperm for slice_shape node in fx_passes,https://github.com/pytorch/pytorch/pull/184276,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/b200, ciflow/dtensor, ciflow/h100, ciflow/inductor","identify whether block is CI, merge conflict, or review gate",2026-07-07T19:39:05Z
pr,185084,Avoid contiguity guards on maybe-singleton symbolic dims,https://github.com/pytorch/pytorch/pull/185084,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers,"identify whether block is CI, merge conflict, or review gate",2026-07-07T19:23:43Z
pr,185946,[CUDA] Warn instead of assert in `reportProcessMemoryInfo` to prevent crash on Tegra OOM,https://github.com/pytorch/pytorch/pull/185946,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-07-07T19:21:47Z
pr,185300,Fix AttrProxy module reuse in Dynamo,https://github.com/pytorch/pytorch/pull/185300,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T19:17:11Z
pr,185320,Improve scan HOP direct call error,https://github.com/pytorch/pytorch/pull/185320,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T19:11:00Z
pr,184059,Fix split reduction kernel metadata,https://github.com/pytorch/pytorch/pull/184059,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T19:02:38Z
pr,185336,[dynamo] Add regression test for prepended forward pre-hooks,https://github.com/pytorch/pytorch/pull/185336,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2026-07-07T18:55:57Z
pr,185349,Fix flex decoding BLOCK_M validation for short GQA,https://github.com/pytorch/pytorch/pull/185349,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T18:47:13Z
pr,185356,Fix DTensor embedding mask state under compile,https://github.com/pytorch/pytorch/pull/185356,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/dtensor, ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T18:38:46Z
pr,186003,Fix Dynamo condition-dependent frame skip caching,https://github.com/pytorch/pytorch/pull/186003,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T18:37:28Z
pr,185357,Forward container methods on OptimizedModule,https://github.com/pytorch/pytorch/pull/185357,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T18:33:56Z
pr,186622,Fix bf16 autocast precision barriers in Inductor,https://github.com/pytorch/pytorch/pull/186622,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T18:09:44Z
pr,185365,Fix FX subgraph rewrite of adjacent aggregate matches,https://github.com/pytorch/pytorch/pull/185365,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers,"identify whether block is CI, merge conflict, or review gate",2026-07-07T18:06:57Z
pr,188828,Skip conv_transpose3d redispatch test on Windows CI - flaky test,https://github.com/pytorch/pytorch/pull/188828,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-07-07T18:03:47Z
pr,185371,Fix symbolic antialiased resize export shapes,https://github.com/pytorch/pytorch/pull/185371,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-07T17:59:04Z
pr,185373,Fix assume_constant_result tensor constants after graph breaks,https://github.com/pytorch/pytorch/pull/185373,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T17:55:12Z
pr,184103,Fix FakeTensorUpdater HOP dtype propagation,https://github.com/pytorch/pytorch/pull/184103,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-07T17:48:51Z
pr,185419,[dynamo] avoid stale verbose hint for backend failures,https://github.com/pytorch/pytorch/pull/185419,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T17:26:18Z
pr,185436,Preserve eager exception assertions for eager_noexcept,https://github.com/pytorch/pytorch/pull/185436,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T17:22:40Z
pr,185438,Scope distribution validation disabling to Dynamo tracing,https://github.com/pytorch/pytorch/pull/185438,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T17:20:01Z
pr,185451,Wrap internal Dynamo assertion errors,https://github.com/pytorch/pytorch/pull/185451,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T17:12:06Z
pr,185057,Support torch.cuda.use_mem_pool in compile,https://github.com/pytorch/pytorch/pull/185057,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T17:08:52Z
pr,185461,[dynamo] Handle module-level logging ignores,https://github.com/pytorch/pytorch/pull/185461,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T17:06:15Z
pr,185462,Fix NumPy ndarray operators under default device contexts,https://github.com/pytorch/pytorch/pull/185462,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T17:01:40Z
pr,185466,[dynamo] Show carets in graph break user stacks,https://github.com/pytorch/pytorch/pull/185466,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T16:57:14Z
pr,185467,Keep tensors alive for Dynamo pointer reconstruction,https://github.com/pytorch/pytorch/pull/185467,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T16:53:45Z
pr,185471,Handle Dynamo tracing of torch._C._infer_size,https://github.com/pytorch/pytorch/pull/185471,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T16:50:04Z
pr,185474,Fix nested tensor tangent metadata coercion,https://github.com/pytorch/pytorch/pull/185474,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2026-07-07T16:41:41Z
pr,184086,Fix Inductor scaled_mm scale_result lowering,https://github.com/pytorch/pytorch/pull/184086,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/b200, ciflow/h100, ciflow/inductor, ciflow/rocm-mi300","identify whether block is CI, merge conflict, or review gate",2026-07-07T16:30:06Z
pr,184090,Preserve benchmark integer inputs for Inductor debug,https://github.com/pytorch/pytorch/pull/184090,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T16:12:27Z
pr,185504,Support unbacked SymInt closures in cond export,https://github.com/pytorch/pytorch/pull/185504,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T16:09:25Z
pr,185508,[dynamo] Preserve observed exception stack for reraises,https://github.com/pytorch/pytorch/pull/185508,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T16:02:55Z
pr,185528,Fix Dynamo tracing of compile support checks,https://github.com/pytorch/pytorch/pull/185528,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T15:50:42Z
pr,185536,Teach Dynamo PGO about sparse tensor strides,https://github.com/pytorch/pytorch/pull/185536,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T15:46:33Z
pr,185545,Handle symbolic bounds in slice decomposition,https://github.com/pytorch/pytorch/pull/185545,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2026-07-07T15:43:08Z
pr,185568,Fix export of arange with data-dependent float step,https://github.com/pytorch/pytorch/pull/185568,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T15:30:11Z
pr,185588,[inductor] Codegen backed SymInt input runtime asserts,https://github.com/pytorch/pytorch/pull/185588,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T15:19:08Z
pr,185591,Canonicalize avg_pool2d singleton lists during export,https://github.com/pytorch/pytorch/pull/185591,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2026-07-07T15:14:17Z
pr,185604,Improve infer_size unbacked broadcast error,https://github.com/pytorch/pytorch/pull/185604,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T15:11:46Z
pr,185624,Fix subclass static input remapping for SymInt args,https://github.com/pytorch/pytorch/pull/185624,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2026-07-07T15:02:30Z
pr,185632,Make AOTAutograd metadata repr subclass-safe,https://github.com/pytorch/pytorch/pull/185632,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2026-07-07T14:59:34Z
pr,184109,Fix Inductor replacement patterns across view-to-reshape,https://github.com/pytorch/pytorch/pull/184109,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T14:52:56Z
pr,185722,[dynamo] Wrap module.compile leaf module call frames,https://github.com/pytorch/pytorch/pull/185722,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T14:10:05Z
pr,184119,Avoid fp8 bf16 prologue fusion on pre-sm90,https://github.com/pytorch/pytorch/pull/184119,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/b200, ciflow/h100, ciflow/inductor, ciflow/rocm-mi300","identify whether block is CI, merge conflict, or review gate",2026-07-07T14:03:15Z
pr,185724,Fix scalar searchsorted decomposition for unbacked values,https://github.com/pytorch/pytorch/pull/185724,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T13:58:50Z
pr,185726,Improve Dynamo stack for nonstrict_trace input errors,https://github.com/pytorch/pytorch/pull/185726,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T13:55:55Z
pr,185731,Clear stale prefix exception table entries in Dynamo,https://github.com/pytorch/pytorch/pull/185731,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T13:50:06Z
pr,185732,Fix nested compile during subclass fake tracing,https://github.com/pytorch/pytorch/pull/185732,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-07T13:44:45Z
pr,185734,Preserve torch._check messages in runtime asserts,https://github.com/pytorch/pytorch/pull/185734,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2026-07-07T13:40:16Z
pr,185740,Support inspect.signature on nested functions in Dynamo,https://github.com/pytorch/pytorch/pull/185740,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T13:32:17Z
pr,185742,Support None leaves in nonstrict_trace,https://github.com/pytorch/pytorch/pull/185742,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T13:28:04Z
pr,185730,Fix dynamic shapes for variadic kwargs,https://github.com/pytorch/pytorch/pull/185730,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers,"identify whether block is CI, merge conflict, or review gate",2026-07-07T13:27:38Z
pr,188783,[xpu][fix]Fix bmm_outer_product override to support XPU tensors,https://github.com/pytorch/pytorch/pull/188783,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/trunk, ciflow/xpu","identify whether block is CI, merge conflict, or review gate",2026-07-07T13:27:33Z
pr,185902,Enable torch.compile for RNN modules by default,https://github.com/pytorch/pytorch/pull/185902,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T13:19:06Z
pr,185752,Make OptimizedModule state_dict use public keys,https://github.com/pytorch/pytorch/pull/185752,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T13:14:15Z
pr,185762,[Dynamo] Support _AutoDispatchBelowAutograd context manager,https://github.com/pytorch/pytorch/pull/185762,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T13:07:26Z
pr,185763,Improve disable API graph break messages,https://github.com/pytorch/pytorch/pull/185763,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2026-07-07T13:03:37Z
pr,185766,Dynamo: support torch.jit.isinstance without graph breaks,https://github.com/pytorch/pytorch/pull/185766,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T13:00:43Z
pr,185768,Improve Dynamo error for dynamic module hook registration,https://github.com/pytorch/pytorch/pull/185768,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T12:56:11Z
pr,185769,Preserve generator return values in Dynamo,https://github.com/pytorch/pytorch/pull/185769,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2026-07-07T12:46:39Z
pr,185771,Support Dynamo tracing through textwrap.indent string formatting,https://github.com/pytorch/pytorch/pull/185771,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T12:44:02Z
pr,186469,Wrap Dynamo-generated bytecode runtime errors,https://github.com/pytorch/pytorch/pull/186469,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T12:40:35Z
pr,185773,Use max dynamic bounds for Inductor autotuning,https://github.com/pytorch/pytorch/pull/185773,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T12:33:23Z
pr,185456,Fix Dynamo tracing for Tensor.random_,https://github.com/pytorch/pytorch/pull/185456,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T12:26:47Z
pr,185781,Avoid unbound symbols in insignificant output strides,https://github.com/pytorch/pytorch/pull/185781,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T12:20:40Z
pr,185784,Fix flex attention captured vector gradient scatters,https://github.com/pytorch/pytorch/pull/185784,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T12:07:52Z
pr,185795,Preserve overlapping input layouts when cloning Dynamo inputs,https://github.com/pytorch/pytorch/pull/185795,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T11:28:13Z
pr,185799,Avoid dataclass constructors in Dynamo and make_fx graphs,https://github.com/pytorch/pytorch/pull/185799,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T11:24:34Z
pr,185809,Fix as_strided_copy storage bounds in Inductor,https://github.com/pytorch/pytorch/pull/185809,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T11:09:45Z
pr,185916,Fix FlexAttention nested index backward,https://github.com/pytorch/pytorch/pull/185916,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T11:01:20Z
pr,185823,[Dynamo] Handle class method_descriptor calls generically,https://github.com/pytorch/pytorch/pull/185823,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T10:58:48Z
pr,185920,[Dynamo] Detect tensor mutations when reconstructing generators,https://github.com/pytorch/pytorch/pull/185920,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T10:53:51Z
pr,185829,Fix autograd.grad graph breaks under aot_eager,https://github.com/pytorch/pytorch/pull/185829,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T10:24:02Z
pr,184820,[dynamo] respect custom metaclass __call__ during class construction,https://github.com/pytorch/pytorch/pull/184820,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T09:07:38Z
pr,186043,Lower symbolic bool not to torch.sym_not,https://github.com/pytorch/pytorch/pull/186043,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T09:04:49Z
pr,186051,Fix disabled no-op module pre-hooks in compile,https://github.com/pytorch/pytorch/pull/186051,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T09:01:09Z
pr,186052,Fix Inductor C++ pybinding source literal escaping,https://github.com/pytorch/pytorch/pull/186052,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T08:58:18Z
pr,186554,Preserve hooks for dependent AOTAutograd outputs,https://github.com/pytorch/pytorch/pull/186554,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T08:54:46Z
pr,187589,Make FX symbolic tracing state thread-local,https://github.com/pytorch/pytorch/pull/187589,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2026-07-07T08:44:58Z
pr,184185,Use memory stats for persistent reductions,https://github.com/pytorch/pytorch/pull/184185,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T08:42:28Z
pr,186053,Fix AOTAutograd cache save for call-free lazy backwards,https://github.com/pytorch/pytorch/pull/186053,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T08:37:22Z
pr,184189,Make generated Triton code use runtime device properties,https://github.com/pytorch/pytorch/pull/184189,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T08:34:30Z
pr,186058,Preserve export graph metadata during normalization,https://github.com/pytorch/pytorch/pull/186058,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T08:30:24Z
pr,186063,[dynamo] Route OptimizedModule delegated methods through compiled call,https://github.com/pytorch/pytorch/pull/186063,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T08:27:25Z
pr,186069,Fix Dynamo round on NumPy scalar results,https://github.com/pytorch/pytorch/pull/186069,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T08:24:40Z
pr,189009,[xpu][fix] Enable test_fallback_kernel_with_symexpr_output on XPU with correct head_dim,https://github.com/pytorch/pytorch/pull/189009,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-07T08:21:52Z
pr,186614,Delay RNN graph break until module call,https://github.com/pytorch/pytorch/pull/186614,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T07:24:16Z
pr,186072,Make constant warnings reorderable in Dynamo by default,https://github.com/pytorch/pytorch/pull/186072,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2026-07-07T07:20:53Z
pr,186898,Add always-on AOTI input rank checks,https://github.com/pytorch/pytorch/pull/186898,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T07:18:27Z
pr,185866,Fix CPU Inductor numerics for BatchNorm erf normalize,https://github.com/pytorch/pytorch/pull/185866,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T07:15:18Z
pr,186078,Fix OptimizedModule _parameters access,https://github.com/pytorch/pytorch/pull/186078,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T07:07:26Z
pr,184460,Remove obsolete Python wrapper symbol extraction,https://github.com/pytorch/pytorch/pull/184460,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-07T07:05:51Z
pr,188556,[inductor] Preserve eager semantics for infinite Bessel inputs,https://github.com/pytorch/pytorch/pull/188556,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T07:04:47Z
pr,187353,Preserve signed zero in Triton neg lowering,https://github.com/pytorch/pytorch/pull/187353,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T07:03:49Z
pr,185333,Graph break existing list tensor assignments in Dynamo,https://github.com/pytorch/pytorch/pull/185333,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T07:01:15Z
pr,186094,Fix AOTI cond predicate unbacked symbol dependency,https://github.com/pytorch/pytorch/pull/186094,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2026-07-07T06:55:13Z
pr,186900,Improve pytree missing registration error,https://github.com/pytorch/pytorch/pull/186900,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2026-07-07T06:43:21Z
pr,186942,[inductor] Round fmod/remainder scalars for bf16/fp16 CPU,https://github.com/pytorch/pytorch/pull/186942,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T06:40:25Z
pr,186119,Improve Dynamo observed exception messages,https://github.com/pytorch/pytorch/pull/186119,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T06:36:59Z
pr,186166,Respect fake/functional modes for backend autograd kernels,https://github.com/pytorch/pytorch/pull/186166,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers,"identify whether block is CI, merge conflict, or review gate",2026-07-07T06:29:52Z
pr,186044,Stop Dynamo from importing optional libraries at import time,https://github.com/pytorch/pytorch/pull/186044,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-07T06:24:21Z
pr,186168,Fix Dynamo tracing of torch.random.fork_rng,https://github.com/pytorch/pytorch/pull/186168,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T06:19:28Z
pr,186180,Fix Dynamo cProfile under active Python profilers,https://github.com/pytorch/pytorch/pull/186180,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T06:16:58Z
pr,186920,Route overrideable SDPA AOTI calls through dispatcher,https://github.com/pytorch/pytorch/pull/186920,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T06:06:41Z
pr,184367,Fix AotAutograd aliased-output replay for traceable wrapper subclasses,https://github.com/pytorch/pytorch/pull/184367,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/dtensor, ciflow/inductor, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-07T06:00:06Z
pr,186192,Fix CompiledFxGraph callable state across threads,https://github.com/pytorch/pytorch/pull/186192,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T05:58:43Z
pr,184256,[inductor] Cap reduction XBLOCK to max block,https://github.com/pytorch/pytorch/pull/184256,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T05:50:52Z
pr,184705,[dynamo] Preserve exception chaining in InternalTorchDynamoError rewrapping,https://github.com/pytorch/pytorch/pull/184705,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T05:47:17Z
pr,186501,Collapse foreach input mutation copies in AOTAutograd,https://github.com/pytorch/pytorch/pull/186501,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T05:40:00Z
pr,186210,Fix cond branch-local shape assertions,https://github.com/pytorch/pytorch/pull/186210,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T05:36:21Z
pr,186214,Profile lazy backward compilation with cProfile,https://github.com/pytorch/pytorch/pull/186214,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T05:32:50Z
pr,186217,Move FX graph cache hash details out of debug logs,https://github.com/pytorch/pytorch/pull/186217,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T05:28:59Z
pr,186222,Include epilogue guards in Dynamo code_parts,https://github.com/pytorch/pytorch/pull/186222,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T05:23:25Z
pr,186249,[dynamo] Support time.time as a runtime scalar input,https://github.com/pytorch/pytorch/pull/186249,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T05:05:55Z
pr,187890,Fix Dynamo metadata sync after inplace tensor mutations,https://github.com/pytorch/pytorch/pull/187890,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T04:57:36Z
pr,186263,Increase small fake tensor constant limit,https://github.com/pytorch/pytorch/pull/186263,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T04:55:06Z
pr,186280,Disable profiler Python tracing during compile,https://github.com/pytorch/pytorch/pull/186280,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T04:50:59Z
pr,186287,Reject non-functional export decomposition entries,https://github.com/pytorch/pytorch/pull/186287,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2026-07-07T04:46:25Z
pr,184294,Avoid unused storage size guards for indexing,https://github.com/pytorch/pytorch/pull/184294,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T04:40:13Z
pr,186340,Ensure explicit compile attempts skipped code,https://github.com/pytorch/pytorch/pull/186340,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T04:35:47Z
pr,188546,[Inductor] Fix log_sigmoid_forward flushing float32 subnormals to +0.0 on CUDA,https://github.com/pytorch/pytorch/pull/188546,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T03:07:28Z
pr,188466,Fix inductor scatter crash with empty index,https://github.com/pytorch/pytorch/pull/188466,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-07T00:07:42Z
pr,184133,[aot_autograd] Reduce runtime_wrapper Python overhead: 4 micro-optimizations,https://github.com/pytorch/pytorch/pull/184133,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2026-07-06T22:02:10Z
pr,175081,Add shim config includes for linting,https://github.com/pytorch/pytorch/pull/175081,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/binaries, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-06T21:59:20Z
pr,132135,"[torch.special] Adding betainc, betaincc, betaincinv, betainccinv, betaln and beta with backward operation",https://github.com/pytorch/pytorch/pull/132135,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-06T21:23:21Z
pr,188286,Fix out= variant recompilations for CompositeImplicitAutograd linalg ops,https://github.com/pytorch/pytorch/pull/188286,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2026-07-06T20:53:33Z
pr,188902,[ATen] Add reduction tag to missing base reduction ops,https://github.com/pytorch/pytorch/pull/188902,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers,"identify whether block is CI, merge conflict, or review gate",2026-07-06T20:52:22Z
pr,188406,[dynamo] preserve Stream/Event subclass type across graph-break rec…,https://github.com/pytorch/pytorch/pull/188406,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-06T20:13:36Z
pr,188853,Register DTensorSpec as an opaque reference type for torch.export,https://github.com/pytorch/pytorch/pull/188853,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/dtensor, ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-06T20:11:31Z
pr,188859,Raise a clear error when linalg.eigh/svd workspace exceeds 32-bit LAPACK limit,https://github.com/pytorch/pytorch/pull/188859,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers,"identify whether block is CI, merge conflict, or review gate",2026-07-06T18:51:14Z
pr,186294,[dynamo] Proxy OptimizedModule __class__ to wrapped module,https://github.com/pytorch/pytorch/pull/186294,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-06T18:38:06Z
pr,186306,Support tensor torch._assert in export,https://github.com/pytorch/pytorch/pull/186306,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-06T18:35:10Z
pr,186874,Fix Dynamo resume lifetime for _EnableTorchFunction,https://github.com/pytorch/pytorch/pull/186874,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-06T18:26:10Z
pr,186319,Avoid full simplify in vector-range index analysis,https://github.com/pytorch/pytorch/pull/186319,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-06T18:15:25Z
pr,186330,Fix compiled autograd.Function manual vmap rules,https://github.com/pytorch/pytorch/pull/186330,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-06T18:03:39Z
pr,184305,Fix zero-bias SDPA flash selection,https://github.com/pytorch/pytorch/pull/184305,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-06T17:13:30Z
pr,186336,Support multiword CXX commands in Inductor cpp_builder,https://github.com/pytorch/pytorch/pull/186336,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-06T17:06:55Z
pr,187596,Fall back on AOTAutograd version-check capture errors,https://github.com/pytorch/pytorch/pull/187596,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2026-07-06T17:03:15Z
pr,185791,Fix cond branch shape guards leaking to export constraints,https://github.com/pytorch/pytorch/pull/185791,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-06T16:59:35Z
pr,188268,Fix positional arg overflow in TritonBenchmarkRequest when template inputs are deduplicated,https://github.com/pytorch/pytorch/pull/188268,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/b200, ciflow/h100, ciflow/inductor, ciflow/rocm-mi300","identify whether block is CI, merge conflict, or review gate",2026-07-06T10:01:56Z
pr,188328,[meta] Add output_padding and bias validation to transposed convolution meta path,https://github.com/pytorch/pytorch/pull/188328,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-07-06T05:32:45Z
pr,179619,[PowerPC] Guard deviceTypeFromActivity with USE_KINETO check,https://github.com/pytorch/pytorch/pull/179619,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers,"identify whether block is CI, merge conflict, or review gate",2026-07-05T22:54:03Z
pr,187782,[dynamo] Skip guard creation for unused function inputs,https://github.com/pytorch/pytorch/pull/187782,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-05T14:29:15Z
pr,188401,dtensor: include local tensor device in _stable_hash_for_caching,https://github.com/pytorch/pytorch/pull/188401,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/dtensor, ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-05T14:03:48Z
pr,187664,[MPS] Fix `baddbmm` operation to handle zero-sized dimensions correctly,https://github.com/pytorch/pytorch/pull/187664,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/mps,"identify whether block is CI, merge conflict, or review gate",2026-07-05T12:45:29Z
pr,186479,[dynamo] Respect mark_dynamic on Parameter inputs,https://github.com/pytorch/pytorch/pull/186479,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-05T09:40:32Z
pr,168999,[Pt. 1] Make some aspects of Dynamo/Inductor global state thread safe,https://github.com/pytorch/pytorch/pull/168999,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-05T05:49:00Z
pr,178664,Implement igamma/igammac backward w.r.t. first argument,https://github.com/pytorch/pytorch/pull/178664,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers,"identify whether block is CI, merge conflict, or review gate",2026-07-04T22:49:12Z
pr,188394,Speed up CPU quantile/nanquantile with partial selection,https://github.com/pytorch/pytorch/pull/188394,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers,"identify whether block is CI, merge conflict, or review gate",2026-07-04T19:03:01Z
pr,182216,Add Half/BFloat16 copysign overloads to fix NaN sign bit loss on CUDA,https://github.com/pytorch/pytorch/pull/182216,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers,"identify whether block is CI, merge conflict, or review gate",2026-07-04T18:50:13Z
pr,187653,Add NanDetectMode for forward-pass NaN/Inf detection,https://github.com/pytorch/pytorch/pull/187653,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers,"identify whether block is CI, merge conflict, or review gate",2026-07-04T15:13:42Z
pr,175082,[pytree] make `torch.utils.pytree` as public API,https://github.com/pytorch/pytorch/pull/175082,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-04T12:38:27Z
pr,188291,"Fix dims docstring for torch.flip and torch.rot90 (axes, element type)",https://github.com/pytorch/pytorch/pull/188291,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers,"identify whether block is CI, merge conflict, or review gate",2026-07-04T07:01:06Z
pr,175947,[dynamo] Remove the dict over guarding on dict setitem,https://github.com/pytorch/pytorch/pull/175947,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/dynamo, ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-03T00:10:44Z
pr,186127,[AI Codemod][PerfAICT-General] fbcode/caffe2/torch/csrc/jit/serialization/pickler.cpp (#186127),https://github.com/pytorch/pytorch/pull/186127,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-07-02T19:51:19Z
pr,181582,[AI Codemod][RlAutotuneFbsourceWearablesMemory] torch::jit::mobile::Function::append_instruction (function.cpp) (#181582),https://github.com/pytorch/pytorch/pull/181582,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-07-02T19:15:32Z
pr,188426,Fix incorrect CPU bf16 conv3d when innermost output dim is 1 and strided,https://github.com/pytorch/pytorch/pull/188426,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers,"identify whether block is CI, merge conflict, or review gate",2026-07-02T17:33:35Z
pr,185911,[dynamo] Keep relational guards out of dict-tag skips,https://github.com/pytorch/pytorch/pull/185911,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/tsan","identify whether block is CI, merge conflict, or review gate",2026-07-02T09:24:15Z
pr,186532,Fix index_add decomposition bounds checks,https://github.com/pytorch/pytorch/pull/186532,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/dtensor, ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-02T08:37:48Z
pr,187032,[dynamo] Support member access on protobuf enums,https://github.com/pytorch/pytorch/pull/187032,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-01T21:09:05Z
pr,182092,Allow torch.meshgrid() with empty input to return empty tuple,https://github.com/pytorch/pytorch/pull/182092,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers,"identify whether block is CI, merge conflict, or review gate",2026-07-01T21:03:34Z
pr,188337,[inductor] Handle non-comparable arguments in index_propagation Min/Max,https://github.com/pytorch/pytorch/pull/188337,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-01T12:50:55Z
pr,185661,[fx] Fix pytree codegen for tuple/dict positional inputs,https://github.com/pytorch/pytorch/pull/185661,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-01T11:59:59Z
pr,181571,Generalize test_pooling.py to be device-agnostic and enable it on XPU,https://github.com/pytorch/pytorch/pull/181571,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-07-01T09:44:24Z
pr,185738,Make HOP alias and mutation support opt-in,https://github.com/pytorch/pytorch/pull/185738,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-01T08:19:03Z
pr,185988,Fix FlexAttention scalar captured grad backward,https://github.com/pytorch/pytorch/pull/185988,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-01T02:59:52Z
pr,187786,Support int step tensors in foreach optimizer implementations,https://github.com/pytorch/pytorch/pull/187786,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers,"identify whether block is CI, merge conflict, or review gate",2026-07-01T01:54:05Z
pr,187207,Use abs() in lp_pool to avoid NaN for non-integer norm_type,https://github.com/pytorch/pytorch/pull/187207,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers,"identify whether block is CI, merge conflict, or review gate",2026-07-01T01:54:04Z
pr,188368,[Inductor] Fix ValueRangeError in Inductor for zero-iteration loops,https://github.com/pytorch/pytorch/pull/188368,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-01T01:22:00Z
pr,185891,[dynamo] Guard AOTAutograd input mutation overlap changes,https://github.com/pytorch/pytorch/pull/185891,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-30T18:31:24Z
pr,188511,[AI Codemod][PerfAICT-General] Replace cast with castRaw in TensorType::isSubtypeOfExt to avoid shared_ptr refcount overhead (#188511),https://github.com/pytorch/pytorch/pull/188511,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-06-30T18:30:30Z
pr,188357,[ROCm] Add MIOpen convolution ops to FORBIDDEN_CUDAGRAPH_OPS for cudagraph safety,https://github.com/pytorch/pytorch/pull/188357,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-30T17:57:21Z
pr,176044,[opaque object] Disallow creation of reference-type opaque objects under custom ops in torch.compile,https://github.com/pytorch/pytorch/pull/176044,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers,"identify whether block is CI, merge conflict, or review gate",2026-06-30T17:19:21Z
pr,188549,[AI Codemod][PerfAICT-General] Vectorize index_select bounds check via fused unsigned compare (#188549),https://github.com/pytorch/pytorch/pull/188549,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers,"identify whether block is CI, merge conflict, or review gate",2026-06-30T16:30:32Z
pr,176084,ci: Cleanup dead centos code,https://github.com/pytorch/pytorch/pull/176084,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/binaries_wheel,"identify whether block is CI, merge conflict, or review gate",2026-06-30T05:55:32Z
pr,187714,test: Command to reproduce error is incorrect,https://github.com/pytorch/pytorch/pull/187714,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers,"identify whether block is CI, merge conflict, or review gate",2026-06-30T04:53:50Z
pr,187927,docs: document FlopCounterMode semantics,https://github.com/pytorch/pytorch/pull/187927,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-06-29T22:34:38Z
pr,188320,[c10d] Integrate ncclGather into PT,https://github.com/pytorch/pytorch/pull/188320,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-06-29T22:04:19Z
pr,188410,Use typing_extensions.TypeAliasType for better reexport of __module__,https://github.com/pytorch/pytorch/pull/188410,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/dtensor, ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-29T21:21:22Z
pr,188314,Fix tutorial link for GPU Direct Storage,https://github.com/pytorch/pytorch/pull/188314,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers,"identify whether block is CI, merge conflict, or review gate",2026-06-29T18:14:58Z
pr,169241,[XPU] Migrate 10 dynamo test cases for XPU,https://github.com/pytorch/pytorch/pull/169241,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/xpu","identify whether block is CI, merge conflict, or review gate",2026-06-29T11:55:14Z
pr,169798,"[XPU] Migrate 2 test cases autograd, creation_ops for XPU",https://github.com/pytorch/pytorch/pull/169798,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/xpu,"identify whether block is CI, merge conflict, or review gate",2026-06-29T11:55:00Z
pr,188380,docs: fix broken gpu_direct_storage tutorial URL in serialization notes,https://github.com/pytorch/pytorch/pull/188380,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers,"identify whether block is CI, merge conflict, or review gate",2026-06-29T11:54:58Z
pr,187871,Fix scaled_dot_product_attention: explicitly check for both attn_mask…,https://github.com/pytorch/pytorch/pull/187871,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/mps,"identify whether block is CI, merge conflict, or review gate",2026-06-28T23:36:42Z
pr,181561,[AI Codemod][DevmateCppModernizeFindToContains] fbcode/caffe2/torch/csrc/jit/backends/xnnpack/xnnpack_graph_builder.cpp (#181561),https://github.com/pytorch/pytorch/pull/181561,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-06-28T20:49:58Z
pr,181516,Fix heap-buffer-overflow in torch.linalg.eig on subnormal real inputs,https://github.com/pytorch/pytorch/pull/181516,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers,"identify whether block is CI, merge conflict, or review gate",2026-06-27T17:50:56Z
pr,181475,Add `tensor` and `from_numpy` to `torch.__all__`,https://github.com/pytorch/pytorch/pull/181475,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers,"identify whether block is CI, merge conflict, or review gate",2026-06-27T15:53:33Z
pr,184986,Handle unbatched input for non-zeros padding modes in Conv1d/2d/3d,https://github.com/pytorch/pytorch/pull/184986,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers,"identify whether block is CI, merge conflict, or review gate",2026-06-27T08:37:46Z
pr,185177,Fix DTensor nll_loss weighted mean bug (closes #185167),https://github.com/pytorch/pytorch/pull/185177,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/dtensor, ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-26T20:50:55Z
pr,180561,[optim] Support ndim >= 2 in Muon and add param_groups_for_muon helper,https://github.com/pytorch/pytorch/pull/180561,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers,"identify whether block is CI, merge conflict, or review gate",2026-06-26T09:09:03Z
pr,181546,Fix AssertionError: failed to set ranges (#181546),https://github.com/pytorch/pytorch/pull/181546,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-26T07:48:29Z
pr,185077,Avoid folding mutated buffer storages,https://github.com/pytorch/pytorch/pull/185077,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-26T01:27:26Z
pr,179592,Fix MPS SDPA correctness when score matrix exceeds 2^32 elements,https://github.com/pytorch/pytorch/pull/179592,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/mps,"identify whether block is CI, merge conflict, or review gate",2026-06-25T22:02:58Z
pr,183509,[distributed] Add fused scaled_mm_v2 + reduce_scatter for NVFP4 support,https://github.com/pytorch/pytorch/pull/183509,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/h100-symm-mem, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-25T17:01:21Z
pr,184193,Preserve FX graph cache guard provenance,https://github.com/pytorch/pytorch/pull/184193,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-25T11:35:14Z
pr,186779,Error on unsupported batch norm third derivatives,https://github.com/pytorch/pytorch/pull/186779,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2026-06-25T07:44:26Z
pr,187368,[inductor] add input validation to normal and bernoulli decompositions,https://github.com/pytorch/pytorch/pull/187368,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/inductor-pallas, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-25T06:03:51Z
pr,184704,Capturable param in optimizers restored using custom vts and cleanup hooks,https://github.com/pytorch/pytorch/pull/184704,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-25T04:59:44Z
pr,181481,[Codemod][MigrateInternmcUrls] fbcode/caffe2 [A] (#181481),https://github.com/pytorch/pytorch/pull/181481,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-06-25T01:24:33Z
pr,175526,while_loop vmap batching rule,https://github.com/pytorch/pytorch/pull/175526,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-06-25T01:18:10Z
pr,175246,Fix NLLLoss backward failure for non-contiguous 4D inputs,https://github.com/pytorch/pytorch/pull/175246,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers,"identify whether block is CI, merge conflict, or review gate",2026-06-24T21:58:09Z
pr,186085,Fix transient dispatch-mode skips in Dynamo,https://github.com/pytorch/pytorch/pull/186085,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-24T19:58:40Z
pr,186178,Fix matmul folding when an inner folded dim has size 1,https://github.com/pytorch/pytorch/pull/186178,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-06-24T19:21:25Z
pr,182342,[Inductor] Deal with choices that fail to compile in async pipelined autotuning path (#182342),https://github.com/pytorch/pytorch/pull/182342,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/b200, ciflow/h100, ciflow/inductor, ciflow/rocm-mi300","identify whether block is CI, merge conflict, or review gate",2026-06-24T16:00:27Z
pr,183668,[ROCm] fix flaky test_honor_sm_carveout,https://github.com/pytorch/pytorch/pull/183668,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/b200, ciflow/h100, ciflow/rocm-mi300","identify whether block is CI, merge conflict, or review gate",2026-06-24T06:14:11Z
pr,178584,Remove unused expected failure decorator from FakeTensorOperatorInvariants test and clean up condition in Module class,https://github.com/pytorch/pytorch/pull/178584,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/dynamo, ciflow/inductor, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-23T19:24:57Z
pr,187726,Expose public API for torch.cuda.clear_cublas_workspaces,https://github.com/pytorch/pytorch/pull/187726,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/h100-symm-mem, ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-23T15:22:42Z
pr,181252,[Codemod][MigrateInternmcUrls] fbcode/caffe2 (#181252),https://github.com/pytorch/pytorch/pull/181252,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-06-22T23:54:49Z
pr,166876,[ci][cpu] Update AArch64 manylinux compiler to GCC14,https://github.com/pytorch/pytorch/pull/166876,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/docker, ciflow/linux-aarch64, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-22T16:03:37Z
pr,180557,[docs] Improve documentation of normalization layers,https://github.com/pytorch/pytorch/pull/180557,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers,"identify whether block is CI, merge conflict, or review gate",2026-06-22T14:33:15Z
pr,186049,Fix bool argmax meta validation for Inductor,https://github.com/pytorch/pytorch/pull/186049,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2026-06-21T16:27:42Z
pr,187250,Support NumPy 2 __array__ copy kwarg and __array_wrap__ context on Tensor,https://github.com/pytorch/pytorch/pull/187250,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers,"identify whether block is CI, merge conflict, or review gate",2026-06-21T15:57:16Z
pr,176070,[rnn] Prevent LSTMCell segfault on input/hidden shape mismatch,https://github.com/pytorch/pytorch/pull/176070,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers,"identify whether block is CI, merge conflict, or review gate",2026-06-21T03:48:22Z
pr,183525,[MPS] truncate fractional start/end to integer type in linspace when …,https://github.com/pytorch/pytorch/pull/183525,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/mps,"identify whether block is CI, merge conflict, or review gate",2026-06-20T14:23:28Z
pr,180783,"docs: fix index_add_ `index` description (points into self, not source)",https://github.com/pytorch/pytorch/pull/180783,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers,"identify whether block is CI, merge conflict, or review gate",2026-06-19T13:44:46Z
pr,180782,docs: remove unsupported out= from torch.linalg._powsum docstring,https://github.com/pytorch/pytorch/pull/180782,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers,"identify whether block is CI, merge conflict, or review gate",2026-06-19T13:44:45Z
pr,180319,[ROCm] Handle profiler trace format differences in combo kernel tests,https://github.com/pytorch/pytorch/pull/180319,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/inductor-rocm-mi200, ciflow/inductor-rocm-mi300, ciflow/inductor-rocm-mi355","identify whether block is CI, merge conflict, or review gate",2026-06-19T13:44:45Z
pr,180315,Add _Float16 support for Tensor::item() C++ API,https://github.com/pytorch/pytorch/pull/180315,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/mps,"identify whether block is CI, merge conflict, or review gate",2026-06-19T08:36:53Z
pr,187661,Return empty result for FFT with empty batch on MKL backend,https://github.com/pytorch/pytorch/pull/187661,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers,"identify whether block is CI, merge conflict, or review gate",2026-06-18T19:36:04Z
pr,184696,Normalize export slice default bounds from getitem,https://github.com/pytorch/pytorch/pull/184696,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-18T13:50:10Z
pr,185316,[dynamo] Resolve autograd.Function static attrs in getattr,https://github.com/pytorch/pytorch/pull/185316,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-18T01:24:49Z
pr,187405,[dtensor] Propagate `_scaled_mm` tensor meta without eager scale checks,https://github.com/pytorch/pytorch/pull/187405,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/dtensor, ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-17T23:32:38Z
pr,180570,Fix torch.cond crash with FunctionalTensor constants (#180354),https://github.com/pytorch/pytorch/pull/180570,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2026-06-17T21:23:49Z
pr,180810,[inductor][static launcher] validate pointer arg dtypes before launch,https://github.com/pytorch/pytorch/pull/180810,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-17T21:22:49Z
pr,180737,[inductor] Support pin_memory for torch.rand and torch.tensor,https://github.com/pytorch/pytorch/pull/180737,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-17T15:47:07Z
pr,185578,Fix addmm max-autotune SliceView bias guard,https://github.com/pytorch/pytorch/pull/185578,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/b200, ciflow/h100, ciflow/inductor, ciflow/rocm-mi300","identify whether block is CI, merge conflict, or review gate",2026-06-17T14:54:20Z
pr,179235,[Inductor] Add _debug_assert_fused option to foreach_map,https://github.com/pytorch/pytorch/pull/179235,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-17T08:52:35Z
pr,184968,Fix autocast state propagation for normalization layers in functorch batched autograd,https://github.com/pytorch/pytorch/pull/184968,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers,"identify whether block is CI, merge conflict, or review gate",2026-06-16T21:25:19Z
pr,182431,[Inductor] Elide redundant .to(tl.float32) in Triton codegen when value already has target dtype,https://github.com/pytorch/pytorch/pull/182431,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-16T19:05:49Z
pr,178209,Fix CUDA batch_norm eval returning populated save_mean/save_invstd,https://github.com/pytorch/pytorch/pull/178209,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers,"identify whether block is CI, merge conflict, or review gate",2026-06-16T18:53:29Z
pr,178298,torch.linalg.solve_triangular: support for eligible non-dense args,https://github.com/pytorch/pytorch/pull/178298,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-16T12:23:01Z
pr,169937,Fix vectorized float32 to uint8 conversion on AArch64,https://github.com/pytorch/pytorch/pull/169937,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/linux-aarch64, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-15T12:50:01Z
pr,170538,[XPU] Migrate test_testing for xpu,https://github.com/pytorch/pytorch/pull/170538,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/trunk, ciflow/xpu","identify whether block is CI, merge conflict, or review gate",2026-06-15T02:42:17Z
pr,179627,Fix all masked Transformer eager/compile inconsistency,https://github.com/pytorch/pytorch/pull/179627,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers,"identify whether block is CI, merge conflict, or review gate",2026-06-14T16:59:38Z
pr,179584,fix inference_mode getitem compile error,https://github.com/pytorch/pytorch/pull/179584,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers,"identify whether block is CI, merge conflict, or review gate",2026-06-14T16:59:32Z
pr,179404,[DTensor] Replace __module__ hacks with @exposed_in decorators,https://github.com/pytorch/pytorch/pull/179404,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers,"identify whether block is CI, merge conflict, or review gate",2026-06-14T16:59:31Z
pr,184143,Bound Inductor op count for high-order diff,https://github.com/pytorch/pytorch/pull/184143,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-13T14:18:46Z
pr,179658,Add OpaqueGenerator to allow make_fx trace Generator,https://github.com/pytorch/pytorch/pull/179658,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-12T23:00:10Z
pr,187169,Support weights_only and safetensors format in torch.export.save and …,https://github.com/pytorch/pytorch/pull/187169,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers,"identify whether block is CI, merge conflict, or review gate",2026-06-12T17:55:13Z
pr,185122,Validate parameter/buffer storage bounds at export time,https://github.com/pytorch/pytorch/pull/185122,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers,"identify whether block is CI, merge conflict, or review gate",2026-06-12T17:39:33Z
pr,186573,[Inductor] Stabilize checkpointed reduction autotuning,https://github.com/pytorch/pytorch/pull/186573,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-12T17:00:15Z
pr,186619,[pytorch] Fix Inductor missing wait on AsyncCollectiveTensor nested inside DTensor,https://github.com/pytorch/pytorch/pull/186619,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2026-06-11T18:11:55Z
pr,179526,inductor: handle matmuls gracefully in foreach_map,https://github.com/pytorch/pytorch/pull/179526,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-11T16:40:41Z
pr,176295,[inductor] Fix cudagraph output->input feedback crashes,https://github.com/pytorch/pytorch/pull/176295,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-11T15:45:18Z
pr,171617,[FSDP] Preserve container object identity in _recursive_to function,https://github.com/pytorch/pytorch/pull/171617,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-11T13:06:41Z
pr,186061,Fixing torch.pow crash with integral dtypes on CUDA by gating reciprocal/sqrt/rsqrt shortcuts,https://github.com/pytorch/pytorch/pull/186061,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers,"identify whether block is CI, merge conflict, or review gate",2026-06-11T11:41:37Z
pr,180162,Add CPU tensor validation to FBGEMM operators to prevent segfault,https://github.com/pytorch/pytorch/pull/180162,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers,"identify whether block is CI, merge conflict, or review gate",2026-06-11T06:06:57Z
pr,180971,"[inductor] Using is_gpu() instead of hardcoded ""cuda"" check",https://github.com/pytorch/pytorch/pull/180971,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-11T01:24:30Z
pr,176670,[CUDAGraph] Enable `graph_capture_record_stream_reuse` with conditional nodes,https://github.com/pytorch/pytorch/pull/176670,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers,"identify whether block is CI, merge conflict, or review gate",2026-06-10T23:50:28Z
pr,178805,[openreg] Add comprehensive test documentation for OpenReg backend test suite,https://github.com/pytorch/pytorch/pull/178805,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers,"identify whether block is CI, merge conflict, or review gate",2026-06-10T07:55:21Z
pr,184334,[CUDA] Enable bf16 grouped_mm CUTLASS fast path on SM12x and Windows,https://github.com/pytorch/pytorch/pull/184334,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/b200, ciflow/h100, ciflow/rocm-mi300","identify whether block is CI, merge conflict, or review gate",2026-06-10T06:44:23Z
pr,184459,Decompose supported aten._trilinear in Inductor,https://github.com/pytorch/pytorch/pull/184459,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-10T00:11:38Z
pr,184155,Fix Inductor bool vector mask casts,https://github.com/pytorch/pytorch/pull/184155,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-09T23:14:06Z
pr,184114,Add cpp wrapper fallback alignment assertions,https://github.com/pytorch/pytorch/pull/184114,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-09T23:13:58Z
pr,179882,[Devmate Tasks][Butterfly] Fix flaky fbcode//caffe2/test/distributed/tensor:dtensor_ops - test_single_dim_strategy_i0_cpu_float (#179882),https://github.com/pytorch/pytorch/pull/179882,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/dtensor, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-09T05:53:40Z
pr,176234,[autograd] Explicitly error when layer_norm computes 3rd order derivatives,https://github.com/pytorch/pytorch/pull/176234,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-06-08T18:20:19Z
pr,184409,Fix local_map deferred inlining to functionalize mutable operations,https://github.com/pytorch/pytorch/pull/184409,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers,"identify whether block is CI, merge conflict, or review gate",2026-06-05T12:26:59Z
pr,185415,[ROCm] Fix LayerNorm backward kernel for AMD Strix Halo GPUs,https://github.com/pytorch/pytorch/pull/185415,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers,"identify whether block is CI, merge conflict, or review gate",2026-06-05T12:18:38Z
pr,177546,[inductor] Fix pad_mm leaking padded strides to user-visible outputs,https://github.com/pytorch/pytorch/pull/177546,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-04T18:58:21Z
pr,174241,[MPS] Fix conv backward to preserve input memory format,https://github.com/pytorch/pytorch/pull/174241,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/mps, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-04T01:30:31Z
pr,185973,[Accelerator] Enable Default Stream API,https://github.com/pytorch/pytorch/pull/185973,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-03T22:01:12Z
pr,185851,[autograd] Fall back to tensor backward for symbolic pow exponent,https://github.com/pytorch/pytorch/pull/185851,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2026-06-03T18:37:37Z
pr,185834,[inductor] Preserve signed zero in Triton neg codegen,https://github.com/pytorch/pytorch/pull/185834,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-03T03:44:46Z
pr,185846,[decomp] Match eager's device-dependent threshold comparison dtype,https://github.com/pytorch/pytorch/pull/185846,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-03T00:04:20Z
pr,185842,[decomp] Fix multilabel_margin_loss for padded targets,https://github.com/pytorch/pytorch/pull/185842,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2026-06-02T21:59:02Z
pr,185838,[inductor] Propagate subgraph tensor constants to the root graph,https://github.com/pytorch/pytorch/pull/185838,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-02T17:39:02Z
pr,181915,Support returning None from torch.cond() true_fn and false_fn,https://github.com/pytorch/pytorch/pull/181915,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-02T03:46:13Z
pr,185727,"Document bernoulli(input, p) overload in torch docs",https://github.com/pytorch/pytorch/pull/185727,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers,"identify whether block is CI, merge conflict, or review gate",2026-06-01T23:46:24Z
pr,179205,[functorch] Fix vmap/functionalize under inference_mode,https://github.com/pytorch/pytorch/pull/179205,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers,"identify whether block is CI, merge conflict, or review gate",2026-06-01T22:47:09Z
pr,185518,[inductor] Add probability checks for Bernoulli lowering,https://github.com/pytorch/pytorch/pull/185518,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-01T07:24:06Z
pr,185358,[PrivateUse1] Add dedicated LSTM dispatch branch for PU1 backends.,https://github.com/pytorch/pytorch/pull/185358,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers,"identify whether block is CI, merge conflict, or review gate",2026-05-29T21:41:39Z
pr,185579,Improve error message in linalg.eigh for large matrices,https://github.com/pytorch/pytorch/pull/185579,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers,"identify whether block is CI, merge conflict, or review gate",2026-05-29T18:28:37Z
pr,185231,[C++] Ensure wrapped tensors can be printed,https://github.com/pytorch/pytorch/pull/185231,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers,"identify whether block is CI, merge conflict, or review gate",2026-05-29T14:55:50Z
pr,184164,Fix while_loop compile with empty carry and mutation-only state,https://github.com/pytorch/pytorch/pull/184164,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-05-29T08:25:29Z
pr,182068,Make custom op aliasing check warn (not error) in CI,https://github.com/pytorch/pytorch/pull/182068,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-05-29T02:03:32Z
pr,177170,[FSDP] Fix alias annotation for fsdp::all_gather_copy_in,https://github.com/pytorch/pytorch/pull/177170,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-05-29T01:57:56Z
pr,182414,fix(decomp): cast padding to int in replication/reflection pad decomposition,https://github.com/pytorch/pytorch/pull/182414,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-05-28T17:01:06Z
pr,178106,Support Optional[list[torch.Tensor]] in infer_schema,https://github.com/pytorch/pytorch/pull/178106,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers,"identify whether block is CI, merge conflict, or review gate",2026-05-27T12:58:49Z
pr,185227,Error out when getting symmetric memory device pointers for NCCL < 2.28,https://github.com/pytorch/pytorch/pull/185227,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/h100-symm-mem,"identify whether block is CI, merge conflict, or review gate",2026-05-27T08:38:20Z
pr,184420,"Fix pow(-0.0, +/-0.5) sign on sqrt/rsqrt shortcut",https://github.com/pytorch/pytorch/pull/184420,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/mps, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-05-26T23:49:52Z
pr,184822,Add Sphinx redirects for renamed autosummary Stream/stream pages,https://github.com/pytorch/pytorch/pull/184822,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers,"identify whether block is CI, merge conflict, or review gate",2026-05-26T16:10:55Z
pr,184631,[MPS] Skip strided NDArray path in _mps_linear_nograph when inputs are contiguous,https://github.com/pytorch/pytorch/pull/184631,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/mps,"identify whether block is CI, merge conflict, or review gate",2026-05-26T13:45:10Z
pr,181823,Docs: update torch.multinomial num_samples requirement note for bette…,https://github.com/pytorch/pytorch/pull/181823,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers,"identify whether block is CI, merge conflict, or review gate",2026-05-25T11:25:22Z
pr,183857,Support float8 dtypes in index_copy on CPU and CUDA,https://github.com/pytorch/pytorch/pull/183857,ready for maintainer decision,72,approved with no detected failing/pending checks; PR closes or references open issue numbers; PR has not updated for at least 45 days,final maintainer merge/release decision,2026-05-18T16:41:02Z
pr,171101,workaround for transpose_mxn test failure on AArch64,https://github.com/pytorch/pytorch/pull/171101,ready for maintainer decision,72,"approved with no detected failing/pending checks; PR closes or references open issue numbers; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/trunk",final maintainer merge/release decision,2026-05-11T20:08:35Z
pr,141912,Add stdlib includes to c10d debug handlers,https://github.com/pytorch/pytorch/pull/141912,ready for maintainer decision,72,approved with no detected failing/pending checks; PR closes or references open issue numbers; PR has not updated for at least 45 days,final maintainer merge/release decision,2025-03-10T19:21:46Z
pr,121445,[NJT] Actually inline NT torch function during dynamo,https://github.com/pytorch/pytorch/pull/121445,ready for maintainer decision,72,"approved with no detected failing/pending checks; PR closes or references open issue numbers; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/trunk",final maintainer merge/release decision,2024-08-06T22:35:39Z
issue,170672,"torch.export.export Fails with RuntimeError and ""Please file an issue on github"" for In-Place Parameter Updates in forward",https://github.com/pytorch/pytorch/issues/170672,needs triage,71,"PyTorch issue lacks triaged label; open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: oncall: export, oncall: pt2",review linked PR state before touching issue,2026-01-04T04:13:24Z
issue,170539,"test_export_opinfo: 9 tests fail with ""Unexpected success"" due to outdated xfail markers",https://github.com/pytorch/pytorch/issues/170539,needs triage,71,"PyTorch issue lacks triaged label; open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: ci, oncall: export, oncall: pt2",review linked PR state before touching issue,2025-12-17T00:26:40Z
issue,163713,[export] 8*s72 is not tracked with proxy for torch.fx.experimental.proxy_tensor._ModuleStackTracer object,https://github.com/pytorch/pytorch/issues/163713,needs triage,71,"PyTorch issue lacks triaged label; open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: oncall: export, oncall: pt2",review linked PR state before touching issue,2025-12-15T10:45:37Z
issue,163323,[export] Bug Report: module.to(device) on exported programs causes assertion failures (device-agnostic export broken since 2.8),https://github.com/pytorch/pytorch/issues/163323,needs triage,71,"PyTorch issue lacks triaged label; open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: oncall: export, oncall: pt2",review linked PR state before touching issue,2025-12-12T01:09:12Z
issue,169991,aot_export_joint_with_descriptors don't work with compile inside due to dynamo config,https://github.com/pytorch/pytorch/issues/169991,needs triage,71,"PyTorch issue lacks triaged label; open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: oncall: export, oncall: pt2",review linked PR state before touching issue,2025-12-09T23:28:48Z
issue,169210,Segmentation Fault when Converting QInt32Storage to bfloat16,https://github.com/pytorch/pytorch/issues/169210,needs triage,71,"PyTorch issue lacks triaged label; open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: error checking, oncall: quantization",review linked PR state before touching issue,2025-12-05T00:42:42Z
issue,164313,Torch export with dynamic shapes does not work with Hugging Face models that has TransformersKwargs in forward signature,https://github.com/pytorch/pytorch/issues/164313,needs triage,71,"PyTorch issue lacks triaged label; open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: oncall: export, oncall: pt2",review linked PR state before touching issue,2025-11-11T22:17:58Z
issue,167007,"torch.export failed with tensor subclass, worked in 2.9.0",https://github.com/pytorch/pytorch/issues/167007,needs triage,71,"PyTorch issue lacks triaged label; open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: oncall: export, oncall: pt2",review linked PR state before touching issue,2025-11-11T07:44:20Z
issue,166121,[profile] profile_memory doesn't work if _ExperimentalConfig(profile_all_threads = True) is set,https://github.com/pytorch/pytorch/issues/166121,needs triage,71,PyTorch issue lacks triaged label; open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: oncall: profiler,review linked PR state before touching issue,2025-11-10T22:01:15Z
issue,146719,Operations that fail under `torch.export.export`(`torch.autograd.grad`) -> `torch.compile`,https://github.com/pytorch/pytorch/issues/146719,needs triage,71,"PyTorch issue lacks triaged label; open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: oncall: export, oncall: pt2",review linked PR state before touching issue,2025-11-07T16:22:01Z
issue,149497,symbolic_trace failed on deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B,https://github.com/pytorch/pytorch/issues/149497,needs triage,71,"PyTorch issue lacks triaged label; open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: fx, oncall: fx",review linked PR state before touching issue,2025-10-31T03:37:04Z
issue,165641,strict export doesn't work with global constants,https://github.com/pytorch/pytorch/issues/165641,needs triage,71,"PyTorch issue lacks triaged label; open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: oncall: export, oncall: pt2",review linked PR state before touching issue,2025-10-21T21:37:23Z
issue,165444,AOTInductor not updating buffers inplace,https://github.com/pytorch/pytorch/issues/165444,needs triage,71,"PyTorch issue lacks triaged label; open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: aotinductor, oncall: export, oncall: pt2",review linked PR state before touching issue,2025-10-19T23:42:43Z
issue,154559,[cond] support for unbacked symbols in compiled region,https://github.com/pytorch/pytorch/issues/154559,needs triage,71,"PyTorch issue lacks triaged label; open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: oncall: export, oncall: pt2",review linked PR state before touching issue,2025-10-18T05:59:01Z
issue,165032,"torch.export fails when you slice an input at a dynamic dimension and Dim(“t”, min=, max=)",https://github.com/pytorch/pytorch/issues/165032,needs triage,71,"PyTorch issue lacks triaged label; open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamic shapes, oncall: export, oncall: pt2",review linked PR state before touching issue,2025-10-13T23:34:32Z
issue,165073,torch.export lift symint input to a graph during dynamo,https://github.com/pytorch/pytorch/issues/165073,needs triage,71,"PyTorch issue lacks triaged label; open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: oncall: export, oncall: pt2",review linked PR state before touching issue,2025-10-10T17:00:38Z
issue,153247,Cannot use torch.arange in torch.ops.higher_order.scan,https://github.com/pytorch/pytorch/issues/153247,needs triage,71,"PyTorch issue lacks triaged label; open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: oncall: export, oncall: pt2",review linked PR state before touching issue,2025-10-02T05:53:16Z
issue,159468,Dynamo ONNX Export of resize operation doesn't work as expected,https://github.com/pytorch/pytorch/issues/159468,needs triage,71,"PyTorch issue lacks triaged label; open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: onnx, oncall: export, oncall: pt2",review linked PR state before touching issue,2025-10-01T07:16:38Z
issue,159918,[export] Tensor subclass decomposition breaks state dict contracts,https://github.com/pytorch/pytorch/issues/159918,needs triage,71,"PyTorch issue lacks triaged label; open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: oncall: export, oncall: pt2",review linked PR state before touching issue,2025-09-24T15:20:27Z
issue,163303,[export] non-strict export creates extra tensor constants when forward runs with a container of params,https://github.com/pytorch/pytorch/issues/163303,needs triage,71,"PyTorch issue lacks triaged label; open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: oncall: export, oncall: pt2",review linked PR state before touching issue,2025-09-23T16:52:11Z
issue,163312,[symbolic trace] fuction with decorator,https://github.com/pytorch/pytorch/issues/163312,needs triage,71,"PyTorch issue lacks triaged label; open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: fx, oncall: fx",review linked PR state before touching issue,2025-09-20T00:51:03Z
issue,162734,Run decomposition fails for prims.broadcast_in_dim,https://github.com/pytorch/pytorch/issues/162734,needs triage,71,"PyTorch issue lacks triaged label; open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: oncall: export, oncall: pt2",review linked PR state before touching issue,2025-09-16T00:08:19Z
issue,162386,Cannot trace out joint from non-strict export for model with activation checkpointing (AC),https://github.com/pytorch/pytorch/issues/162386,needs triage,71,"PyTorch issue lacks triaged label; open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: aotdispatch, module: pt2-dispatcher, oncall: export, oncall: pt2",review linked PR state before touching issue,2025-09-16T00:07:20Z
issue,140607,Bad error message for malformed args during export,https://github.com/pytorch/pytorch/issues/140607,needs triage,71,"PyTorch issue lacks triaged label; open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: compile ux, module: dynamo, oncall: export, oncall: pt2",review linked PR state before touching issue,2025-09-12T21:16:08Z
issue,161807,torch.fx.experimental.optimization.optimize_for_inference changes OrderedDict outputs to immutable_dict in torch.export CallSpec,https://github.com/pytorch/pytorch/issues/161807,needs triage,71,"PyTorch issue lacks triaged label; open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: oncall: export, oncall: pt2",review linked PR state before touching issue,2025-09-08T18:50:49Z
issue,161671,Saving a modified ExportedProgram,https://github.com/pytorch/pytorch/issues/161671,needs triage,71,"PyTorch issue lacks triaged label; open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: oncall: export, oncall: pt2",review linked PR state before touching issue,2025-08-29T18:44:31Z
issue,159346,[AOTI] Severe Performance Regression with FP16 Autocast in AOTInductor for Small Batch Sizes,https://github.com/pytorch/pytorch/issues/159346,needs triage,71,"PyTorch issue lacks triaged label; open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: aotinductor, oncall: export, oncall: pt2",review linked PR state before touching issue,2025-08-21T23:25:42Z
issue,156786,[inductor] [cpu] `torch.nn.Embedding-torch.index_copy` outputs inconsistent results on cpu inductor,https://github.com/pytorch/pytorch/issues/156786,needs triage,71,"PyTorch issue lacks triaged label; open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: oncall: cpu inductor, oncall: pt2",review linked PR state before touching issue,2025-08-12T02:56:15Z
issue,158540,can't torch.export.export tinyllama model,https://github.com/pytorch/pytorch/issues/158540,needs triage,71,"PyTorch issue lacks triaged label; open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: oncall: export, oncall: pt2",review linked PR state before touching issue,2025-07-26T00:27:15Z
issue,156511,functorch_maml_omniglot is a bad CPU performance smoketest model,https://github.com/pytorch/pytorch/issues/156511,needs triage,71,"PyTorch issue lacks triaged label; open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: infra, oncall: cpu inductor, oncall: pt2",review linked PR state before touching issue,2025-07-07T22:25:25Z
issue,147380,[export] fail to export joint graph of a model with tied weights using experimental `_export_forward_backward` API,https://github.com/pytorch/pytorch/issues/147380,needs triage,71,"PyTorch issue lacks triaged label; open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: oncall: export, oncall: pt2",review linked PR state before touching issue,2025-06-06T22:59:38Z
issue,154647,"Recreation of unbacked symint leads to ""possible memo disaster"" when running decompositions",https://github.com/pytorch/pytorch/issues/154647,needs triage,71,"PyTorch issue lacks triaged label; open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: oncall: export, oncall: pt2",review linked PR state before touching issue,2025-06-05T00:27:42Z
issue,154137,[Dynamo][Inductor] `detectron2_fcos_r_50_fpn` in export config failure on dashboard,https://github.com/pytorch/pytorch/issues/154137,needs triage,71,"PyTorch issue lacks triaged label; open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, module: guards, oncall: export, oncall: pt2",review linked PR state before touching issue,2025-05-27T18:21:07Z
issue,153387,Pre-dispatch export doesn't work with non-param/buffer tensor subclasses,https://github.com/pytorch/pytorch/issues/153387,needs triage,71,"PyTorch issue lacks triaged label; open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: oncall: export, oncall: pt2",review linked PR state before touching issue,2025-05-20T15:45:46Z
issue,153056,Export doesn't move embedding to correct device,https://github.com/pytorch/pytorch/issues/153056,needs triage,71,"PyTorch issue lacks triaged label; open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: oncall: export, oncall: pt2",review linked PR state before touching issue,2025-05-07T19:09:01Z
issue,150371,Dynamic shapes doesn't work with kwargs,https://github.com/pytorch/pytorch/issues/150371,needs triage,71,"PyTorch issue lacks triaged label; open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: oncall: export, oncall: pt2",review linked PR state before touching issue,2025-04-02T20:09:53Z
issue,145687,[Inductor-CPU] `add` is not being fused with templated int8 WoQ GEMM while running LLaMA2,https://github.com/pytorch/pytorch/issues/145687,needs triage,71,"PyTorch issue lacks triaged label; open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: oncall: cpu inductor, oncall: pt2",review linked PR state before touching issue,2025-02-04T22:33:10Z
issue,145785,export serde turn hop's tuple arg into list,https://github.com/pytorch/pytorch/issues/145785,needs triage,71,"PyTorch issue lacks triaged label; open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: oncall: export, oncall: pt2",review linked PR state before touching issue,2025-01-31T23:44:47Z
issue,145574,[compile / strict export] torch._dynamo.exc.Unsupported: CollectiveFunctionRewriteVariable can't support async_op=True for <function all_reduce at 0x7f40be5724d0>,https://github.com/pytorch/pytorch/issues/145574,needs triage,71,"PyTorch issue lacks triaged label; open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: oncall: export, oncall: pt2",review linked PR state before touching issue,2025-01-24T17:53:07Z
issue,144247,[inductor][cpu] torch.bitwise_and/or/xor incorrectly accepts float32 tensors,https://github.com/pytorch/pytorch/issues/144247,needs triage,71,"PyTorch issue lacks triaged label; open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: error checking, oncall: cpu inductor, oncall: pt2",review linked PR state before touching issue,2025-01-06T18:04:38Z
issue,140683,[inductor][cpu]maml fp32/amp_fp16 performance regression in 2024-11-11 nightly release,https://github.com/pytorch/pytorch/issues/140683,needs triage,71,"PyTorch issue lacks triaged label; open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: oncall: cpu inductor, oncall: pt2",review linked PR state before touching issue,2024-12-17T01:19:15Z
issue,141640,Relaxing Export IR node.meta requirements,https://github.com/pytorch/pytorch/issues/141640,needs triage,71,"PyTorch issue lacks triaged label; open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: oncall: export, oncall: pt2",review linked PR state before touching issue,2024-12-04T18:17:29Z
issue,141115,[AOTInductor] AOT compiled models do not consistently check input shapes,https://github.com/pytorch/pytorch/issues/141115,needs triage,71,"PyTorch issue lacks triaged label; open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: oncall: export, oncall: pt2",review linked PR state before touching issue,2024-11-20T19:06:14Z
issue,140615,Non-actionable error message when output type is not pytree-able for aoti artifact,https://github.com/pytorch/pytorch/issues/140615,needs triage,71,"PyTorch issue lacks triaged label; open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: oncall: export, oncall: pt2",review linked PR state before touching issue,2024-11-18T01:45:45Z
issue,140842,torch.cond with unbacked symint failed to compile in AOTI,https://github.com/pytorch/pytorch/issues/140842,needs triage,71,"PyTorch issue lacks triaged label; open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: aotinductor, oncall: export, oncall: pt2",review linked PR state before touching issue,2024-11-16T00:02:36Z
pr,187631,[dynamo] Fix segfault in list comprehension graph break,https://github.com/pytorch/pytorch/pull/187631,ready for maintainer decision,70,"approved with no detected failing/pending checks; CI lane labels: ciflow/inductor, ciflow/torchtitan",final maintainer merge/release decision,2026-07-09T01:04:06Z
pr,188175,[dynamo] fix autograd cache and split_module tests under canonicalization,https://github.com/pytorch/pytorch/pull/188175,ready for maintainer decision,70,approved with no detected failing/pending checks; CI lane labels: ciflow/inductor,final maintainer merge/release decision,2026-07-08T23:22:38Z
pr,189292,Fix dead ExecuTorch XNNPACK doc link in mobile_optimizer,https://github.com/pytorch/pytorch/pull/189292,ready for maintainer decision,70,approved with no detected failing/pending checks; CI lane labels: ciflow/trunk,final maintainer merge/release decision,2026-07-08T23:22:32Z
pr,186809,[cudaMallocAsync] Count graph-mem pool in memory_reserved(),https://github.com/pytorch/pytorch/pull/186809,ready for maintainer decision,70,approved with no detected failing/pending checks,final maintainer merge/release decision,2026-07-08T22:47:34Z
pr,189256,[BE][MPS] Factor host-buffer wrapping into buffer_with_offset_from_tensor,https://github.com/pytorch/pytorch/pull/189256,ready for maintainer decision,70,approved with no detected failing/pending checks; CI lane labels: ciflow/mps,final maintainer merge/release decision,2026-07-08T21:46:11Z
pr,188901,Implement Double Backward for `cdist` and `pdist`,https://github.com/pytorch/pytorch/pull/188901,ready for maintainer decision,70,approved with no detected failing/pending checks; CI lane labels: ciflow/inductor,final maintainer merge/release decision,2026-07-08T21:37:19Z
pr,189187,Retune against triton 3.8,https://github.com/pytorch/pytorch/pull/189187,ready for maintainer decision,70,"approved with no detected failing/pending checks; CI lane labels: ciflow/inductor, ciflow/torchtitan",final maintainer merge/release decision,2026-07-08T21:22:11Z
pr,189021,[dynamo] Support empty tuple-subclass construction (MyTuple()),https://github.com/pytorch/pytorch/pull/189021,ready for maintainer decision,70,"approved with no detected failing/pending checks; CI lane labels: ciflow/inductor, ciflow/torchtitan",final maintainer merge/release decision,2026-07-08T21:21:53Z
pr,187887,[Inductor] Add decomposition for prims.uniform,https://github.com/pytorch/pytorch/pull/187887,ready for maintainer decision,70,"approved with no detected failing/pending checks; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk",final maintainer merge/release decision,2026-07-08T21:12:15Z
pr,188825,"[Dynamo] Implement subgenerator support in "".throw"" / "".close"" methods",https://github.com/pytorch/pytorch/pull/188825,ready for maintainer decision,70,"approved with no detected failing/pending checks; CI lane labels: ciflow/inductor, ciflow/torchtitan",final maintainer merge/release decision,2026-07-08T21:06:07Z
pr,188834,[Dynamo] Track generator attribute mutations for proper closure handling,https://github.com/pytorch/pytorch/pull/188834,ready for maintainer decision,70,"approved with no detected failing/pending checks; CI lane labels: ciflow/inductor, ciflow/torchtitan",final maintainer merge/release decision,2026-07-08T21:06:06Z
pr,188639,[Dynamo] Corretly implement SEND opcode,https://github.com/pytorch/pytorch/pull/188639,ready for maintainer decision,70,"approved with no detected failing/pending checks; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk",final maintainer merge/release decision,2026-07-08T21:06:04Z
pr,188638,[Dynamo] Fix CLEANUP_THROW,https://github.com/pytorch/pytorch/pull/188638,ready for maintainer decision,70,"approved with no detected failing/pending checks; CI lane labels: ciflow/inductor, ciflow/torchtitan",final maintainer merge/release decision,2026-07-08T21:06:03Z
pr,185309,Add sparse COO dispatch for linalg.vector_norm,https://github.com/pytorch/pytorch/pull/185309,ready for maintainer decision,70,approved with no detected failing/pending checks; CI lane labels: ciflow/trunk,final maintainer merge/release decision,2026-07-08T20:52:04Z
pr,182906,Port 11 tests under test/distributed folder to XPU,https://github.com/pytorch/pytorch/pull/182906,ready for maintainer decision,70,approved with no detected failing/pending checks; CI lane labels: ciflow/torchtitan,final maintainer merge/release decision,2026-07-08T20:48:29Z
pr,189201,[TEST][Inductor] Fix some `test_buffer_mutation*` tests,https://github.com/pytorch/pytorch/pull/189201,ready for maintainer decision,70,approved with no detected failing/pending checks; CI lane labels: ciflow/inductor,final maintainer merge/release decision,2026-07-08T20:40:29Z
pr,189172,Fix protobuf resource-compiler build with clang-cl (RC1106 /bigobj),https://github.com/pytorch/pytorch/pull/189172,ready for maintainer decision,70,approved with no detected failing/pending checks,final maintainer merge/release decision,2026-07-08T20:20:05Z
pr,186854,Simplify endianness check using std::endian,https://github.com/pytorch/pytorch/pull/186854,ready for maintainer decision,70,"approved with no detected failing/pending checks; CI lane labels: ciflow/trunk, ciflow/vllm",final maintainer merge/release decision,2026-07-08T19:23:37Z
pr,184261,[Test] Make test_serialization.py device-agnostic for out-of-tree backends,https://github.com/pytorch/pytorch/pull/184261,ready for maintainer decision,70,approved with no detected failing/pending checks; CI lane labels: ciflow/trunk,final maintainer merge/release decision,2026-07-08T19:12:10Z
issue,189287,"[vllm] vllm-x-pytorch-test multimodal pooling tests fail with EngineCore ""CUDA driver initialization failed"" since Triton 3.8 bump (#188251)",https://github.com/pytorch/pytorch/issues/189287,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: ci, module: cuda, module: regression, module: tests",review early; this can change release or regression risk,2026-07-08T18:15:41Z
issue,189283,"[CPU] operator_benchmark: isolated ~2x regressions (index_add, baddbmm) since ~Nov 2025",https://github.com/pytorch/pytorch/issues/189283,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: cpu, module: performance, module: regression",review early; this can change release or regression risk,2026-07-08T17:34:26Z
pr,187509,test: add backend-agnostic c10d collective tests,https://github.com/pytorch/pytorch/pull/187509,ready for maintainer decision,70,approved with no detected failing/pending checks,final maintainer merge/release decision,2026-07-08T17:00:29Z
pr,188472,[inductor] Bucket all_reduce in the manual overlap scheduler,https://github.com/pytorch/pytorch/pull/188472,ready for maintainer decision,70,"approved with no detected failing/pending checks; CI lane labels: ciflow/inductor, ciflow/torchtitan",final maintainer merge/release decision,2026-07-08T16:51:13Z
pr,187236,Make Overlap Runtime Estimation Device Aware,https://github.com/pytorch/pytorch/pull/187236,ready for maintainer decision,70,"approved with no detected failing/pending checks; CI lane labels: ciflow/inductor, ciflow/torchtitan",final maintainer merge/release decision,2026-07-08T15:25:16Z
pr,188330,Dispatch `index_put_` to `masked_fill_.Scalar` when possible,https://github.com/pytorch/pytorch/pull/188330,ready for maintainer decision,70,approved with no detected failing/pending checks; CI lane labels: ciflow/trunk,final maintainer merge/release decision,2026-07-08T12:14:25Z
pr,188480,[Dynamo] Add UNSPECIFIED_BACKEND lint rule requiring explicit backend=,https://github.com/pytorch/pytorch/pull/188480,ready for maintainer decision,70,"approved with no detected failing/pending checks; CI lane labels: ciflow/docker, ciflow/inductor, ciflow/trunk",final maintainer merge/release decision,2026-07-08T05:57:36Z
issue,185405,"[Inductor] AssertionError: len(index) == len(stride) in ir.py when using torch.compile(mode=""max-autotune"") on CPU BMM/GEMM",https://github.com/pytorch/pytorch/issues/185405,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: complex, module: cpu, module: crash, module: inductor",review early; this can change release or regression risk,2026-07-08T05:55:34Z
pr,189181,[CI] Advisor: split garbage verdict into infra_issue + garbage (minimal),https://github.com/pytorch/pytorch/pull/189181,ready for maintainer decision,70,approved with no detected failing/pending checks,final maintainer merge/release decision,2026-07-08T03:15:15Z
pr,189017,Respect the blocking parameter in torch.Event,https://github.com/pytorch/pytorch/pull/189017,ready for maintainer decision,70,"approved with no detected failing/pending checks; CI lane labels: ciflow/mps, ciflow/trunk",final maintainer merge/release decision,2026-07-08T01:25:26Z
pr,189004,Add priority&priority_range to torch.Stream,https://github.com/pytorch/pytorch/pull/189004,ready for maintainer decision,70,approved with no detected failing/pending checks; CI lane labels: ciflow/trunk,final maintainer merge/release decision,2026-07-08T01:09:05Z
pr,189033,Auto-triage open source PRs when a maintainer engages,https://github.com/pytorch/pytorch/pull/189033,ready for maintainer decision,70,approved with no detected failing/pending checks,final maintainer merge/release decision,2026-07-07T22:10:04Z
pr,188396,[Test] Improve test reuse in test/test_tensor_iterator.py for out-of-tree backends,https://github.com/pytorch/pytorch/pull/188396,ready for maintainer decision,70,approved with no detected failing/pending checks,final maintainer merge/release decision,2026-07-07T18:13:06Z
pr,185207,[Test] Make dynamo/test_aot_autograd_cache.py device-agnostic for out-of-tree backends,https://github.com/pytorch/pytorch/pull/185207,ready for maintainer decision,70,approved with no detected failing/pending checks; CI lane labels: ciflow/inductor,final maintainer merge/release decision,2026-07-07T16:34:21Z
pr,176404,Add prioritized text linker optimization for lld.,https://github.com/pytorch/pytorch/pull/176404,ready for maintainer decision,70,"approved with no detected failing/pending checks; CI lane labels: ciflow/linux-aarch64, ciflow/trunk",final maintainer merge/release decision,2026-07-07T14:18:47Z
pr,184593,[Test] Make test_autograd.py device-agnostic for out-of-tree backends,https://github.com/pytorch/pytorch/pull/184593,ready for maintainer decision,70,approved with no detected failing/pending checks,final maintainer merge/release decision,2026-07-07T13:56:39Z
pr,185586,Update README of OpenReg module to consistent with its structure and principle,https://github.com/pytorch/pytorch/pull/185586,ready for maintainer decision,70,approved with no detected failing/pending checks,final maintainer merge/release decision,2026-07-07T09:43:23Z
issue,188332,Inductor: allgather fusion creates rank-specific cache that crashes when loaded by other ranks,https://github.com/pytorch/pytorch/issues/188332,high-priority/blocker,70,high-priority/blocker/regression label present; component/oncall labels: oncall: pt2,review early; this can change release or regression risk,2026-07-07T06:55:08Z
issue,185839,"torch.linalg.qr (float64) repeated in a loop crashes the process (SIGSEGV, exit 139) on ROCm 7.x",https://github.com/pytorch/pytorch/issues/185839,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: crash, module: cudacachingallocator, module: linear algebra, module: regression",review early; this can change release or regression risk,2026-07-07T03:32:56Z
pr,188666,Enable xpu for TestRefs and TestDecomp.,https://github.com/pytorch/pytorch/pull/188666,ready for maintainer decision,70,"approved with no detected failing/pending checks; CI lane labels: ciflow/trunk, ciflow/xpu",final maintainer merge/release decision,2026-07-07T02:30:51Z
pr,182784,"[xpu][test] Port distributed _tools,algorithms,launcher tests cases on Intel GPUs",https://github.com/pytorch/pytorch/pull/182784,ready for maintainer decision,70,approved with no detected failing/pending checks; CI lane labels: ciflow/h100-distributed,final maintainer merge/release decision,2026-07-07T01:50:18Z
pr,189016,[Test] Improve test reuse in test_torch.py for out-of-tree backends,https://github.com/pytorch/pytorch/pull/189016,ready for maintainer decision,70,approved with no detected failing/pending checks,final maintainer merge/release decision,2026-07-07T01:49:04Z
issue,189018,`_scaled_mm_v2` meta implementation checks for device-specific constraints,https://github.com/pytorch/pytorch/issues/189018,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: dispatch, module: meta tensors, module: regression",review early; this can change release or regression risk,2026-07-06T22:46:50Z
pr,186639,[CUDA graphs] Add resolve_and_remap convenience for kernel annotations,https://github.com/pytorch/pytorch/pull/186639,ready for maintainer decision,70,approved with no detected failing/pending checks; CI lane labels: ciflow/trunk,final maintainer merge/release decision,2026-07-06T20:11:11Z
pr,188832,[pytorch][cuda] make_graphed_callables: expose capture_error_mode (#188832),https://github.com/pytorch/pytorch/pull/188832,ready for maintainer decision,70,approved with no detected failing/pending checks; CI lane labels: ciflow/trunk,final maintainer merge/release decision,2026-07-06T20:11:00Z
pr,182062,[CUDA] Early-exit in `release_cached_blocks` if `captures_underway` isn't empty,https://github.com/pytorch/pytorch/pull/182062,ready for maintainer decision,70,approved with no detected failing/pending checks,final maintainer merge/release decision,2026-07-06T19:08:20Z
issue,188733,[vllm] [2.12.1 regression][Spec Decode] CUDA WARP_ILLEGAL_ADDRESS in FlashAttnFwdSm90 during CUDA graph capture (EAGLE),https://github.com/pytorch/pytorch/issues/188733,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: crash, module: cuda, module: cuda graphs, module: regression",review early; this can change release or regression risk,2026-07-06T18:05:06Z
issue,188835,DistillGPT2 fails accuracy on A100 in inductor smoke test,https://github.com/pytorch/pytorch/issues/188835,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: inductor, module: numerical-stability, module: regression, oncall: pt2",review early; this can change release or regression risk,2026-07-06T17:32:29Z
issue,188958,[CUDA] illegal memory read on batch_norm_backward_reduce_channels_last_kernel,https://github.com/pytorch/pytorch/issues/188958,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: crash, module: cuda, module: edge cases, module: norms and normalization",review early; this can change release or regression risk,2026-07-06T17:31:39Z
issue,188959,[CUDA] illegal memory read on col2im_batched_kernel,https://github.com/pytorch/pytorch/issues/188959,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: convolution, module: crash, module: cuda, module: edge cases",review early; this can change release or regression risk,2026-07-06T17:31:22Z
issue,188841,triton 3.8 pin update regressing compile + float8 training numerics,https://github.com/pytorch/pytorch/issues/188841,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: floatx (formerly float8), module: inductor, module: numerical-stability, module: regression",review early; this can change release or regression risk,2026-07-06T17:23:33Z
issue,188557,operator_microbenchmark ROCm gfx950 job never completes -- runs 8h+ and is cancelled at the workflow timeout,https://github.com/pytorch/pytorch/issues/188557,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: benchmark, module: ci, module: regression, module: rocm",review early; this can change release or regression risk,2026-07-06T17:19:02Z
pr,188656,[Test] Improve test reuse in test/test_unary_ufuncs.py for out-of-tree backends,https://github.com/pytorch/pytorch/pull/188656,ready for maintainer decision,70,approved with no detected failing/pending checks,final maintainer merge/release decision,2026-07-06T15:58:54Z
pr,187926,[Test] Refactor test/test_shape_ops.py,https://github.com/pytorch/pytorch/pull/187926,ready for maintainer decision,70,approved with no detected failing/pending checks,final maintainer merge/release decision,2026-07-06T13:43:44Z
pr,186351,[Test] Refactor test/test_scatter_gather_ops.py,https://github.com/pytorch/pytorch/pull/186351,ready for maintainer decision,70,approved with no detected failing/pending checks,final maintainer merge/release decision,2026-07-06T13:43:43Z
pr,186352,[Test] Refactor test/test_ops.py,https://github.com/pytorch/pytorch/pull/186352,ready for maintainer decision,70,approved with no detected failing/pending checks,final maintainer merge/release decision,2026-07-06T13:43:37Z
pr,183278,[Inductor][Refactor] 4/N Move reduction heuristics from triton_heuristics.py into triton_codegen/,https://github.com/pytorch/pytorch/pull/183278,ready for maintainer decision,70,"approved with no detected failing/pending checks; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk",final maintainer merge/release decision,2026-07-06T10:55:42Z
issue,188674,DISABLED test_workspace_runtime_integration (__main__.TestNVUniversalGemmEpilogueFusion),https://github.com/pytorch/pytorch/issues/188674,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: inductor, module: tests, oncall: pt2",review early; this can change release or regression risk,2026-07-06T10:19:40Z
issue,188955,rms_norm autocast fp32 policy (2.12) materializes standalone cast kernels under torch.compile when dtypes already match. Peak-memory and throughput regression for bf16 models,https://github.com/pytorch/pytorch/issues/188955,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: amp (automated mixed precision), module: inductor, module: memory usage, module: performance",review early; this can change release or regression risk,2026-07-06T10:13:27Z
issue,188912,vit_base_patch16_siglip_256 amp training accuracy regression on H100 (2.13.0),https://github.com/pytorch/pytorch/issues/188912,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: amp (automated mixed precision), module: correctness (silent), module: cudnn, module: inductor",review early; this can change release or regression risk,2026-07-06T10:12:49Z
pr,188044,[Test] Refactor test/test_view_ops.py,https://github.com/pytorch/pytorch/pull/188044,ready for maintainer decision,70,approved with no detected failing/pending checks,final maintainer merge/release decision,2026-07-06T09:24:25Z
pr,185798,[Test] Refactor test/test_indexing.py,https://github.com/pytorch/pytorch/pull/185798,ready for maintainer decision,70,approved with no detected failing/pending checks; CI lane labels: ciflow/trunk,final maintainer merge/release decision,2026-07-06T08:27:13Z
pr,185802,[Test] Refactor test/test_out_dtype_op.py,https://github.com/pytorch/pytorch/pull/185802,ready for maintainer decision,70,approved with no detected failing/pending checks,final maintainer merge/release decision,2026-07-06T08:26:56Z
pr,183272,[DTensor] Switch test_math_ops and test_matrix_ops to DTensorOpTestBase,https://github.com/pytorch/pytorch/pull/183272,ready for maintainer decision,70,approved with no detected failing/pending checks; CI lane labels: ciflow/dtensor,final maintainer merge/release decision,2026-07-05T17:51:54Z
pr,184947,[dynamo] fix pyright typing on torch.compiler.disable,https://github.com/pytorch/pytorch/pull/184947,ready for maintainer decision,70,"approved with no detected failing/pending checks; CI lane labels: ciflow/inductor, ciflow/torchtitan",final maintainer merge/release decision,2026-07-05T00:19:46Z
pr,182014,[inductor] Fix spmd_check deadlock in recursive post_grad_passes,https://github.com/pytorch/pytorch/pull/182014,ready for maintainer decision,70,"approved with no detected failing/pending checks; CI lane labels: ciflow/inductor, ciflow/torchtitan",final maintainer merge/release decision,2026-07-04T18:50:04Z
pr,178051,[pytree][dynamo] add generic_pytree (torch.utils.pytree) to test_tree_map.py parametrization,https://github.com/pytorch/pytorch/pull/178051,ready for maintainer decision,70,approved with no detected failing/pending checks; CI lane labels: ciflow/inductor,final maintainer merge/release decision,2026-07-04T12:10:38Z
issue,188864,[MPS] torch.fft.hfft aborts on 5D complex64 input with dim=0,https://github.com/pytorch/pytorch/issues/188864,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: complex, module: crash, module: error checking, module: fft",review early; this can change release or regression risk,2026-07-03T03:42:04Z
pr,188229,Fix 64-bit indexing in avg_pool3d backward atomic kernel,https://github.com/pytorch/pytorch/pull/188229,ready for maintainer decision,70,approved with no detected failing/pending checks,final maintainer merge/release decision,2026-07-02T19:08:59Z
pr,184795,Autogenerated operator!= in C++20,https://github.com/pytorch/pytorch/pull/184795,ready for maintainer decision,70,"approved with no detected failing/pending checks; CI lane labels: ciflow/trunk, ciflow/vllm",final maintainer merge/release decision,2026-07-02T17:23:39Z
pr,188785,Upgrade submodule oneDNN to v3.12.2,https://github.com/pytorch/pytorch/pull/188785,ready for maintainer decision,70,approved with no detected failing/pending checks,final maintainer merge/release decision,2026-07-02T16:15:03Z
issue,188691,DISABLED test_stream_invalid_device_index (__main__.TestCuda),https://github.com/pytorch/pytorch/issues/188691,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: cuda, module: regression, module: tests",review early; this can change release or regression risk,2026-07-01T16:15:45Z
issue,188678,DISABLED test_redispatch_nn_functional_conv_transpose3d_cuda_float16 (__main__.TestTorchFunctionRedispatchOpsCUDA),https://github.com/pytorch/pytorch/issues/188678,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: __torch_function__, module: convolution, module: cuda, module: half",review early; this can change release or regression risk,2026-07-01T15:24:50Z
issue,188336,AArch64 CPU - Inductor benchmark Regressions,https://github.com/pytorch/pytorch/issues/188336,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: arm, module: ci, module: performance, module: regression",review early; this can change release or regression risk,2026-07-01T08:42:20Z
pr,181431,test split group,https://github.com/pytorch/pytorch/pull/181431,ready for maintainer decision,70,approved with no detected failing/pending checks; CI lane labels: ciflow/trunk,final maintainer merge/release decision,2026-06-30T21:07:50Z
pr,187952,Fix out= variant recompilations for torch.max/min/topk,https://github.com/pytorch/pytorch/pull/187952,ready for maintainer decision,70,approved with no detected failing/pending checks; CI lane labels: ciflow/inductor,final maintainer merge/release decision,2026-06-30T14:01:44Z
pr,184545,Fix UntypedStorage.new null allocator crash,https://github.com/pytorch/pytorch/pull/184545,has linked issue,70,PR closes or references open issue numbers; high-priority/blocker/regression label present,review when ready,2026-06-29T23:48:11Z
issue,187444,`torch.compile` vs eager DDP divergence with ResNet50 + AdamW lr=1.0 on two-node NCCL,https://github.com/pytorch/pytorch/issues/187444,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: correctness (silent), module: ddp, module: inductor, oncall: distributed",review early; this can change release or regression risk,2026-06-29T22:37:29Z
issue,188425,torch.compile produces drastically incorrect outputs for model with symbolic dimensions and random operations,https://github.com/pytorch/pytorch/issues/188425,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: correctness (silent), module: determinism, module: inductor, module: random",review early; this can change release or regression risk,2026-06-29T18:27:42Z
issue,188127,torch.compile causes persistent RSS growth during repeated inference in PyTorch 2.9 with ROCM,https://github.com/pytorch/pytorch/issues/188127,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: inductor, module: memory usage, module: performance, module: regression",review early; this can change release or regression risk,2026-06-29T18:20:38Z
pr,182033,Add a utility method for interrupting long running subprocess during run_test.py,https://github.com/pytorch/pytorch/pull/182033,ready for maintainer decision,70,approved with no detected failing/pending checks,final maintainer merge/release decision,2026-06-29T18:16:28Z
issue,188218,[inductor] `LoweringException: AssertionError: max not supported for zero-dimension tensors!`,https://github.com/pytorch/pytorch/issues/188218,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: crash, module: inductor, module: pooling, oncall: pt2",review early; this can change release or regression risk,2026-06-29T17:26:28Z
pr,181614,[TEST] Fix `_debug_handlers` import error in `test_stacks_handler_all_success`,https://github.com/pytorch/pytorch/pull/181614,ready for maintainer decision,70,approved with no detected failing/pending checks; CI lane labels: ciflow/torchtitan,final maintainer merge/release decision,2026-06-26T20:10:01Z
pr,186468,Fix incorrect LR curves in Chained/Sequential scheduler doc images,https://github.com/pytorch/pytorch/pull/186468,ready for maintainer decision,70,approved with no detected failing/pending checks,final maintainer merge/release decision,2026-06-26T19:21:51Z
issue,187951,inductor_timm tests has been broken since 6/12,https://github.com/pytorch/pytorch/issues/187951,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: ci, module: inductor, module: regression, module: tests",review early; this can change release or regression risk,2026-06-26T19:18:48Z
issue,187343,expandable_segments:True crashes on ROCm,https://github.com/pytorch/pytorch/issues/187343,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: crash, module: cudacachingallocator, module: rocm",review early; this can change release or regression risk,2026-06-25T23:29:55Z
pr,185123,[c10d] Add missing runCollectiveChecks for allgather_into_tensor_coalesced in ProcessGroupWrapper,https://github.com/pytorch/pytorch/pull/185123,ready for maintainer decision,70,approved with no detected failing/pending checks,final maintainer merge/release decision,2026-06-25T13:27:19Z
pr,187625,Widen optimizer hyperparameter type hints to accept int and Tensor,https://github.com/pytorch/pytorch/pull/187625,ready for maintainer decision,70,approved with no detected failing/pending checks,final maintainer merge/release decision,2026-06-24T19:58:26Z
issue,187449,`torch.compile` vs eager DDP logits-sum divergence with ResNet50 + AdamW lr=0.01 on two-node NCCL,https://github.com/pytorch/pytorch/issues/187449,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: correctness (silent), module: ddp, module: inductor, module: nccl",review early; this can change release or regression risk,2026-06-24T15:11:18Z
issue,65766,Temporarily disabling grad doesn't work if AMP is enabled,https://github.com/pytorch/pytorch/issues/65766,high-priority/blocker,70,high-priority/blocker/regression label present; component/oncall labels: module: amp (automated mixed precision),review early; this can change release or regression risk,2026-06-24T14:17:20Z
pr,187720,Fix stale_issues year cutoff clamping the day to 28 in every month,https://github.com/pytorch/pytorch/pull/187720,ready for maintainer decision,70,approved with no detected failing/pending checks,final maintainer merge/release decision,2026-06-24T14:02:16Z
issue,187319,torch.compile FX graph does not distinguish custom ops with the same name but different mutates_args,https://github.com/pytorch/pytorch/issues/187319,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: correctness (silent), module: custom-operators, module: fx, module: inductor",review early; this can change release or regression risk,2026-06-24T01:32:04Z
pr,187779,Fix MPS fused RMSNorm: do the weight multiply in fp32 to match CPU/CUDA,https://github.com/pytorch/pytorch/pull/187779,ready for maintainer decision,70,approved with no detected failing/pending checks; CI lane labels: ciflow/mps,final maintainer merge/release decision,2026-06-24T01:12:50Z
issue,187445,"`torch.compile` vs eager DDP loss/grad/logits divergence with ResNet34 + SGD lr=0.1, batch=2, TF32 allowed, and 1MB buckets",https://github.com/pytorch/pytorch/issues/187445,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: correctness (silent), module: ddp, module: determinism, module: inductor",review early; this can change release or regression risk,2026-06-23T18:28:31Z
issue,187446,"`torch.compile` vs eager DDP grad/parameter divergence with ResNet50 + Adam lr=0.05, batch=1, sqrt LR scaling and step scheduler",https://github.com/pytorch/pytorch/issues/187446,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: correctness (silent), module: ddp, module: inductor, module: pt2 accuracy",review early; this can change release or regression risk,2026-06-23T18:28:23Z
issue,187447,"`torch.compile` vs eager DDP parameter-fingerprint divergence with ResNet50 + Adam lr=0.1, grad accumulation, grad clipping, and 1MB buckets",https://github.com/pytorch/pytorch/issues/187447,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: correctness (silent), module: ddp, module: inductor, module: pt2 optimizer",review early; this can change release or regression risk,2026-06-23T18:19:52Z
issue,187448,"`torch.compile` vs eager DDP grad/logits divergence with ResNet50 + AdamW lr=0.01, warmup cosine schedule, grad accumulation, and DDP bucket views",https://github.com/pytorch/pytorch/issues/187448,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: correctness (silent), module: ddp, module: inductor, oncall: distributed",review early; this can change release or regression risk,2026-06-23T18:19:45Z
pr,187731,Fix extract_scripts step-index zero-padding off-by-one,https://github.com/pytorch/pytorch/pull/187731,ready for maintainer decision,70,approved with no detected failing/pending checks,final maintainer merge/release decision,2026-06-23T16:59:39Z
issue,176784,[CUDA] AdaptiveAvgPool2d: START_IND macro causes int32 overflow → OOB read,https://github.com/pytorch/pytorch/issues/176784,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: 64-bit, module: crash, module: cuda, module: edge cases",review early; this can change release or regression risk,2026-06-23T08:05:36Z
pr,174275,Adjust to use __riscv_v to detect for RISC-V Vector Extension,https://github.com/pytorch/pytorch/pull/174275,ready for maintainer decision,70,approved with no detected failing/pending checks,final maintainer merge/release decision,2026-06-20T07:48:12Z
pr,174826,add mem_get_info usage notes to CUDA memory management docs,https://github.com/pytorch/pytorch/pull/174826,ready for maintainer decision,70,approved with no detected failing/pending checks; CI lane labels: ciflow/trunk,final maintainer merge/release decision,2026-06-20T01:26:30Z
pr,174667,Update tolerances for tan* functions w/ complex dtypes on XPU,https://github.com/pytorch/pytorch/pull/174667,ready for maintainer decision,70,approved with no detected failing/pending checks,final maintainer merge/release decision,2026-06-19T03:48:39Z
pr,186012,Add ownership headers to GitHub workflows and a linter to enforce them,https://github.com/pytorch/pytorch/pull/186012,ready for maintainer decision,70,"approved with no detected failing/pending checks; CI lane labels: ciflow/docker, ciflow/inductor-pallas",final maintainer merge/release decision,2026-06-19T01:01:43Z
pr,187130,[BE] Do not try to install CUDA-12.8 wheels in torchbench images,https://github.com/pytorch/pytorch/pull/187130,ready for maintainer decision,70,approved with no detected failing/pending checks; CI lane labels: ciflow/docker,final maintainer merge/release decision,2026-06-19T01:01:39Z
issue,184880,[ROCm] RX 9060 XT 16GB hard-capped at ~8GB usable VRAM in PyTorch — segfault on allocation beyond cap,https://github.com/pytorch/pytorch/issues/184880,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: crash, module: rocm",review early; this can change release or regression risk,2026-06-18T15:40:41Z
pr,170051,Add pivoted QR decomposition to ATen and torch.linalg,https://github.com/pytorch/pytorch/pull/170051,ready for maintainer decision,70,"approved with no detected failing/pending checks; CI lane labels: ciflow/inductor, ciflow/torchtitan",final maintainer merge/release decision,2026-06-18T11:06:22Z
issue,187320,Stable ABI isn't stable when PyTorch changes the default CUDA version,https://github.com/pytorch/pytorch/issues/187320,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: abi, module: binaries, module: cpp-extensions, module: cuda",review early; this can change release or regression risk,2026-06-15T23:57:42Z
issue,186952,GPU is not accessible on some MacOS-15 runners,https://github.com/pytorch/pytorch/issues/186952,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: arm, module: ci, module: macos",review early; this can change release or regression risk,2026-06-15T18:23:28Z
issue,186516,torch.cuda.device_count() returns 0 on ROCm when amdsmi is installed (duplicate libamd_smi ODR clash); HIP reports 1,https://github.com/pytorch/pytorch/issues/186516,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: binaries, module: rocm",review early; this can change release or regression risk,2026-06-15T14:23:41Z
pr,176321,test: replaces deprecated numpy.linalg.linalg module,https://github.com/pytorch/pytorch/pull/176321,ready for maintainer decision,70,approved with no detected failing/pending checks,final maintainer merge/release decision,2026-06-14T20:03:43Z
pr,187068,[torch][profiler] Fix refcount leaks in _TensorMetadata layout/dtype getters,https://github.com/pytorch/pytorch/pull/187068,ready for maintainer decision,70,approved with no detected failing/pending checks; CI lane labels: ciflow/trunk,final maintainer merge/release decision,2026-06-14T16:17:54Z
pr,186635,c10/util/ArrayRef.h: specialize std::ranges::enable_borrowed_range,https://github.com/pytorch/pytorch/pull/186635,ready for maintainer decision,70,approved with no detected failing/pending checks; CI lane labels: ciflow/trunk,final maintainer merge/release decision,2026-06-12T21:53:32Z
pr,171780,Add support shared mempool for accelerator Graph,https://github.com/pytorch/pytorch/pull/171780,ready for maintainer decision,70,approved with no detected failing/pending checks; CI lane labels: ciflow/trunk,final maintainer merge/release decision,2026-06-11T16:50:34Z
pr,178258,[submodule] Upgrade to protobuf v33.6,https://github.com/pytorch/pytorch/pull/178258,ready for maintainer decision,70,"approved with no detected failing/pending checks; CI lane labels: ciflow/linux-aarch64, ciflow/trunk",final maintainer merge/release decision,2026-06-11T07:36:22Z
issue,184431,[vllm] [2.12 regression][multimodal] Qwen2-Audio text-then-audio_embeds: prompt_embeds vs raw-text outputs diverge under --enforce-eager,https://github.com/pytorch/pytorch/issues/184431,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: bfloat16, module: correctness (silent), module: numerical-stability, module: regression",review early; this can change release or regression risk,2026-06-11T01:16:13Z
pr,183389,Port 7 tests under test/distributed folder to XPU,https://github.com/pytorch/pytorch/pull/183389,ready for maintainer decision,70,approved with no detected failing/pending checks; CI lane labels: ciflow/inductor,final maintainer merge/release decision,2026-06-10T06:08:23Z
issue,178880,`torch.compile` produces different classification output for model using E8M0 bit-manipulation quantization (`view(int32) → bitshift → clamp → uint8`),https://github.com/pytorch/pytorch/issues/178880,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: correctness (silent), module: inductor, oncall: pt2",review early; this can change release or regression risk,2026-06-09T23:11:14Z
pr,185189,[Test] Make binary_ufuncs file device agnostic for out-of-tree backends,https://github.com/pytorch/pytorch/pull/185189,ready for maintainer decision,70,approved with no detected failing/pending checks; CI lane labels: ciflow/trunk,final maintainer merge/release decision,2026-06-09T21:20:23Z
issue,184549,[vllm] [2.12 regression][Inductor] assert_size_stride (8192 vs 512) in BERT-with-rope AOT-compiled path for Nomic v2 MoE pooling model,https://github.com/pytorch/pytorch/issues/184549,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: custom-operators, module: inductor, module: pt2-dispatcher, module: regression",review early; this can change release or regression risk,2026-06-09T01:23:00Z
issue,186174,ROCm: Flash/mem-efficient SDPA hangs during batched autoregressive generation with left-padded sequences,https://github.com/pytorch/pytorch/issues/186174,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: deadlock, module: regression, module: rocm, module: sdpa",review early; this can change release or regression risk,2026-06-08T17:08:02Z
issue,186212,C++ API: at::batch_norm crashes with an internal assertion on invalid inputs,https://github.com/pytorch/pytorch/issues/186212,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: cpp, module: cpu, module: crash, module: norms and normalization",review early; this can change release or regression risk,2026-06-08T14:29:40Z
pr,186519,c10/util/intrusive_ptr.h: add [[nodiscard]] to query methods,https://github.com/pytorch/pytorch/pull/186519,ready for maintainer decision,70,approved with no detected failing/pending checks; CI lane labels: ciflow/trunk,final maintainer merge/release decision,2026-06-08T13:55:27Z
pr,186575,Update FindMKLDNN CMake configuration,https://github.com/pytorch/pytorch/pull/186575,ready for maintainer decision,70,approved with no detected failing/pending checks,final maintainer merge/release decision,2026-06-08T03:07:22Z
issue,73294,Internal assert failed at rref_context.cpp,https://github.com/pytorch/pytorch/issues/73294,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: crash, module: rpc, oncall: distributed, oncall: distributed infra",review early; this can change release or regression risk,2026-06-06T16:29:49Z
pr,186038,add distortion benchmark,https://github.com/pytorch/pytorch/pull/186038,ready for maintainer decision,70,approved with no detected failing/pending checks,final maintainer merge/release decision,2026-06-05T21:12:04Z
pr,185821,[BE][Ez]: Add c10::SmallVector reserve calls and simplify log func calls,https://github.com/pytorch/pytorch/pull/185821,ready for maintainer decision,70,approved with no detected failing/pending checks; CI lane labels: ciflow/mps,final maintainer merge/release decision,2026-06-05T19:51:27Z
issue,76865,Object-base collectives create tensors at unexpected devices,https://github.com/pytorch/pytorch/issues/76865,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: c10d, oncall: distributed, oncall: distributed infra",review early; this can change release or regression risk,2026-06-05T17:12:58Z
issue,82293,pytorch's checkpoint_wrapper does not save memory while fairscale's checkpoint_wrapper saves huge memory,https://github.com/pytorch/pytorch/issues/82293,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: activation checkpointing, module: fsdp, oncall: distributed, oncall: distributed parallelisms",review early; this can change release or regression risk,2026-06-05T17:11:27Z
pr,186268,[TEST][CUDA] Add some additional cleanup to `TestCudaallocator`and re-enable `test_memory_plots`,https://github.com/pytorch/pytorch/pull/186268,ready for maintainer decision,70,approved with no detected failing/pending checks,final maintainer merge/release decision,2026-06-05T00:21:48Z
issue,169741,ReplicationPad2dwith extremely large padding value causes segmentation fault,https://github.com/pytorch/pytorch/issues/169741,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: crash, module: nn",review early; this can change release or regression risk,2026-06-03T22:45:46Z
pr,185934,[NVSHMEM] Fix link_whole and -fPIC for nvshmem_device_link_target,https://github.com/pytorch/pytorch/pull/185934,ready for maintainer decision,70,"approved with no detected failing/pending checks; CI lane labels: ciflow/h100-symm-mem, ciflow/trunk",final maintainer merge/release decision,2026-06-03T18:44:54Z
issue,185494,"[dynamo] LazyBatchNorm2d fails under torch.compile(dynamic=True) — ""SymIntArrayRef expected to contain only concrete integers""",https://github.com/pytorch/pytorch/issues/185494,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: crash, module: dynamic shapes, module: dynamo, module: nn",review early; this can change release or regression risk,2026-06-03T03:36:56Z
issue,185346,[Inductor] Performance regression of phlippe_densenet model due to non-fused Cat kernels from new fusion mode,https://github.com/pytorch/pytorch/issues/185346,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: inductor, module: performance, module: regression, oncall: pt2",review early; this can change release or regression risk,2026-06-02T18:23:29Z
issue,156819,FakeTensorUpdater doesn't support HOPs,https://github.com/pytorch/pytorch/issues/156819,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: faketensor, module: inductor, module: pt2-dispatcher, oncall: pt2",review early; this can change release or regression risk,2026-06-02T17:04:18Z
pr,183056,[lint] Fix PYI036 violations and re-enable the rule,https://github.com/pytorch/pytorch/pull/183056,ready for maintainer decision,70,"approved with no detected failing/pending checks; CI lane labels: ciflow/inductor, ciflow/torchtitan",final maintainer merge/release decision,2026-06-01T15:07:15Z
issue,113586,mps bug: failed assertion `[MPSNDArrayDescriptor sliceDimension:withSubrange:] error: subRange.start (8192) is not less than length of dimension[1] (512)',https://github.com/pytorch/pytorch/issues/113586,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: crash, module: mps",review early; this can change release or regression risk,2026-06-01T06:06:24Z
pr,182924,[Fix] DLPack toDLPackNonOwning: handle MPS storage_offset,https://github.com/pytorch/pytorch/pull/182924,ready for maintainer decision,70,approved with no detected failing/pending checks,final maintainer merge/release decision,2026-05-31T15:02:25Z
issue,179030,XPU: Segmentation Fault when using newer drivers,https://github.com/pytorch/pytorch/issues/179030,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: crash, module: regression, module: xpu",review early; this can change release or regression risk,2026-05-30T22:16:27Z
pr,180828,Reuse ACCELERATOR_TYPE in test UT files,https://github.com/pytorch/pytorch/pull/180828,ready for maintainer decision,70,"approved with no detected failing/pending checks; CI lane labels: ciflow/dtensor, ciflow/h100-symm-mem, ciflow/inductor, ciflow/trunk",final maintainer merge/release decision,2026-05-29T01:10:22Z
pr,184163,[Test]Refactor test/nn/test_dropout.py to be device-agnostic,https://github.com/pytorch/pytorch/pull/184163,ready for maintainer decision,70,approved with no detected failing/pending checks; CI lane labels: ciflow/trunk,final maintainer merge/release decision,2026-05-28T16:46:02Z
issue,173313,Bad perf for torch.compile for merge_attention with transposed inputs,https://github.com/pytorch/pytorch/issues/173313,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: aotdispatch, module: pt2-dispatcher, oncall: pt2",review early; this can change release or regression risk,2026-05-26T17:55:01Z
issue,181869,[vllm] [torch nightly] AOTAutogradCache.save: _pickle.PicklingError on triton launcher (DFlashDraftModel speculative-decode path),https://github.com/pytorch/pytorch/issues/181869,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: aotdispatch, module: inductor, module: pickle, module: pt2-dispatcher",review early; this can change release or regression risk,2026-05-26T17:54:17Z
issue,184027,[inductor] `torch.compile` crashes on convolutions when weight is a plain tensor — stride prediction mismatch,https://github.com/pytorch/pytorch/issues/184027,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: convolution, module: crash, module: inductor, oncall: pt2",review early; this can change release or regression risk,2026-05-26T16:49:54Z
issue,120601,Torchrun arguments shadowing script arguments,https://github.com/pytorch/pytorch/issues/120601,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: intel, module: regression",review early; this can change release or regression risk,2026-05-26T02:36:50Z
issue,183401,[inductor][cpu] resnet50_quantized_qat CPP wrapper torch._inductor.exc.InductorError: CppCompileError: C++ compile error in 2026-05-11 nightly release,https://github.com/pytorch/pytorch/issues/183401,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: cpu, module: inductor, module: regression, oncall: pt2",review early; this can change release or regression risk,2026-05-21T18:26:50Z
issue,161295,UNSTABLE inductor-periodic / cuda12.8-py3.10-gcc9-sm80 / test (inductor_torchbench_smoketest_perf),https://github.com/pytorch/pytorch/issues/161295,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: ci, module: inductor, oncall: pt2",review early; this can change release or regression risk,2026-05-21T18:16:03Z
issue,177834,`torch.nn.LSTM.forward` produces incorrect outputs (or crashes) on ROCm,https://github.com/pytorch/pytorch/issues/177834,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: correctness (silent), module: crash, module: rnn, module: rocm",review early; this can change release or regression risk,2026-05-20T22:35:45Z
issue,178726,Greatly increased CPU memory use on aarch64 H100 node from torch 2.9.0 to torch 2.10.0,https://github.com/pytorch/pytorch/issues/178726,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: arm, module: cuda, module: memory usage, module: regression",review early; this can change release or regression risk,2026-05-20T21:20:51Z
issue,87960,crash when call `torch.set_num_interop_threads` twice,https://github.com/pytorch/pytorch/issues/87960,high-priority/blocker,70,high-priority/blocker/regression label present; component/oncall labels: module: crash,review early; this can change release or regression risk,2026-05-20T18:32:03Z
issue,183890,torch.compile hits RecursionError on nn.Module with custom __getattr__,https://github.com/pytorch/pytorch/issues/183890,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: crash, module: dynamo, oncall: pt2",review early; this can change release or regression risk,2026-05-18T21:35:07Z
issue,182949,[CD] Drop CPython 3.13t from binary build matrix (manylinux upstream removed it 2026-05-07),https://github.com/pytorch/pytorch/issues/182949,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: binaries, module: ci, module: dependency bug, module: docker",review early; this can change release or regression risk,2026-05-14T14:23:49Z
issue,13246,DataLoader num_workers > 0 causes CPU memory from parent process to be replicated in all worker processes,https://github.com/pytorch/pytorch/issues/13246,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: dataloader, module: dependency bug, module: memory usage, module: molly-guard",review early; this can change release or regression risk,2026-05-10T18:38:26Z
issue,91570,Security policy impractical / lacks contact information?,https://github.com/pytorch/pytorch/issues/91570,high-priority/blocker,70,high-priority/blocker/regression label present; component/oncall labels: module: docs,review early; this can change release or regression risk,2026-05-06T20:41:45Z
issue,182263,"libtorch_cpu.so BLAS symbols are interposable, causing crashes when another BLAS is present in the process",https://github.com/pytorch/pytorch/issues/182263,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: binaries, module: crash, module: openblas",review early; this can change release or regression risk,2026-05-04T17:30:57Z
issue,179891,"torch.xpu.get_device_properties() segfaults on Intel ARC PRO B70 (Battlemage/BMG, device 0xe223)",https://github.com/pytorch/pytorch/issues/179891,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: crash, module: xpu",review early; this can change release or regression risk,2026-05-02T17:28:20Z
issue,177714,XPU: Fatal Level Zero `RuntimeErrors` instead of recoverable `OutOfMemoryError`,https://github.com/pytorch/pytorch/issues/177714,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: crash, module: memory usage, module: xpu",review early; this can change release or regression risk,2026-04-29T21:55:24Z
issue,181362,DISABLED deit_tiny_patch16_224.fb_in1k fail_accuracy on ROCm dynamic_inductor_timm_training,https://github.com/pytorch/pytorch/issues/181362,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: inductor, module: pt2 accuracy, module: regression, module: rocm",review early; this can change release or regression risk,2026-04-24T11:30:22Z
issue,139989,Support AC with graph break,https://github.com/pytorch/pytorch/issues/139989,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: activation checkpointing, module: dynamo, module: higher order operators, module: pt2-dispatcher",review early; this can change release or regression risk,2026-04-23T22:12:01Z
issue,143487,Aborted (core dumped) in `replication_pad1d`,https://github.com/pytorch/pytorch/issues/143487,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: crash, module: error checking",review early; this can change release or regression risk,2026-04-22T05:25:08Z
issue,143484,Segmentation fault (core dumped) in `gru_cell`,https://github.com/pytorch/pytorch/issues/143484,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: crash, module: empty tensor, module: error checking",review early; this can change release or regression risk,2026-04-22T05:23:11Z
issue,181024,[CUDA] `test_dlpack_tensor_on_different_device` fails with `torch.device` instead of `torch.cuda.device`,https://github.com/pytorch/pytorch/issues/181024,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: cuda, module: dlpack, module: regression, module: tests",review early; this can change release or regression risk,2026-04-21T21:05:52Z
issue,179804,torch.compile + aot_eager: backward ignores inner autocast(enabled=False) vs eager,https://github.com/pytorch/pytorch/issues/179804,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: aotdispatch, module: autograd, module: pt2-dispatcher, oncall: pt2",review early; this can change release or regression risk,2026-04-21T15:44:21Z
issue,141896,"CheckpointError with checkpoint(..., use_reentrant=False) & autocast()",https://github.com/pytorch/pytorch/issues/141896,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: activation checkpointing, module: amp (automated mixed precision)",review early; this can change release or regression risk,2026-04-19T10:12:12Z
issue,178939,Pytorch needs Kineto even when using USE_KINETO=OFF,https://github.com/pytorch/pytorch/issues/178939,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: build, module: regression",review early; this can change release or regression risk,2026-04-13T17:10:46Z
issue,168058,[DTensor] inconsistency of distributed RNG (Shard(0) != Shard(1) or 8 GPU != 4 GPU != 1 GPU),https://github.com/pytorch/pytorch/issues/168058,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: dtensor, oncall: distributed",review early; this can change release or regression risk,2026-04-06T17:45:17Z
issue,41486,"LSTMs leak memory in CPU PyTorch 1.5.1, 1.6, and 1.7 on Linux",https://github.com/pytorch/pytorch/issues/41486,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: cpu, module: memory usage, module: rnn",review early; this can change release or regression risk,2026-03-27T05:49:45Z
issue,145702,PyTorch VS2022 official build Windows binary illegal instruction on AVX2(max ISA level) CPU,https://github.com/pytorch/pytorch/issues/145702,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: build, module: cpu, module: crash, module: windows",review early; this can change release or regression risk,2026-03-26T08:22:43Z
issue,165448,`import torch` fails on 3.13 Linux nightly jobs with AST parsing errors,https://github.com/pytorch/pytorch/issues/165448,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: binaries, module: regression",review early; this can change release or regression risk,2026-03-25T11:16:27Z
issue,152302,NCCL out of memory error after updating to PyTorch 2.7,https://github.com/pytorch/pytorch/issues/152302,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: nccl, module: regression, oncall: distributed",review early; this can change release or regression risk,2026-03-21T02:43:04Z
issue,152954,DTensor placement propagation for `as_strided` missing,https://github.com/pytorch/pytorch/issues/152954,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: pt2-dispatcher, oncall: distributed, oncall: pt2",review early; this can change release or regression risk,2026-03-19T22:56:29Z
issue,58743,Python Array API Compatibility Tracker,https://github.com/pytorch/pytorch/issues/58743,high-priority/blocker,70,high-priority/blocker/regression label present; component/oncall labels: module: python array api,review early; this can change release or regression risk,2026-03-16T20:56:13Z
issue,142048,Device check missing in torch.linalg.solve_triangular leading to hard crash,https://github.com/pytorch/pytorch/issues/142048,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: crash, module: linear algebra, module: mps",review early; this can change release or regression risk,2026-03-14T10:13:35Z
issue,172537,[Regression] 20x slower vision model forward pass in PyTorch 2.9.1 vs 2.8.0,https://github.com/pytorch/pytorch/issues/172537,high-priority/blocker,70,high-priority/blocker/regression label present; component/oncall labels: module: performance,review early; this can change release or regression risk,2026-03-13T19:23:37Z
issue,170787,[ARM] Unit test regression AArch64 - int4mm kernels require bf16 hw support,https://github.com/pytorch/pytorch/issues/170787,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: arm, module: regression",review early; this can change release or regression risk,2026-03-12T14:42:25Z
issue,30774,Natively Declarable Fast-path Functions,https://github.com/pytorch/pytorch/issues/30774,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: dispatch, module: internals",review early; this can change release or regression risk,2026-03-11T19:31:57Z
issue,124484,HOP dispatch isn't faithful,https://github.com/pytorch/pytorch/issues/124484,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: pt2-dispatcher, oncall: pt2",review early; this can change release or regression risk,2026-03-11T18:36:02Z
issue,175928,[release 2.11][vllm] Language Models Tests (Extra Standard) 1 test_models[float-jason9693/Qwen2.5-1.5B-apeach],https://github.com/pytorch/pytorch/issues/175928,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: regression, module: tests, module: vllm, oncall: releng",review early; this can change release or regression risk,2026-03-11T14:47:43Z
issue,167373,Torch Tensor -> Nanobind ndarray breaks cuda graphs on torch 2.9,https://github.com/pytorch/pytorch/issues/167373,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: cuda graphs, module: dlpack, module: regression",review early; this can change release or regression risk,2026-03-10T22:12:43Z
issue,163501,Improve Registration Error message with const SymIntArrayRef&in PyTorch 2.5+,https://github.com/pytorch/pytorch/issues/163501,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: cpp-extensions, module: dispatch, module: dynamic shapes, module: regression",review early; this can change release or regression risk,2026-03-10T16:09:43Z
issue,157128,"Windows Source Build Fails with OSError: [WinError 126] on aoti_custom_ops.dll for RTX 5080 (sm_120), Pre-built PyTorch Works",https://github.com/pytorch/pytorch/issues/157128,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: aotinductor, module: build, module: regression, module: windows",review early; this can change release or regression risk,2026-03-09T23:25:25Z
issue,174704,Recompile time with torch.compile 2.10,https://github.com/pytorch/pytorch/issues/174704,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: compile-time, module: dynamic shapes, module: fsdp, module: regression",review early; this can change release or regression risk,2026-03-07T01:47:55Z
issue,176285,"AArch64 test_cpu_repro - Expected to not find ""transpose_mxn"" but found it",https://github.com/pytorch/pytorch/issues/176285,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: arm, module: ci, module: inductor, module: regression",review early; this can change release or regression risk,2026-03-04T11:59:53Z
issue,142836,[MPS] Incorrect output from convolution ops with large dimensions,https://github.com/pytorch/pytorch/issues/142836,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: correctness (silent), module: mps, module: regression",review early; this can change release or regression risk,2026-03-01T20:13:15Z
issue,142834,Segmentation fault in `replication_pad2d_backward`,https://github.com/pytorch/pytorch/issues/142834,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: crash, module: edge cases, module: nn",review early; this can change release or regression risk,2026-02-24T15:42:59Z
issue,175436,Intermittent SIGILL in `torch.sin` on CPU (`torch==2.10.0+cpu`) on Xeon E5-2670; crash in `mkl_vml_kernel_dSin_*`,https://github.com/pytorch/pytorch/issues/175436,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: cpu, module: crash, module: intel, module: mkl",review early; this can change release or regression risk,2026-02-23T18:47:29Z
issue,142459,Segmentation fault (core dumped) in `max_pool3d_with_indices_backward`,https://github.com/pytorch/pytorch/issues/142459,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: crash, module: edge cases, module: error checking, module: nn",review early; this can change release or regression risk,2026-02-18T14:32:11Z
issue,142457,Aborted (core dumped) in `slow_conv_transpose3d`,https://github.com/pytorch/pytorch/issues/142457,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: convolution, module: crash, module: edge cases, module: error checking",review early; this can change release or regression risk,2026-02-18T14:31:11Z
issue,168997,[Umbrella issue] Deployment of index updates to download.pytorch.org - Switch dependencies that are on pypi to utilize pypi's CDN,https://github.com/pytorch/pytorch/issues/168997,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: binaries, module: ci",review early; this can change release or regression risk,2026-02-03T19:38:02Z
issue,94765,MPS internal error in `torch.gather` when last dimension is a singleton dimension,https://github.com/pytorch/pytorch/issues/94765,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: crash, module: mps",review early; this can change release or regression risk,2026-01-30T06:39:30Z
issue,173574,`torch.arange` triggers `SIGFPE` (Floating Point Exception) instead of RuntimeError when writing float results to `int64` out tensor,https://github.com/pytorch/pytorch/issues/173574,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: crash, module: edge cases, module: error checking",review early; this can change release or regression risk,2026-01-28T01:25:30Z
issue,150098,Aborted (core dumped) raw pointer API is exposed directly to Python,https://github.com/pytorch/pytorch/issues/150098,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: crash, module: cuda, module: error checking",review early; this can change release or regression risk,2026-01-27T13:07:06Z
issue,172907,[CI] Changes to install scripts in lintrunner.toml are not tested in CI,https://github.com/pytorch/pytorch/issues/172907,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: ci, module: typing",review early; this can change release or regression risk,2026-01-26T18:14:46Z
issue,172798,`backward` does not respect `with torch.device(device)`,https://github.com/pytorch/pytorch/issues/172798,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: autograd, module: regression",review early; this can change release or regression risk,2026-01-26T18:12:31Z
issue,51039,Status of pip wheels with _GLIBCXX_USE_CXX11_ABI=1,https://github.com/pytorch/pytorch/issues/51039,high-priority/blocker,70,"high-priority/blocker/regression label present; needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: abi, module: binaries, module: cpp",route to design/research decision before patch planning,2025-05-01T18:32:44Z
issue,93880,"`PYTORCH_DEBUG_MODE`, better invalid index embedding lookup error message on cuda",https://github.com/pytorch/pytorch/issues/93880,high-priority/blocker,70,high-priority/blocker/regression label present; needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: python frontend,route to design/research decision before patch planning,2024-09-14T15:43:15Z
issue,23756,[feature request] Core API for invertible/inplace and flow-like ops + memory-saving (hookless?) reversible sequential container for RevNets to allow for much larger batch-sizes in academic setting,https://github.com/pytorch/pytorch/issues/23756,high-priority/blocker,70,"high-priority/blocker/regression label present; needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: distributions, module: nn",route to design/research decision before patch planning,2023-08-21T20:47:19Z
issue,68768,[feature request] making pytorch less noisy,https://github.com/pytorch/pytorch/issues/68768,high-priority/blocker,70,"high-priority/blocker/regression label present; needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: cpp-extensions, module: ux",route to design/research decision before patch planning,2022-06-02T05:17:44Z
issue,44428,Support for CUDA matrix multiplication on long (and other integer) tensors,https://github.com/pytorch/pytorch/issues/44428,high-priority/blocker,70,"high-priority/blocker/regression label present; needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: cublas, module: cuda, module: linear algebra",route to design/research decision before patch planning,2021-11-12T03:38:36Z
pr,186418,Fix AOT export GraphModule serialization,https://github.com/pytorch/pytorch/pull/186418,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,review diff and decide acceptance path,2026-07-08T19:16:31Z
pr,186431,Fix ModuleTracker with Dynamo runtime modules,https://github.com/pytorch/pytorch/pull/186431,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers,review diff and decide acceptance path,2026-07-08T18:56:47Z
pr,186443,Validate random_ and uniform_ meta arguments,https://github.com/pytorch/pytorch/pull/186443,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers,review diff and decide acceptance path,2026-07-08T18:53:27Z
pr,184347,[inductor] Add third reduction tile for strided block pointers,https://github.com/pytorch/pytorch/pull/184347,ready for maintainer decision,67,"needs-review label present; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",review diff and decide acceptance path,2026-07-08T18:41:19Z
pr,186454,Warn on unsupported custom op schema types,https://github.com/pytorch/pytorch/pull/186454,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers,review diff and decide acceptance path,2026-07-08T18:37:58Z
pr,186459,Fix fake view strides for AOT group norm tracing,https://github.com/pytorch/pytorch/pull/186459,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,review diff and decide acceptance path,2026-07-08T18:25:42Z
pr,186463,Split foreach bool inplace error samples,https://github.com/pytorch/pytorch/pull/186463,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers,review diff and decide acceptance path,2026-07-08T18:05:31Z
pr,186464,Fix singleton quotient divisibility replacement,https://github.com/pytorch/pytorch/pull/186464,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,review diff and decide acceptance path,2026-07-08T18:02:24Z
pr,186470,Raise user error for joint parameter mutation export,https://github.com/pytorch/pytorch/pull/186470,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,review diff and decide acceptance path,2026-07-08T17:47:10Z
pr,186471,Handle symbolic tensors in overlap status,https://github.com/pytorch/pytorch/pull/186471,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,review diff and decide acceptance path,2026-07-08T17:42:15Z
pr,185926,Preserve tuple arguments in export serde,https://github.com/pytorch/pytorch/pull/185926,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,review diff and decide acceptance path,2026-07-08T17:39:36Z
pr,185929,Fix Dynamo benchmark model state isolation,https://github.com/pytorch/pytorch/pull/185929,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,review diff and decide acceptance path,2026-07-08T17:33:56Z
pr,186007,Preserve eager post accumulate hook ordering in AOTAutograd,https://github.com/pytorch/pytorch/pull/186007,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,review diff and decide acceptance path,2026-07-08T16:56:34Z
pr,186833,Remove stale export OpInfo xfails,https://github.com/pytorch/pytorch/pull/186833,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers,review diff and decide acceptance path,2026-07-08T16:34:31Z
pr,186478,Fix AOTAutograd resized out tensor mutation replay,https://github.com/pytorch/pytorch/pull/186478,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,review diff and decide acceptance path,2026-07-08T16:27:56Z
pr,186488,Improve schema errors for symbolic int list mismatches,https://github.com/pytorch/pytorch/pull/186488,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,review diff and decide acceptance path,2026-07-08T16:20:31Z
pr,186495,Initialize compile context for non-strict export tracing,https://github.com/pytorch/pytorch/pull/186495,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers,review diff and decide acceptance path,2026-07-08T16:09:24Z
pr,186503,Preserve input aliases for multi-output view returns,https://github.com/pytorch/pytorch/pull/186503,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,review diff and decide acceptance path,2026-07-08T15:31:42Z
pr,186506,Register missing CIA decomposition py_impls,https://github.com/pytorch/pytorch/pull/186506,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,review diff and decide acceptance path,2026-07-08T15:23:46Z
pr,186517,Detect aliased subclass inputs after desugaring,https://github.com/pytorch/pytorch/pull/186517,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,review diff and decide acceptance path,2026-07-08T15:05:12Z
pr,185793,Add ONNX lowering for prims.remainder,https://github.com/pytorch/pytorch/pull/185793,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers,review diff and decide acceptance path,2026-07-08T15:02:11Z
pr,184432,Serialize cudagraph tree runtime,https://github.com/pytorch/pytorch/pull/184432,ready for maintainer decision,67,"needs-review label present; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",review diff and decide acceptance path,2026-07-08T14:57:12Z
pr,184436,Fix Triton async compile cache keys,https://github.com/pytorch/pytorch/pull/184436,ready for maintainer decision,67,"needs-review label present; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",review diff and decide acceptance path,2026-07-08T14:52:50Z
pr,186531,Remove stale Dynamo skip for max_pool1d pooling test,https://github.com/pytorch/pytorch/pull/186531,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers,review diff and decide acceptance path,2026-07-08T14:39:01Z
pr,186533,Fix SymInt vector dispatcher signature detection,https://github.com/pytorch/pytorch/pull/186533,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers,review diff and decide acceptance path,2026-07-08T14:33:01Z
pr,187601,[functorch] Fix dynamic layers under inference_mode,https://github.com/pytorch/pytorch/pull/187601,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers,review diff and decide acceptance path,2026-07-08T14:25:35Z
pr,187592,Support registered HOPs in FlopCounterMode,https://github.com/pytorch/pytorch/pull/187592,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers,review diff and decide acceptance path,2026-07-08T13:16:23Z
pr,186561,Preserve autograd metadata for functionalized inplace views,https://github.com/pytorch/pytorch/pull/186561,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,review diff and decide acceptance path,2026-07-08T11:56:59Z
pr,186409,Fix subclass view replay with dynamic metadata,https://github.com/pytorch/pytorch/pull/186409,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,review diff and decide acceptance path,2026-07-08T11:44:39Z
pr,185861,Fix ViewMeta serialization for special views,https://github.com/pytorch/pytorch/pull/185861,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers,review diff and decide acceptance path,2026-07-08T11:37:57Z
pr,185897,Support concrete enum types in torch.library.define schemas,https://github.com/pytorch/pytorch/pull/185897,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers,review diff and decide acceptance path,2026-07-08T10:30:52Z
pr,183628,Fix fractional max pool RNG in Inductor,https://github.com/pytorch/pytorch/pull/183628,ready for maintainer decision,67,"needs-review label present; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",review diff and decide acceptance path,2026-07-08T09:29:00Z
pr,183701,Fix CPU vectorized reduction tail stores,https://github.com/pytorch/pytorch/pull/183701,ready for maintainer decision,67,"needs-review label present; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",review diff and decide acceptance path,2026-07-08T08:51:21Z
pr,185995,Fix stale pending-backward state for dead CUDA graph paths,https://github.com/pytorch/pytorch/pull/185995,ready for maintainer decision,67,"needs-review label present; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",review diff and decide acceptance path,2026-07-08T08:46:38Z
pr,184667,Fix cond branch tensor constants in run_decompositions,https://github.com/pytorch/pytorch/pull/184667,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers,review diff and decide acceptance path,2026-07-08T08:20:27Z
pr,184695,Fix all-masked TransformerEncoder compile consistency,https://github.com/pytorch/pytorch/pull/184695,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers,review diff and decide acceptance path,2026-07-08T07:35:30Z
pr,184709,Fix Dynamo tensor subclass constructor side effects,https://github.com/pytorch/pytorch/pull/184709,ready for maintainer decision,67,"needs-review label present; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",review diff and decide acceptance path,2026-07-08T07:25:01Z
pr,184917,Make FX traceback preserve state thread-local,https://github.com/pytorch/pytorch/pull/184917,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers,review diff and decide acceptance path,2026-07-08T07:03:00Z
pr,184928,Fix linalg_pinv tensor tolerance broadcasting,https://github.com/pytorch/pytorch/pull/184928,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,review diff and decide acceptance path,2026-07-08T06:57:51Z
pr,185059,Fix non-strict export tensor subclass attr tracing,https://github.com/pytorch/pytorch/pull/185059,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers,review diff and decide acceptance path,2026-07-08T06:35:29Z
pr,184015,[inductor] Handle scalar workaround args in pattern checks,https://github.com/pytorch/pytorch/pull/184015,ready for maintainer decision,67,"needs-review label present; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",review diff and decide acceptance path,2026-07-08T06:31:50Z
pr,185088,Avoid graphsafe RNG inputs in descriptor compile,https://github.com/pytorch/pytorch/pull/185088,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,review diff and decide acceptance path,2026-07-08T06:04:40Z
pr,184747,Remove no-op SDPA masks before backend dispatch,https://github.com/pytorch/pytorch/pull/184747,ready for maintainer decision,67,"needs-review label present; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",review diff and decide acceptance path,2026-07-08T04:16:39Z
pr,183879,[inductor][cpu] Enable int8 WOQ GEMM template on Arm,https://github.com/pytorch/pytorch/pull/183879,ready for maintainer decision,67,"needs-review label present; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",review diff and decide acceptance path,2026-07-08T04:13:41Z
pr,183908,Fix deferred runtime assert symbol availability,https://github.com/pytorch/pytorch/pull/183908,ready for maintainer decision,67,"needs-review label present; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",review diff and decide acceptance path,2026-07-08T04:05:11Z
pr,183958,Enable SDPA constraint cache by default,https://github.com/pytorch/pytorch/pull/183958,ready for maintainer decision,67,"needs-review label present; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",review diff and decide acceptance path,2026-07-08T03:48:03Z
pr,184948,Fix dynamo export for tensor closures in output pytrees,https://github.com/pytorch/pytorch/pull/184948,ready for maintainer decision,67,"needs-review label present; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",review diff and decide acceptance path,2026-07-08T02:25:10Z
pr,184966,Preserve unflattened module forward signatures,https://github.com/pytorch/pytorch/pull/184966,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers,review diff and decide acceptance path,2026-07-08T02:11:28Z
pr,184974,Fix register_autograd wrappers under torch.func.grad,https://github.com/pytorch/pytorch/pull/184974,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers,review diff and decide acceptance path,2026-07-08T01:57:49Z
pr,184975,Support FakeTensor module materialization via to_empty,https://github.com/pytorch/pytorch/pull/184975,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers,review diff and decide acceptance path,2026-07-08T01:53:49Z
pr,184976,Fix FakeTensor Module._apply conversion with weakrefs,https://github.com/pytorch/pytorch/pull/184976,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers,review diff and decide acceptance path,2026-07-08T01:51:21Z
pr,184978,Improve export error for untraceable lifted constants,https://github.com/pytorch/pytorch/pull/184978,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers,review diff and decide acceptance path,2026-07-08T01:31:12Z
pr,184981,Fix torch._numpy bool numeric subtract,https://github.com/pytorch/pytorch/pull/184981,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,review diff and decide acceptance path,2026-07-08T01:27:26Z
pr,184983,Make SDPA backend checks safe for unbacked head dims,https://github.com/pytorch/pytorch/pull/184983,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers,review diff and decide acceptance path,2026-07-08T01:23:08Z
pr,184987,Fix index_select refs bounds semantics,https://github.com/pytorch/pytorch/pull/184987,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,review diff and decide acceptance path,2026-07-08T01:18:24Z
pr,184995,Fix AOT getitem replay under inference_mode,https://github.com/pytorch/pytorch/pull/184995,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,review diff and decide acceptance path,2026-07-08T01:01:23Z
pr,185008,Add DebugMode tensor hash export,https://github.com/pytorch/pytorch/pull/185008,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers; CI lane labels: ciflow/dtensor,review diff and decide acceptance path,2026-07-08T00:53:37Z
pr,185009,Add fuzzy DebugMode hash mismatch comparison,https://github.com/pytorch/pytorch/pull/185009,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers; CI lane labels: ciflow/dtensor,review diff and decide acceptance path,2026-07-08T00:50:28Z
pr,184079,Enable graph deduplication for Inductor compile,https://github.com/pytorch/pytorch/pull/184079,ready for maintainer decision,67,"needs-review label present; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",review diff and decide acceptance path,2026-07-08T00:35:13Z
pr,185025,[export][serde] Initialize symbol ranges before constants,https://github.com/pytorch/pytorch/pull/185025,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,review diff and decide acceptance path,2026-07-08T00:32:29Z
pr,186953,Fix CPU amin/amax backward rematerialization mismatch,https://github.com/pytorch/pytorch/pull/186953,ready for maintainer decision,67,"needs-review label present; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",review diff and decide acceptance path,2026-07-08T00:16:47Z
pr,185050,Handle SymInt sizes in lazy module initialization,https://github.com/pytorch/pytorch/pull/185050,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,review diff and decide acceptance path,2026-07-08T00:03:22Z
pr,185051,Fix FX GraphModule serialization of string annotations,https://github.com/pytorch/pytorch/pull/185051,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers,review diff and decide acceptance path,2026-07-07T23:59:51Z
pr,185052,Fix Inductor reuse for disjoint strided slices,https://github.com/pytorch/pytorch/pull/185052,ready for maintainer decision,67,"needs-review label present; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",review diff and decide acceptance path,2026-07-07T23:56:48Z
pr,186879,Fix export dynamic shapes for variadic kwargs,https://github.com/pytorch/pytorch/pull/186879,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers,review diff and decide acceptance path,2026-07-07T23:46:33Z
pr,185105,Fix export move_to_device_pass device rewrites,https://github.com/pytorch/pytorch/pull/185105,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers,review diff and decide acceptance path,2026-07-07T23:28:41Z
pr,185106,Fix non-strict export state aliases,https://github.com/pytorch/pytorch/pull/185106,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,review diff and decide acceptance path,2026-07-07T23:24:33Z
pr,185121,Raise TypeError for invalid torch.export.load inputs,https://github.com/pytorch/pytorch/pull/185121,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers,review diff and decide acceptance path,2026-07-07T23:18:53Z
pr,185126,Optimize GB_REGISTRY linter scanning,https://github.com/pytorch/pytorch/pull/185126,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers,review diff and decide acceptance path,2026-07-07T22:44:44Z
pr,185132,Fix custom op autograd defaults with tensorlist outputs,https://github.com/pytorch/pytorch/pull/185132,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers,review diff and decide acceptance path,2026-07-07T22:41:33Z
pr,185140,Preserve reentrant checkpoint in non-strict export,https://github.com/pytorch/pytorch/pull/185140,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,review diff and decide acceptance path,2026-07-07T22:34:40Z
pr,185147,Fix Z3 translation validation for torch.sym_not,https://github.com/pytorch/pytorch/pull/185147,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,review diff and decide acceptance path,2026-07-07T22:32:09Z
pr,185155,Avoid warning spam for backend fallback graph breaks,https://github.com/pytorch/pytorch/pull/185155,ready for maintainer decision,67,"needs-review label present; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",review diff and decide acceptance path,2026-07-07T22:25:38Z
pr,185166,Fix export slice bounds across dynamic dim,https://github.com/pytorch/pytorch/pull/185166,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,review diff and decide acceptance path,2026-07-07T22:16:53Z
pr,185170,Fix ExportedProgram pickle after graph_module edits,https://github.com/pytorch/pytorch/pull/185170,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers,review diff and decide acceptance path,2026-07-07T22:13:22Z
pr,185191,Support wrapper subclass storages in HOP alias checks,https://github.com/pytorch/pytorch/pull/185191,ready for maintainer decision,67,"needs-review label present; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",review diff and decide acceptance path,2026-07-07T21:53:38Z
pr,185293,Require bool conditions for torch.where,https://github.com/pytorch/pytorch/pull/185293,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers; CI lane labels: ciflow/mps,review diff and decide acceptance path,2026-07-07T21:18:40Z
pr,186859,Recover fake/proxy views from functional tensor bases,https://github.com/pytorch/pytorch/pull/186859,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,review diff and decide acceptance path,2026-07-07T21:17:17Z
pr,184787,[export] Deserialize PT2 schema in C++ during load,https://github.com/pytorch/pytorch/pull/184787,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,review diff and decide acceptance path,2026-07-07T21:11:39Z
pr,185073,[export] Treat free new-tracer state as constants,https://github.com/pytorch/pytorch/pull/185073,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers,review diff and decide acceptance path,2026-07-07T20:37:31Z
pr,187012,Support optional Tensor list schemas,https://github.com/pytorch/pytorch/pull/187012,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers,review diff and decide acceptance path,2026-07-07T18:59:07Z
pr,185369,Fix dynamic output sizes for adaptive avg pool lowering,https://github.com/pytorch/pytorch/pull/185369,ready for maintainer decision,67,"needs-review label present; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",review diff and decide acceptance path,2026-07-07T18:01:43Z
pr,185404,[inductor] Fold small-row expanded bmm to mm in post-grad,https://github.com/pytorch/pytorch/pull/185404,ready for maintainer decision,67,"needs-review label present; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",review diff and decide acceptance path,2026-07-07T17:44:24Z
pr,186894,Fix non-strict export of vmap tensor indexing,https://github.com/pytorch/pytorch/pull/186894,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,review diff and decide acceptance path,2026-07-07T17:37:06Z
pr,184065,Fix static launcher 16-bit float scalars,https://github.com/pytorch/pytorch/pull/184065,ready for maintainer decision,67,"needs-review label present; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",review diff and decide acceptance path,2026-07-07T17:30:43Z
pr,185450,Fix reinplace alias checks for index_put,https://github.com/pytorch/pytorch/pull/185450,ready for maintainer decision,67,"needs-review label present; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",review diff and decide acceptance path,2026-07-07T17:16:47Z
pr,184083,Preserve user Triton allocator in Inductor,https://github.com/pytorch/pytorch/pull/184083,ready for maintainer decision,67,"needs-review label present; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",review diff and decide acceptance path,2026-07-07T16:45:45Z
pr,185483,Add fake support for pad_packed_sequence,https://github.com/pytorch/pytorch/pull/185483,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,review diff and decide acceptance path,2026-07-07T16:36:25Z
pr,185564,Handle pre-dispatch export of subclass tensor constants,https://github.com/pytorch/pytorch/pull/185564,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,review diff and decide acceptance path,2026-07-07T15:33:38Z
pr,187956,Propagate FX subgraph replacement meta values,https://github.com/pytorch/pytorch/pull/187956,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers,review diff and decide acceptance path,2026-07-07T15:27:38Z
pr,185569,Handle symbolic bool short-circuit with unbacked terms,https://github.com/pytorch/pytorch/pull/185569,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,review diff and decide acceptance path,2026-07-07T15:22:19Z
pr,185718,Fix fake functorch-wrapped tensor handling in jacfwd vmap,https://github.com/pytorch/pytorch/pull/185718,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,review diff and decide acceptance path,2026-07-07T14:44:05Z
pr,185755,Use gmem efficient attention kernel for long fp32 KV,https://github.com/pytorch/pytorch/pull/185755,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers,review diff and decide acceptance path,2026-07-07T14:12:52Z
pr,185772,Fix CPU small-output bilinear upsample precision in Inductor,https://github.com/pytorch/pytorch/pull/185772,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,review diff and decide acceptance path,2026-07-07T12:38:12Z
pr,185774,Fix non-strict export for Tensor.numpy,https://github.com/pytorch/pytorch/pull/185774,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,review diff and decide acceptance path,2026-07-07T12:30:33Z
pr,185408,Fix torch logging to avoid closed pytest stderr,https://github.com/pytorch/pytorch/pull/185408,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers,review diff and decide acceptance path,2026-07-07T12:19:00Z
pr,185526,Fix Dynamo tracing of LazyGraphModule,https://github.com/pytorch/pytorch/pull/185526,ready for maintainer decision,67,"needs-review label present; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",review diff and decide acceptance path,2026-07-07T12:16:19Z
pr,185786,Handle non-natural SymPy powers in value range interp,https://github.com/pytorch/pytorch/pull/185786,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers,review diff and decide acceptance path,2026-07-07T12:03:15Z
pr,185803,Fix joint export for tied parameter aliases,https://github.com/pytorch/pytorch/pull/185803,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers,review diff and decide acceptance path,2026-07-07T11:21:52Z
pr,185914,Log aot_export fw_metadata instead of printing it,https://github.com/pytorch/pytorch/pull/185914,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,review diff and decide acceptance path,2026-07-07T11:13:26Z
pr,185922,Fix fake one_hot invalid class count checks,https://github.com/pytorch/pytorch/pull/185922,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,review diff and decide acceptance path,2026-07-07T10:34:05Z
pr,184178,Fix masked Triton conv load pointers,https://github.com/pytorch/pytorch/pull/184178,ready for maintainer decision,67,"needs-review label present; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",review diff and decide acceptance path,2026-07-07T10:19:14Z
pr,186074,Fix multilabel_margin_loss decomposition sentinel handling,https://github.com/pytorch/pytorch/pull/186074,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,review diff and decide acceptance path,2026-07-07T07:11:13Z
pr,184207,Release max autotune benchmark artifacts,https://github.com/pytorch/pytorch/pull/184207,ready for maintainer decision,67,"needs-review label present; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",review diff and decide acceptance path,2026-07-07T06:52:34Z
pr,186111,Fix stale mutated input history in AOTAutograd,https://github.com/pytorch/pytorch/pull/186111,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,review diff and decide acceptance path,2026-07-07T06:50:00Z
pr,186118,Preserve guard eval defaulting without callback churn,https://github.com/pytorch/pytorch/pull/186118,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,review diff and decide acceptance path,2026-07-07T06:46:54Z
pr,186188,Fix FakeTensor wrapping for meta tensor constructor,https://github.com/pytorch/pytorch/pull/186188,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers,review diff and decide acceptance path,2026-07-07T06:14:15Z
pr,186194,Canonicalize trunc of symbolic integer division by one,https://github.com/pytorch/pytorch/pull/186194,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers,review diff and decide acceptance path,2026-07-07T05:55:45Z
pr,186199,Fix functionalization return mapping for mixed mutable out ops,https://github.com/pytorch/pytorch/pull/186199,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,review diff and decide acceptance path,2026-07-07T05:47:07Z
pr,186201,Deduplicate symbolic cond inputs after export passes,https://github.com/pytorch/pytorch/pull/186201,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,review diff and decide acceptance path,2026-07-07T05:44:48Z
pr,187604,Fix fake slice pending unbacked replacements,https://github.com/pytorch/pytorch/pull/187604,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers,review diff and decide acceptance path,2026-07-07T05:25:43Z
pr,186225,Preserve NJT nested ints in AOT subclass outputs,https://github.com/pytorch/pytorch/pull/186225,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,review diff and decide acceptance path,2026-07-07T05:20:37Z
pr,186261,Fix pre-dispatch tracing for saved tensor detach,https://github.com/pytorch/pytorch/pull/186261,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,review diff and decide acceptance path,2026-07-07T05:03:06Z
pr,186291,Fix export distribution validation assertions,https://github.com/pytorch/pytorch/pull/186291,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers,review diff and decide acceptance path,2026-07-07T04:31:47Z
pr,186315,Preserve extra autocast keys on FakeTensor,https://github.com/pytorch/pytorch/pull/186315,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,review diff and decide acceptance path,2026-07-06T18:18:35Z
pr,186321,Fix PythonMod axiom reuse after size refinement,https://github.com/pytorch/pytorch/pull/186321,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,review diff and decide acceptance path,2026-07-06T18:12:26Z
pr,186322,Add explicit schema registration for TorchBind methods,https://github.com/pytorch/pytorch/pull/186322,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers,review diff and decide acceptance path,2026-07-06T18:10:11Z
pr,186328,Preserve strides for functionalized in-place mutations,https://github.com/pytorch/pytorch/pull/186328,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers,review diff and decide acceptance path,2026-07-06T18:06:16Z
pr,186339,Improve torch.compile sibling output grad error,https://github.com/pytorch/pytorch/pull/186339,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,review diff and decide acceptance path,2026-07-06T16:55:55Z
pr,186342,Preserve nested fake subclass inputs in AOTAutograd,https://github.com/pytorch/pytorch/pull/186342,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,review diff and decide acceptance path,2026-07-06T16:48:47Z
pr,185079,Fix invalid Dynamo setattr exception timing,https://github.com/pytorch/pytorch/pull/185079,ready for maintainer decision,67,"needs-review label present; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",review diff and decide acceptance path,2026-07-03T03:44:16Z
pr,186527,Support export of packed LSTM with data-dependent lengths,https://github.com/pytorch/pytorch/pull/186527,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,review diff and decide acceptance path,2026-06-23T19:28:50Z
pr,185000,debug_mode: avoid hashing async collectives before wait,https://github.com/pytorch/pytorch/pull/185000,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers; CI lane labels: ciflow/dtensor,review diff and decide acceptance path,2026-06-11T03:18:46Z
issue,189322,DISABLED test_aoti_debug_printer_codegen_cuda (__main__.AOTInductorTestDualWrapper),https://github.com/pytorch/pytorch/issues/189322,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: flaky-tests, module: inductor, oncall: pt2",categorize and decide next maintainer owner/action,2026-07-09T01:17:51Z
issue,189323,DISABLED test_fake_export_contiguous_cpu_float32 (__main__.TestExportOpInfoCPU),https://github.com/pytorch/pytorch/issues/189323,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: flaky-tests, oncall: export, oncall: pt2",categorize and decide next maintainer owner/action,2026-07-09T01:17:16Z
issue,119537,DISABLED test_source_multithreaded_open_in_scope_work_in_main_thread_False (__main__.TestProfiler),https://github.com/pytorch/pytorch/issues/119537,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: flaky-tests, oncall: profiler",categorize and decide next maintainer owner/action,2026-07-09T01:17:04Z
issue,119576,DISABLED test_source_multithreaded_multiple_preexisting_work_in_main_thread_True (__main__.TestProfiler),https://github.com/pytorch/pytorch/issues/119576,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: flaky-tests, oncall: profiler",categorize and decide next maintainer owner/action,2026-07-09T01:17:01Z
issue,189308,`torch.distributed.checkpoint`: `.metadata` loaded with raw `pickle.load` on the public `load()` path — same class you already warn about on `torch.load`,https://github.com/pytorch/pytorch/issues/189308,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: oncall: distributed, oncall: distributed checkpointing",categorize and decide next maintainer owner/action,2026-07-08T21:48:14Z
issue,189309,[Stable ABI] Add a has_side_effects op tag. DCE silently deletes side-effect-only op calls,https://github.com/pytorch/pytorch/issues/189309,needs triage,65,PyTorch issue lacks triaged label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2026-07-08T21:45:56Z
issue,189298,torch.jit.trace fails with MaxUnpool in 2.13,https://github.com/pytorch/pytorch/issues/189298,needs triage,65,PyTorch issue lacks triaged label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2026-07-08T19:55:36Z
issue,189293,AttributeError: Can't get local object 'WeakValueDictionary.__init__.<locals>.remove' during AOTAutograd graph caching on Python 3.13,https://github.com/pytorch/pytorch/issues/189293,needs triage,65,"triage signal present; component/oncall labels: module: aotdispatch, module: pickle, module: pt2-dispatcher, module: python version",categorize and decide next maintainer owner/action,2026-07-08T19:26:04Z
issue,119526,DISABLED test_source_multithreaded_multiple_preexisting_work_in_main_thread_False (__main__.TestProfiler),https://github.com/pytorch/pytorch/issues/119526,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: flaky-tests, oncall: profiler",categorize and decide next maintainer owner/action,2026-07-08T19:02:44Z
issue,189280,[CPU][quant] operator_benchmark: qcat ~3-4x slower (likely) and fake_quantize ~700x slower (possible) since ~Nov 2025,https://github.com/pytorch/pytorch/issues/189280,needs triage,65,PyTorch issue lacks triaged label; component/oncall labels: oncall: quantization,categorize and decide next maintainer owner/action,2026-07-08T17:31:39Z
issue,188833,[pipelining] Interleaved1F1B warmup head-of-line bubble from front-loaded RECV,https://github.com/pytorch/pytorch/issues/188833,needs triage,65,PyTorch issue lacks triaged label; component/oncall labels: oncall: distributed,categorize and decide next maintainer owner/action,2026-07-08T17:31:15Z
issue,188836,DISABLED test_dropout_parity_dynamic_shapes (__main__.TestDropoutAlignRandomEager),https://github.com/pytorch/pytorch/issues/188836,needs triage,65,PyTorch issue lacks triaged label; component/oncall labels: oncall: pt2,categorize and decide next maintainer owner/action,2026-07-08T17:17:44Z
issue,188896,DISABLED test_launch_elastic_worker_raise_exception (__main__.ElasticLaunchTest),https://github.com/pytorch/pytorch/issues/188896,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: flaky-tests, oncall: distributed, oncall: r2p",categorize and decide next maintainer owner/action,2026-07-08T17:16:02Z
issue,189245,[cpp_extension] Stale .lock file from FileBaton causes permanent deadlock when compilation process is killed,https://github.com/pytorch/pytorch/issues/189245,needs triage,65,"triage signal present; component/oncall labels: module: cpp-extensions, module: deadlock",categorize and decide next maintainer owner/action,2026-07-08T16:35:40Z
issue,119536,DISABLED test_source_multithreaded_complex_work_in_main_thread_True (__main__.TestProfiler),https://github.com/pytorch/pytorch/issues/119536,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: flaky-tests, oncall: profiler",categorize and decide next maintainer owner/action,2026-07-08T13:12:38Z
issue,119490,DISABLED test_source_multithreaded_complex_work_in_main_thread_False (__main__.TestProfiler),https://github.com/pytorch/pytorch/issues/119490,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: flaky-tests, oncall: profiler",categorize and decide next maintainer owner/action,2026-07-08T07:30:07Z
issue,189228,[symm_mem] one/two-shot all-reduce deadlocks when issued from multiple streams: ops lack channel plumbing,https://github.com/pytorch/pytorch/issues/189228,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: oncall: distributed, oncall: distributed infra",categorize and decide next maintainer owner/action,2026-07-08T03:18:28Z
issue,119514,DISABLED test_source_multithreaded_close_in_scope_work_in_main_thread_True (__main__.TestProfiler),https://github.com/pytorch/pytorch/issues/119514,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: flaky-tests, oncall: profiler",categorize and decide next maintainer owner/action,2026-07-08T01:09:15Z
issue,119364,DISABLED test_source_multithreaded_close_in_scope_work_in_main_thread_False (__main__.TestProfiler),https://github.com/pytorch/pytorch/issues/119364,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: flaky-tests, oncall: profiler",categorize and decide next maintainer owner/action,2026-07-07T13:25:19Z
issue,75171,torch.jit.load fails when path contains non-ascii characters,https://github.com/pytorch/pytorch/issues/75171,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: windows, oncall: jit",categorize and decide next maintainer owner/action,2026-07-07T09:23:51Z
issue,119357,DISABLED test_source_multithreaded_basic_work_in_main_thread_True (__main__.TestProfiler),https://github.com/pytorch/pytorch/issues/119357,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: flaky-tests, oncall: profiler",categorize and decide next maintainer owner/action,2026-07-07T07:52:38Z
issue,119346,DISABLED test_source_multithreaded_basic_work_in_main_thread_False (__main__.TestProfiler),https://github.com/pytorch/pytorch/issues/119346,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: flaky-tests, oncall: profiler",categorize and decide next maintainer owner/action,2026-07-07T01:15:49Z
issue,189098,[RFC] Expose Kineto custom logger registration and protocol-based export to PyTorch users,https://github.com/pytorch/pytorch/issues/189098,needs triage,65,PyTorch issue lacks triaged label; component/oncall labels: oncall: profiler,categorize and decide next maintainer owner/action,2026-07-06T21:38:18Z
issue,188553,test_cholesky_solve_batched_many_batches_cuda_float64 fails on ROCm gfx942 (Tensor-likes are not close),https://github.com/pytorch/pytorch/issues/188553,needs triage,65,"triage signal present; component/oncall labels: module: linear algebra, module: rocm",categorize and decide next maintainer owner/action,2026-07-06T17:17:16Z
issue,144683,torch.export treats two of the same parameters as the same node,https://github.com/pytorch/pytorch/issues/144683,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: oncall: export, oncall: pt2",categorize and decide next maintainer owner/action,2026-07-06T04:36:19Z
issue,188567,"torch.distributed._symmetric_memory: NCCL-backend rendezvous() fails (""Failed to window register / invalid argument"") when it is the first NCCL op on the process group",https://github.com/pytorch/pytorch/issues/188567,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: oncall: distributed, oncall: distributed infra",categorize and decide next maintainer owner/action,2026-06-30T18:41:40Z
issue,188490,test_profiler_stack_trace_augmentation fails on Windows due to stale profiler golden,https://github.com/pytorch/pytorch/issues/188490,needs triage,65,PyTorch issue lacks triaged label; component/oncall labels: oncall: profiler,categorize and decide next maintainer owner/action,2026-06-29T22:35:00Z
issue,188481,export_chrome_trace produces invalid timestamps,https://github.com/pytorch/pytorch/issues/188481,needs triage,65,PyTorch issue lacks triaged label; component/oncall labels: oncall: profiler,categorize and decide next maintainer owner/action,2026-06-29T21:05:13Z
issue,188227,[inductor] generated kernel calls `empty_strided_cpu` with a negative stride → `Expected result >= 0`,https://github.com/pytorch/pytorch/issues/188227,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: oncall: cpu inductor, oncall: pt2",categorize and decide next maintainer owner/action,2026-06-28T13:48:56Z
issue,172568,Control flow operators issue tracker,https://github.com/pytorch/pytorch/issues/172568,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: higher order operators, module: pt2-dispatcher, oncall: export, oncall: pt2",categorize and decide next maintainer owner/action,2026-06-27T08:39:48Z
issue,188304,Update NVSHMEM to Resolve RoCE Failures,https://github.com/pytorch/pytorch/issues/188304,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: oncall: distributed, oncall: distributed infra",categorize and decide next maintainer owner/action,2026-06-26T22:26:32Z
issue,188266,SIGABRT inside `gloo::transport::tcp::Address::fromPeerName(int)`,https://github.com/pytorch/pytorch/issues/188266,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: oncall: distributed, oncall: distributed infra",categorize and decide next maintainer owner/action,2026-06-26T10:23:22Z
issue,185268,DISABLED test_correctness_with_reusing_ir_ne_cpu_float32 (__main__.TestLazyOpInfoCPU),https://github.com/pytorch/pytorch/issues/185268,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: flaky-tests, oncall: jit",categorize and decide next maintainer owner/action,2026-06-26T01:29:50Z
issue,185267,DISABLED test_correctness_ne_cpu_float32 (__main__.TestLazyOpInfoCPU),https://github.com/pytorch/pytorch/issues/185267,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: flaky-tests, oncall: jit",categorize and decide next maintainer owner/action,2026-06-26T01:29:49Z
issue,188133,Pallas CPU mishandles synthetic-base mutated aliased inputs,https://github.com/pytorch/pytorch/issues/188133,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: inductor, oncall: cpu inductor, oncall: pt2",categorize and decide next maintainer owner/action,2026-06-25T05:43:56Z
issue,181086,[Symmetric Memory] symm buffer init and rendezvous cannot be assigned to a specific CUDA stream,https://github.com/pytorch/pytorch/issues/181086,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: oncall: distributed, oncall: distributed infra",categorize and decide next maintainer owner/action,2026-06-19T11:10:46Z
issue,52741,torch.jit.script is swallowing a rfloordiv warning,https://github.com/pytorch/pytorch/issues/52741,needs triage,65,PyTorch issue lacks triaged label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2026-06-19T10:24:32Z
issue,182373,"[ROCm] torch.profiler drops GPU activity in spawn child if parent calls torch.cuda.init() (regression in 2.12, 2.10/2.11 pass)",https://github.com/pytorch/pytorch/issues/182373,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: rocm, oncall: profiler",categorize and decide next maintainer owner/action,2026-06-18T16:12:02Z
issue,187659,Segmentation fault in `TestQuantizedOps.test_empty_batch` on ARM,https://github.com/pytorch/pytorch/issues/187659,needs triage,65,PyTorch issue lacks triaged label; component/oncall labels: oncall: quantization,categorize and decide next maintainer owner/action,2026-06-18T14:07:03Z
issue,177254,AArch64 Unit Test Failure - test_aot_inductor_arrayref - RuntimeError: unknown architecure,https://github.com/pytorch/pytorch/issues/177254,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: arm, oncall: quantization",categorize and decide next maintainer owner/action,2026-06-18T10:11:47Z
issue,186363,test_quantization.py int8 tests fail on AArch64,https://github.com/pytorch/pytorch/issues/186363,needs triage,65,PyTorch issue lacks triaged label; component/oncall labels: oncall: quantization,categorize and decide next maintainer owner/action,2026-06-18T10:10:52Z
issue,187432,TypeError: can only join an iterable within `default_planner.py`,https://github.com/pytorch/pytorch/issues/187432,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: oncall: distributed, oncall: distributed checkpointing",categorize and decide next maintainer owner/action,2026-06-16T11:51:08Z
issue,186591,Importing Pytorch results in undefined symbol ncclCommResume,https://github.com/pytorch/pytorch/issues/186591,needs triage,65,"triage signal present; component/oncall labels: oncall: distributed, oncall: distributed infra",categorize and decide next maintainer owner/action,2026-06-16T05:35:09Z
issue,184147,[AMD] MoRI support for MoE EP Combine/Dispatch torch.distributed.TokenSwitch Interface & AMD Pollara Support,https://github.com/pytorch/pytorch/issues/184147,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: oncall: distributed, oncall: distributed infra",categorize and decide next maintainer owner/action,2026-06-15T23:24:39Z
issue,187094,torch._int_mm dispatches Ampere-era cutlass_80 kernels on sm_103 (GB300): 6.4-13.4x SLOWER than the fp32 tf32 matmul of the same shape,https://github.com/pytorch/pytorch/issues/187094,needs triage,65,PyTorch issue lacks triaged label; component/oncall labels: oncall: quantization,categorize and decide next maintainer owner/action,2026-06-14T13:45:25Z
issue,173815,[RFC] Abort a DeviceMesh,https://github.com/pytorch/pytorch/issues/173815,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: devicemesh, oncall: distributed, oncall: distributed infra",categorize and decide next maintainer owner/action,2026-06-13T16:39:14Z
issue,176439,[wip][RFC] Symmetric Memory pool explicit lending semantics,https://github.com/pytorch/pytorch/issues/176439,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: symm_mem, oncall: distributed, oncall: distributed infra",categorize and decide next maintainer owner/action,2026-06-13T16:39:11Z
issue,167966,Symmtric_memory: torch.ops.symm_mem.all_to_all_vdev_2d restricts the number of experts per rank (ne) to less than 16,https://github.com/pytorch/pytorch/issues/167966,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: symm_mem, oncall: distributed, oncall: distributed infra",categorize and decide next maintainer owner/action,2026-06-13T16:38:39Z
issue,172873,Need for all_reduce to initialize NCCL communicator,https://github.com/pytorch/pytorch/issues/172873,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: nccl, module: symm_mem, oncall: distributed, oncall: distributed infra",categorize and decide next maintainer owner/action,2026-06-13T16:38:22Z
issue,172617,Symmetric utils missing,https://github.com/pytorch/pytorch/issues/172617,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: symm_mem, oncall: distributed, oncall: distributed infra",categorize and decide next maintainer owner/action,2026-06-13T16:38:18Z
issue,167975,"Multi-node distributed training hangs indefinitely after NCCL initialization completes, specifically blocking at `verify_param_shape_across_processes` in `torch.distributed.utils`.",https://github.com/pytorch/pytorch/issues/167975,needs triage,65,"triage signal present; component/oncall labels: module: nccl, oncall: distributed, oncall: distributed infra",categorize and decide next maintainer owner/action,2026-06-13T16:34:44Z
issue,172348,Build failure in NCCLSymmetricMemory.cu: undefined identifier `offset_`,https://github.com/pytorch/pytorch/issues/172348,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: build, module: nccl, oncall: distributed, oncall: distributed infra",categorize and decide next maintainer owner/action,2026-06-12T17:12:35Z
issue,170707,Update Rendezvous etcd backend to use etcd v3 API,https://github.com/pytorch/pytorch/issues/170707,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: c10d, module: elastic, oncall: distributed, oncall: distributed infra",categorize and decide next maintainer owner/action,2026-06-12T17:11:17Z
issue,172218,`_expand_group` can get DeviceMesh ranks more directly,https://github.com/pytorch/pytorch/issues/172218,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: devicemesh, oncall: distributed, oncall: distributed infra",categorize and decide next maintainer owner/action,2026-06-12T17:11:16Z
issue,170045,ncclUnhandledCudaError: Call to CUDA function failed. Last error: Cuda failure 999 'unknown error',https://github.com/pytorch/pytorch/issues/170045,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: nccl, oncall: distributed, oncall: distributed infra",categorize and decide next maintainer owner/action,2026-06-12T17:10:48Z
issue,169448,FSDP2 model saving broken unless ignoring frozen params,https://github.com/pytorch/pytorch/issues/169448,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: fsdp, oncall: distributed, oncall: distributed checkpointing",categorize and decide next maintainer owner/action,2026-06-12T17:09:50Z
issue,161147,RuntimeError in DTensor evenly shard view op,https://github.com/pytorch/pytorch/issues/161147,needs triage,65,"triage signal present; component/oncall labels: module: dtensor, oncall: distributed, oncall: distributed parallelisms",categorize and decide next maintainer owner/action,2026-06-12T17:04:15Z
issue,163193,Incorrect results when adding partial DTensor with a scalar,https://github.com/pytorch/pytorch/issues/163193,needs triage,65,"triage signal present; component/oncall labels: module: correctness (silent), module: dtensor, oncall: distributed, oncall: distributed parallelisms",categorize and decide next maintainer owner/action,2026-06-12T17:04:06Z
issue,186026,Annotate Cuda Graph kernel nodes from profiler with the originating fully qualified layer names,https://github.com/pytorch/pytorch/issues/186026,needs triage,65,PyTorch issue lacks triaged label; component/oncall labels: oncall: profiler,categorize and decide next maintainer owner/action,2026-06-11T21:36:50Z
issue,167304,RPC cannot run in jetson orin because of the specific uuid of orin,https://github.com/pytorch/pytorch/issues/167304,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: cuda, module: rpc, oncall: distributed, oncall: distributed infra",categorize and decide next maintainer owner/action,2026-06-11T17:24:35Z
issue,166577,Pipelining - cannot calculate dWeight if a custom torch.autograd.Function is used with a schedule that performs separate dWeight and dInput backward passes,https://github.com/pytorch/pytorch/issues/166577,needs triage,65,"triage signal present; component/oncall labels: module: autograd, module: pipelining, oncall: distributed, oncall: distributed parallelisms",categorize and decide next maintainer owner/action,2026-06-11T17:24:09Z
issue,155924,"Activation Checkpointing breaks ""torch.distributed.checkpoint.state_dict._get_fqns""",https://github.com/pytorch/pytorch/issues/155924,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: activation checkpointing, oncall: distributed, oncall: distributed checkpointing",categorize and decide next maintainer owner/action,2026-06-10T17:14:28Z
issue,186823,DISABLED test_cupti_monitor_trace_has_expected_events (__main__.TestProfilerCUDA),https://github.com/pytorch/pytorch/issues/186823,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: flaky-tests, oncall: profiler",categorize and decide next maintainer owner/action,2026-06-10T01:29:42Z
issue,182988,ROCm MI300 (gfx942): native segfault in mixed-precision BatchNorm2d backward,https://github.com/pytorch/pytorch/issues/182988,needs triage,65,"triage signal present; component/oncall labels: module: fsdp, module: rocm, oncall: distributed, oncall: distributed parallelisms",categorize and decide next maintainer owner/action,2026-06-09T17:29:37Z
issue,153779,NCCL timeout while training on multi-gpu,https://github.com/pytorch/pytorch/issues/153779,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: nccl, oncall: distributed, oncall: distributed infra",categorize and decide next maintainer owner/action,2026-06-09T17:04:13Z
issue,130251,[funcol] numerical error when funcol.reduce-scatter a non-contiguous tensor,https://github.com/pytorch/pytorch/issues/130251,needs triage,65,"triage signal present; component/oncall labels: module: c10d, oncall: distributed, oncall: distributed infra",categorize and decide next maintainer owner/action,2026-06-09T16:58:51Z
issue,140183,FSDP Hybrid Shard worse loss than than Full Shard,https://github.com/pytorch/pytorch/issues/140183,needs triage,65,"triage signal present; component/oncall labels: module: fsdp, oncall: distributed, oncall: distributed parallelisms",categorize and decide next maintainer owner/action,2026-06-09T16:58:45Z
issue,186773,DISABLED test_cupti_monitor_record_shapes (__main__.TestProfilerCUDA),https://github.com/pytorch/pytorch/issues/186773,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: flaky-tests, oncall: profiler",categorize and decide next maintainer owner/action,2026-06-09T13:33:41Z
issue,186670,_load_mobile_module_from_bytes SIGSEGVs on malformed flatbuffer model (unchecked Function.class_type index),https://github.com/pytorch/pytorch/issues/186670,needs triage,65,PyTorch issue lacks triaged label; component/oncall labels: oncall: mobile,categorize and decide next maintainer owner/action,2026-06-08T23:59:37Z
issue,186637,DISABLED test_cupti_monitor_multithread_runtime_thread_assignment (__main__.TestProfilerCUDA),https://github.com/pytorch/pytorch/issues/186637,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: flaky-tests, oncall: profiler",categorize and decide next maintainer owner/action,2026-06-08T19:28:23Z
issue,143574,Torch elastic restart fails with torch-2.6.0 nightly build: NCCL unhandled system error,https://github.com/pytorch/pytorch/issues/143574,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: nccl, oncall: distributed, oncall: distributed infra",categorize and decide next maintainer owner/action,2026-06-08T17:17:44Z
issue,137016,[RFC] Accelerate intra-node communication via NCCL buffer registration,https://github.com/pytorch/pytorch/issues/137016,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: nccl, oncall: distributed, oncall: distributed infra",categorize and decide next maintainer owner/action,2026-06-08T17:16:28Z
issue,143593,TCPStore crash when initializing from multiple threads,https://github.com/pytorch/pytorch/issues/143593,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: c10d, oncall: distributed, oncall: distributed infra",categorize and decide next maintainer owner/action,2026-06-08T17:16:05Z
issue,137007,[RFC] ProcessGroupNCCL uses non-blocking API by default,https://github.com/pytorch/pytorch/issues/137007,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: c10d, module: nccl, oncall: distributed, oncall: distributed infra",categorize and decide next maintainer owner/action,2026-06-08T17:14:29Z
issue,130113,ncclUnhandledCudaError: Call to CUDA function failed.,https://github.com/pytorch/pytorch/issues/130113,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: ddp, module: nccl, oncall: distributed, oncall: distributed infra",categorize and decide next maintainer owner/action,2026-06-08T17:14:11Z
issue,129341,NCCL Blocking Send/Recv are Non-blocking in practice,https://github.com/pytorch/pytorch/issues/129341,needs triage,65,"triage signal present; component/oncall labels: module: nccl, oncall: distributed, oncall: distributed infra",categorize and decide next maintainer owner/action,2026-06-08T17:14:02Z
issue,186587,DISABLED test_cupti_monitor_matches_stock_op_and_kernel_names (__main__.TestProfilerCUDA),https://github.com/pytorch/pytorch/issues/186587,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: flaky-tests, oncall: profiler",categorize and decide next maintainer owner/action,2026-06-08T14:01:08Z
issue,115388,`torch.distributed.destroy_process_group()` hangs after CUDA graph capture of NCCL operations,https://github.com/pytorch/pytorch/issues/115388,needs triage,65,"triage signal present; component/oncall labels: module: c10d, module: cuda graphs, oncall: distributed, oncall: distributed infra",categorize and decide next maintainer owner/action,2026-06-07T16:38:36Z
issue,114396,is PGNCCL abortCommsFromMap working correctly?,https://github.com/pytorch/pytorch/issues/114396,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: nccl, oncall: distributed, oncall: distributed infra",categorize and decide next maintainer owner/action,2026-06-07T16:37:33Z
issue,106633,Refcount problem for torch.distributed.Store objects defined in Python,https://github.com/pytorch/pytorch/issues/106633,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: c10d, oncall: distributed, oncall: distributed infra",categorize and decide next maintainer owner/action,2026-06-07T16:37:12Z
issue,106711,'CUDA out of memory' when using a GPU services for reinforcement learning in Torch rpc tutorial,https://github.com/pytorch/pytorch/issues/106711,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: rpc, oncall: distributed, oncall: distributed infra",categorize and decide next maintainer owner/action,2026-06-07T16:36:53Z
issue,109941,Need latest NCCL support to reduce GPU HBM consumption,https://github.com/pytorch/pytorch/issues/109941,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: nccl, oncall: distributed, oncall: distributed infra",categorize and decide next maintainer owner/action,2026-06-07T16:36:40Z
issue,85671,nn.Embedding weights are not synced across processes with DistributedDataParallel when other parameters are present,https://github.com/pytorch/pytorch/issues/85671,needs triage,65,"triage signal present; component/oncall labels: module: ddp, oncall: distributed, oncall: distributed parallelisms",categorize and decide next maintainer owner/action,2026-06-07T16:33:46Z
issue,102596,Jetson NX with torch 1.12.0 :cannot import name 'ProcessGroup' from 'torch.distributed'.,https://github.com/pytorch/pytorch/issues/102596,needs triage,65,"triage signal present; component/oncall labels: oncall: distributed, oncall: distributed infra",categorize and decide next maintainer owner/action,2026-06-06T16:36:35Z
issue,103221,[help] did torch.distributed.launch can be applied on k8s cluster with pytorch-operator,https://github.com/pytorch/pytorch/issues/103221,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: elastic, oncall: distributed, oncall: distributed infra",categorize and decide next maintainer owner/action,2026-06-06T16:36:24Z
issue,101962,SummaryWriter background thread holds the GIL for too long,https://github.com/pytorch/pytorch/issues/101962,needs triage,65,"triage signal present; component/oncall labels: module: tensorboard, oncall: distributed",categorize and decide next maintainer owner/action,2026-06-06T16:35:54Z
issue,100310,Using ddp training with different machine,https://github.com/pytorch/pytorch/issues/100310,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: elastic, oncall: distributed, oncall: distributed infra",categorize and decide next maintainer owner/action,2026-06-06T16:35:38Z
issue,91173,not able to import pipelines as torch.distributed is missing,https://github.com/pytorch/pytorch/issues/91173,needs triage,65,triage signal present; component/oncall labels: oncall: distributed,categorize and decide next maintainer owner/action,2026-06-06T16:35:12Z
issue,86962,PyTorch RPC crashed when using IB,https://github.com/pytorch/pytorch/issues/86962,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: rpc, oncall: distributed, oncall: distributed infra",categorize and decide next maintainer owner/action,2026-06-06T16:35:02Z
issue,86849,`torch.distributed.all_reduce` allocates excess GPU memory when using NCCL backend,https://github.com/pytorch/pytorch/issues/86849,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: nccl, oncall: distributed, oncall: distributed infra",categorize and decide next maintainer owner/action,2026-06-06T16:34:44Z
issue,74590,[CTA] Let's Stamp Out Flaky Tests!,https://github.com/pytorch/pytorch/issues/74590,needs triage,65,"triage signal present; component/oncall labels: module: autograd, module: cuda, module: data parallel, module: dataloader",categorize and decide next maintainer owner/action,2026-06-06T16:34:00Z
issue,83824,RuntimeError: Interrupted system call when doing distributed training,https://github.com/pytorch/pytorch/issues/83824,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: c10d, oncall: distributed, oncall: distributed infra",categorize and decide next maintainer owner/action,2026-06-06T16:33:27Z
issue,85234,c10d all_gather aborts with Signal 8 (SIGFPE) when tensor.numel() == 0,https://github.com/pytorch/pytorch/issues/85234,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: c10d, oncall: distributed, oncall: distributed infra",categorize and decide next maintainer owner/action,2026-06-06T16:33:19Z
issue,82306,when distribute training load pretrain model error,https://github.com/pytorch/pytorch/issues/82306,needs triage,65,"triage signal present; component/oncall labels: module: serialization, oncall: distributed",categorize and decide next maintainer owner/action,2026-06-06T16:33:02Z
issue,76039,[RFC] NPU device for PyTorch,https://github.com/pytorch/pytorch/issues/76039,needs triage,65,triage signal present; component/oncall labels: oncall: distributed,categorize and decide next maintainer owner/action,2026-06-06T16:32:36Z
issue,71049,Distributed broadcast fails with simple GPU tensor on Windows + GLOO,https://github.com/pytorch/pytorch/issues/71049,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: c10d, oncall: distributed, oncall: distributed infra",categorize and decide next maintainer owner/action,2026-06-06T16:32:08Z
issue,73542,Investigate why tensor shapes are not populated when printing WorkNCCL for broadcast,https://github.com/pytorch/pytorch/issues/73542,needs triage,65,"triage signal present; component/oncall labels: module: c10d, module: ddp, oncall: distributed, oncall: distributed infra",categorize and decide next maintainer owner/action,2026-06-06T16:29:40Z
issue,74809,Incorrect results for `torch.distributed.gather` for tensor created from permuted NumPy array,https://github.com/pytorch/pytorch/issues/74809,needs triage,65,"triage signal present; component/oncall labels: module: c10d, oncall: distributed, oncall: distributed infra",categorize and decide next maintainer owner/action,2026-06-06T16:29:06Z
issue,76282,`torch.cuda.amp.GradScaler` may skip parameter synchronization required by post localSGD optimizer,https://github.com/pytorch/pytorch/issues/76282,needs triage,65,"triage signal present; component/oncall labels: module: amp (automated mixed precision), oncall: distributed, oncall: distributed parallelisms",categorize and decide next maintainer owner/action,2026-06-06T16:28:23Z
issue,186400,DISABLED test_cupti_monitor_collection_repeated_lifecycle (__main__.TestProfilerCUDA),https://github.com/pytorch/pytorch/issues/186400,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: flaky-tests, oncall: profiler",categorize and decide next maintainer owner/action,2026-06-05T19:15:42Z
issue,82308,torch.einsum gets wrong results randomly when training with multi-gpu,https://github.com/pytorch/pytorch/issues/82308,needs triage,65,"triage signal present; component/oncall labels: module: ddp, oncall: distributed, oncall: distributed parallelisms",categorize and decide next maintainer owner/action,2026-06-05T17:11:22Z
issue,82464,SyncBatchNorm does not work on CPU,https://github.com/pytorch/pytorch/issues/82464,needs triage,65,"triage signal present; component/oncall labels: module: nn, oncall: distributed, oncall: distributed parallelisms",categorize and decide next maintainer owner/action,2026-06-05T17:11:17Z
issue,87995,multi-node distributed training rank0 hang at dataloader after a few epochs,https://github.com/pytorch/pytorch/issues/87995,needs triage,65,"triage signal present; component/oncall labels: module: dataloader, oncall: distributed, oncall: distributed parallelisms",categorize and decide next maintainer owner/action,2026-06-05T17:11:01Z
issue,91535,"When dist.broadcast float32 to int64, it will silently generate wrong results",https://github.com/pytorch/pytorch/issues/91535,needs triage,65,"triage signal present; component/oncall labels: oncall: distributed, oncall: distributed infra",categorize and decide next maintainer owner/action,2026-06-05T17:10:02Z
issue,102186,torch/distributed/_spmd/api.py should aot_module_export instead of make_fx directly,https://github.com/pytorch/pytorch/issues/102186,needs triage,65,"triage signal present; component/oncall labels: oncall: distributed, oncall: distributed parallelisms",categorize and decide next maintainer owner/action,2026-06-05T17:09:11Z
issue,106690,DDP grads not synced when static_graph=True and module output is a dict subclass?,https://github.com/pytorch/pytorch/issues/106690,needs triage,65,"triage signal present; component/oncall labels: module: ddp, module: pytree, oncall: distributed, oncall: distributed parallelisms",categorize and decide next maintainer owner/action,2026-06-05T17:08:16Z
issue,111138,Module states cannot be fully synchronized due to the DDP broadcast_buffers breaking change,https://github.com/pytorch/pytorch/issues/111138,needs triage,65,"triage signal present; component/oncall labels: module: ddp, oncall: distributed, oncall: distributed parallelisms",categorize and decide next maintainer owner/action,2026-06-05T17:07:32Z
issue,112152,DDP backpropagated gradients not the same across all gpus when forward inference not using all declared modules,https://github.com/pytorch/pytorch/issues/112152,needs triage,65,"triage signal present; component/oncall labels: module: ddp, oncall: distributed, oncall: distributed parallelisms",categorize and decide next maintainer owner/action,2026-06-05T17:07:07Z
issue,123983,Segfault in TCPStore and FileStore compare_set(),https://github.com/pytorch/pytorch/issues/123983,needs triage,65,"triage signal present; component/oncall labels: oncall: distributed, oncall: distributed infra",categorize and decide next maintainer owner/action,2026-06-05T17:04:50Z
issue,124666,"When using DDP, model outputs that are are custom dataclasses would make things silently wrong",https://github.com/pytorch/pytorch/issues/124666,needs triage,65,"triage signal present; component/oncall labels: module: ddp, oncall: distributed, oncall: distributed parallelisms",categorize and decide next maintainer owner/action,2026-06-05T17:04:00Z
issue,130345,Provide shutdown() method for ProcessGroupGloo distributed backend,https://github.com/pytorch/pytorch/issues/130345,needs triage,65,"triage signal present; component/oncall labels: oncall: distributed, oncall: distributed infra",categorize and decide next maintainer owner/action,2026-06-05T17:03:45Z
issue,130515,[Bug Report] inconsistent global src and local rank in dist.nn.functional.broadcast,https://github.com/pytorch/pytorch/issues/130515,needs triage,65,"triage signal present; component/oncall labels: oncall: distributed, oncall: distributed infra",categorize and decide next maintainer owner/action,2026-06-05T17:03:40Z
issue,130958,Document ProcessGroupOptions,https://github.com/pytorch/pytorch/issues/130958,needs triage,65,"triage signal present; component/oncall labels: module: docs, oncall: distributed, oncall: distributed infra",categorize and decide next maintainer owner/action,2026-06-05T17:03:35Z
issue,143536,FSDP learning hangs when the program tries to save the model,https://github.com/pytorch/pytorch/issues/143536,needs triage,65,"triage signal present; component/oncall labels: module: fsdp, oncall: distributed, oncall: distributed parallelisms",categorize and decide next maintainer owner/action,2026-06-05T17:01:25Z
issue,144359,Incorrect Results with Tensor Parallelism,https://github.com/pytorch/pytorch/issues/144359,needs triage,65,"triage signal present; component/oncall labels: module: dtensor, oncall: distributed, oncall: distributed parallelisms",categorize and decide next maintainer owner/action,2026-06-05T17:01:16Z
issue,146331,DeepSeek: hierarchical a2a,https://github.com/pytorch/pytorch/issues/146331,needs triage,65,"triage signal present; component/oncall labels: oncall: distributed, oncall: distributed parallelisms",categorize and decide next maintainer owner/action,2026-06-05T17:00:33Z
issue,146907,"INTERNAL ASSERT FAILED at ""/pytorch/torch/csrc/distributed/c10d/reducer.cpp"":2134, please report a bug to PyTorch.",https://github.com/pytorch/pytorch/issues/146907,needs triage,65,"triage signal present; component/oncall labels: module: ddp, oncall: distributed, oncall: distributed parallelisms",categorize and decide next maintainer owner/action,2026-06-05T17:00:09Z
issue,165727,DDP hangs at first AllGather during parameter broadcast with tied weights on Blackwell GPUs (sm_120),https://github.com/pytorch/pytorch/issues/165727,needs triage,65,"triage signal present; component/oncall labels: module: ddp, module: nccl, oncall: distributed, oncall: distributed infra",categorize and decide next maintainer owner/action,2026-06-05T16:54:20Z
issue,166301,Improve Docs: Reproducibility and Randomness in dist training with multi-process data loading and explicit RNG seed,https://github.com/pytorch/pytorch/issues/166301,needs triage,65,"triage signal present; component/oncall labels: oncall: distributed, oncall: distributed parallelisms",categorize and decide next maintainer owner/action,2026-06-05T16:54:07Z
issue,169970,Does torch._grouped.mm work with cudagraphs over multiple nodes?,https://github.com/pytorch/pytorch/issues/169970,needs triage,65,"triage signal present; component/oncall labels: module: cuda, module: cuda graphs, oncall: distributed, oncall: distributed parallelisms",categorize and decide next maintainer owner/action,2026-06-05T16:52:53Z
issue,186344,DISABLED test_cupti_monitor_collection_raw_dump_smoke (__main__.TestProfilerCUDA),https://github.com/pytorch/pytorch/issues/186344,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: flaky-tests, oncall: profiler",categorize and decide next maintainer owner/action,2026-06-05T13:39:39Z
issue,153705,Unable to export a model using scan with inplace modification,https://github.com/pytorch/pytorch/issues/153705,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: oncall: export, oncall: pt2",categorize and decide next maintainer owner/action,2026-06-04T20:08:38Z
issue,151654,[sparse][exported-graph] Wrong Exported Graph for the Sparse Tensor,https://github.com/pytorch/pytorch/issues/151654,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: oncall: export, oncall: pt2",categorize and decide next maintainer owner/action,2026-06-04T20:08:31Z
issue,185443,[JIT] Uncontrolled memory allocation in Pickle Unpickler opcode handlers (setInput / readBytes),https://github.com/pytorch/pytorch/issues/185443,needs triage,65,PyTorch issue lacks triaged label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2026-06-04T17:36:27Z
issue,184647,Replacing a submodule's `__call__` with `fx.Interpreter` on its exported `gm` causes the parent's output to contain `FakeTensor`,https://github.com/pytorch/pytorch/issues/184647,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: oncall: export, oncall: pt2",categorize and decide next maintainer owner/action,2026-06-04T00:50:16Z
issue,121471,[profiler]Not passed shapes for lists of tensors to kineto_result,https://github.com/pytorch/pytorch/issues/121471,needs triage,65,PyTorch issue lacks triaged label; component/oncall labels: oncall: profiler,categorize and decide next maintainer owner/action,2026-06-01T16:59:16Z
issue,181621,torch.compile (Inductor) generates invalid C++ kernel: tmp8 used outside its declared scope when torch.gather index is a self-referential expression,https://github.com/pytorch/pytorch/issues/181621,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: cpu, module: inductor, module: scatter & gather ops, oncall: cpu inductor",categorize and decide next maintainer owner/action,2026-06-01T03:58:14Z
issue,181638,torch.compile(backend='inductor') hangs indefinitely on LayerNorm → flatten → cumprod,https://github.com/pytorch/pytorch/issues/181638,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: oncall: cpu inductor, oncall: pt2",categorize and decide next maintainer owner/action,2026-06-01T03:56:17Z
issue,151525,[inductor] [cpu] [silent incorrectness] `nn.LazyConvTranspose2d-torch.randn-F.linear-torch.argmax` output incorrect results on CPU inductor,https://github.com/pytorch/pytorch/issues/151525,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: oncall: cpu inductor, oncall: pt2",categorize and decide next maintainer owner/action,2026-05-30T01:11:09Z
issue,151726,Exported Module cannot call train() or eval(),https://github.com/pytorch/pytorch/issues/151726,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: oncall: export, oncall: pt2",categorize and decide next maintainer owner/action,2026-05-29T16:37:47Z
issue,185442,[JIT] Use-after-free in Source::_M_dispose() during TorchScript error-path cleanup,https://github.com/pytorch/pytorch/issues/185442,needs triage,65,PyTorch issue lacks triaged label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2026-05-27T22:43:56Z
issue,185440,[JIT] Out-of-bounds read in MemoryReadAdapter::read() via crafted TorchScript model,https://github.com/pytorch/pytorch/issues/185440,needs triage,65,PyTorch issue lacks triaged label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2026-05-27T22:42:07Z
issue,184010,[Feature Request] Support PrivateUse1 device in ProcessGroupGloo,https://github.com/pytorch/pytorch/issues/184010,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: oncall: distributed, oncall: distributed infra",categorize and decide next maintainer owner/action,2026-05-27T20:29:52Z
issue,183913,[RFC] Unify Python dependency declarations under pyproject.toml + uv.lock,https://github.com/pytorch/pytorch/issues/183913,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: ci, oncall: releng",categorize and decide next maintainer owner/action,2026-05-27T14:33:16Z
issue,172561,[export] Constraint expression suggested by torch export is not supported,https://github.com/pytorch/pytorch/issues/172561,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: onnx, oncall: export, oncall: pt2",categorize and decide next maintainer owner/action,2026-05-25T20:30:42Z
issue,184982,"deprecation of derived quantized dtypes (quint8, qint8, qint32)",https://github.com/pytorch/pytorch/issues/184982,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: deprecation, oncall: quantization",categorize and decide next maintainer owner/action,2026-05-25T17:33:27Z
issue,152532,The 2.7.0 release tarball is missing `.ci/docker/ci_commit_pins/nccl-cu12.txt` required for building,https://github.com/pytorch/pytorch/issues/152532,needs triage,65,PyTorch issue lacks triaged label; component/oncall labels: oncall: package/deploy,categorize and decide next maintainer owner/action,2026-05-23T23:24:23Z
issue,181429,"Make torchrun options ( --nnodes, --nproc-per-node, etc.) same a mpirun/mpiexec",https://github.com/pytorch/pytorch/issues/181429,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: elastic, oncall: distributed, oncall: distributed infra",categorize and decide next maintainer owner/action,2026-05-23T16:28:35Z
issue,180550,[Discussion] Parallelizing serial for-loops with free-threaded Python — and the `torch.distributed` challenge,https://github.com/pytorch/pytorch/issues/180550,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: oncall: distributed, oncall: distributed infra",categorize and decide next maintainer owner/action,2026-05-23T16:27:54Z
issue,178138,Conflict between PyTorch CUDACachingAllocator and NCCL symmetric memory (ncclCommWindowRegister) causing incorrect all-reduce results,https://github.com/pytorch/pytorch/issues/178138,needs triage,65,"triage signal present; component/oncall labels: module: nccl, module: symm_mem, oncall: distributed, oncall: distributed infra",categorize and decide next maintainer owner/action,2026-05-23T16:24:32Z
issue,179848,Race condition updating pgStatus_ in ProcessGroupGloo::runLoop,https://github.com/pytorch/pytorch/issues/179848,needs triage,65,"triage signal present; component/oncall labels: module: c10d, oncall: distributed, oncall: distributed infra",categorize and decide next maintainer owner/action,2026-05-23T16:24:21Z
issue,179238,Shutdown race in Gloo backend,https://github.com/pytorch/pytorch/issues/179238,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: oncall: distributed, oncall: distributed infra",categorize and decide next maintainer owner/action,2026-05-22T16:51:58Z
issue,178756,Case mismatch in `_new_process_group_helper` prevents default backend type from being set for custom out-of-tree backends,https://github.com/pytorch/pytorch/issues/178756,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: oncall: distributed, oncall: distributed infra",categorize and decide next maintainer owner/action,2026-05-22T16:51:57Z
issue,178191,[distributed] ProcessGroupMPI: cudaAwareMpiCheck() does not support Cray MPICH GPU-aware MPI,https://github.com/pytorch/pytorch/issues/178191,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: oncall: distributed, oncall: distributed infra",categorize and decide next maintainer owner/action,2026-05-22T16:51:13Z
pr,180501,Fix clip_grad_norm_ failure with mixed DeviceMesh DTensor norms,https://github.com/pytorch/pytorch/pull/180501,PR blocked,65,failed checks or blocked/unstable merge state; PR closes or references open issue numbers; PR has not updated for at least 45 days; CI lane labels: ciflow/dtensor,"identify whether block is CI, merge conflict, or review gate",2026-05-22T14:26:14Z
pr,184587,Fix torch.compile addmm failure with beta=0 nonbroadcastable bias,https://github.com/pytorch/pytorch/pull/184587,PR blocked,65,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; PR has not updated for at least 45 days; CI lane labels: ciflow/b200, ciflow/h100, ciflow/inductor, ciflow/rocm-mi300","identify whether block is CI, merge conflict, or review gate",2026-05-22T07:46:30Z
issue,179770,[symm_mem] small-M FP8 fused_scaled_matmul_reduce_scatter is slower than unfused baseline on B300/SM100,https://github.com/pytorch/pytorch/issues/179770,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: oncall: distributed, oncall: distributed infra",categorize and decide next maintainer owner/action,2026-05-21T17:04:03Z
issue,177371,LocalTensorMode: all_to_all_single crashes with uneven split sizes,https://github.com/pytorch/pytorch/issues/177371,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: oncall: distributed, oncall: distributed infra",categorize and decide next maintainer owner/action,2026-05-21T17:00:50Z
issue,177815,[distributed][Dynamo] Add get_process_group_all_ranks() for SPMD compilation backends,https://github.com/pytorch/pytorch/issues/177815,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: c10d, module: devicemesh, oncall: distributed, oncall: distributed infra",categorize and decide next maintainer owner/action,2026-05-21T17:00:43Z
issue,177427,[RFC] Communication primitives for non-contiguous tensors,https://github.com/pytorch/pytorch/issues/177427,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: c10d, module: fsdp, module: nccl, module: pipelining",categorize and decide next maintainer owner/action,2026-05-21T17:00:43Z
issue,177309,[ROCm][NCCL watchdog] Cross-thread stream-capture mode restrictions in hipEventQuery/hipEventSynchronize cause false watchdog failures,https://github.com/pytorch/pytorch/issues/177309,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: rocm, oncall: distributed, oncall: distributed infra",categorize and decide next maintainer owner/action,2026-05-21T17:00:00Z
issue,177059,Way to have DTensor ops go through torch_function rather than go directly through torch_dispatch?,https://github.com/pytorch/pytorch/issues/177059,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: dtensor, oncall: distributed, oncall: distributed parallelisms",categorize and decide next maintainer owner/action,2026-05-21T16:59:36Z
issue,175145,[torch.disitributed] Persistent Deadlock when overlapping NCCL P2P and Collectives from separate Python Threads,https://github.com/pytorch/pytorch/issues/175145,needs triage,65,"triage signal present; component/oncall labels: oncall: distributed, oncall: distributed infra",categorize and decide next maintainer owner/action,2026-05-21T16:57:40Z
pr,183922,Unskip test_vector_norm on AArch64,https://github.com/pytorch/pytorch/pull/183922,PR blocked,65,failed checks or blocked/unstable merge state; PR closes or references open issue numbers; PR has not updated for at least 45 days; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-05-21T08:20:29Z
pr,181060,[fft] Validate _fft_c2r last_dim_size to fix heap-buffer overflow,https://github.com/pytorch/pytorch/pull/181060,PR blocked,65,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; PR has not updated for at least 45 days; CI lane labels: ciflow/mps, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-05-21T05:51:18Z
issue,175469,[DTensor] ExportedProgram.run_decompositions() fails with AssertionError: out is not NotImplemented,https://github.com/pytorch/pytorch/issues/175469,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: dtensor, oncall: distributed, oncall: distributed parallelisms, oncall: export",categorize and decide next maintainer owner/action,2026-05-20T17:18:22Z
issue,176877,[RFC] Adding Distributed Support to OpenReg,https://github.com/pytorch/pytorch/issues/176877,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: openreg, module: privateuse1, oncall: distributed, oncall: distributed infra",categorize and decide next maintainer owner/action,2026-05-20T17:15:23Z
issue,174669,`TypeError: cannot pickle code objects` with Python 3.13,https://github.com/pytorch/pytorch/issues/174669,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: oncall: distributed, oncall: distributed checkpointing",categorize and decide next maintainer owner/action,2026-05-20T17:15:07Z
issue,175724,[Distributed][NCCL] Request for Backport: SIGSEGV in HeartbeatMonitor due to thread-unsafe getenv/setenv race (#167523),https://github.com/pytorch/pytorch/issues/175724,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: oncall: distributed, oncall: distributed infra",categorize and decide next maintainer owner/action,2026-05-20T17:15:06Z
issue,176276,Path Traversal in torch.distributed.FileStore allows writing files outside intended directory,https://github.com/pytorch/pytorch/issues/176276,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: oncall: distributed, oncall: distributed infra",categorize and decide next maintainer owner/action,2026-05-20T17:14:44Z
issue,173680,Large CPU overhead with FSDP2,https://github.com/pytorch/pytorch/issues/173680,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: fsdp, module: performance, oncall: distributed, oncall: distributed parallelisms",categorize and decide next maintainer owner/action,2026-05-20T17:14:06Z
issue,174960,Feature Request: Activation offloading with async prefetch in FSDP,https://github.com/pytorch/pytorch/issues/174960,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: fsdp, oncall: distributed, oncall: distributed parallelisms",categorize and decide next maintainer owner/action,2026-05-20T17:13:21Z
issue,174276,[DTensor] Avoid forced redistribution in layer_norm strategy,https://github.com/pytorch/pytorch/issues/174276,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: dtensor, module: performance, oncall: distributed, oncall: distributed parallelisms",categorize and decide next maintainer owner/action,2026-05-20T17:12:45Z
pr,184151,[CUDA] Fix float32 std/var M2 overflow by using double accumulation,https://github.com/pytorch/pytorch/pull/184151,PR blocked,65,failed checks or blocked/unstable merge state; PR closes or references open issue numbers; PR has not updated for at least 45 days,"identify whether block is CI, merge conflict, or review gate",2026-05-20T15:57:32Z
pr,180602,Enable ruff rule PLW1510: require explicit check= in subprocess calls,https://github.com/pytorch/pytorch/pull/180602,PR blocked,65,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; PR has not updated for at least 45 days; CI lane labels: ciflow/dtensor, ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-05-20T15:55:18Z
issue,183888,torch.compile Inductor fails on unfold followed by in-place abs_,https://github.com/pytorch/pytorch/issues/183888,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: oncall: cpu inductor, oncall: pt2",categorize and decide next maintainer owner/action,2026-05-20T11:50:23Z
pr,182055,[Inductor] Fix flaky test_allow_reuse_disable_if_exceed_peak by checking reuse marker,https://github.com/pytorch/pytorch/pull/182055,PR blocked,65,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-05-20T07:44:18Z
issue,180995,ConvTranspose + BatchNorm fusion broken for `groups != 1`,https://github.com/pytorch/pytorch/issues/180995,needs triage,65,PyTorch issue lacks triaged label; component/oncall labels: oncall: quantization,categorize and decide next maintainer owner/action,2026-05-19T19:34:38Z
issue,182641,[v2.11.0] Cannot jvp through FSDP2,https://github.com/pytorch/pytorch/issues/182641,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: fsdp, oncall: distributed, oncall: distributed parallelisms",categorize and decide next maintainer owner/action,2026-05-19T17:21:39Z
issue,72482,Improve torch::deploy documentation,https://github.com/pytorch/pytorch/issues/72482,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: deploy, oncall: package/deploy",categorize and decide next maintainer owner/action,2026-05-19T15:26:27Z
issue,183702,`grouped fully_shard(list) fails when a grouped module is reused later in the same iteration`,https://github.com/pytorch/pytorch/issues/183702,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: fsdp, oncall: distributed, oncall: distributed parallelisms",categorize and decide next maintainer owner/action,2026-05-18T19:46:44Z
issue,184078,add_params metric_dict values not shown in tensorboard hparams tab,https://github.com/pytorch/pytorch/issues/184078,needs triage,65,PyTorch issue lacks triaged label; component/oncall labels: oncall: visualization,categorize and decide next maintainer owner/action,2026-05-17T07:20:32Z
pr,173459,Add dtype mismatch validation for mm/bmm in meta registrations (#171190),https://github.com/pytorch/pytorch/pull/173459,PR blocked,65,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/mps, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-05-16T08:53:35Z
pr,180118,[dynamo] Skip pad_packed_sequence during tracing,https://github.com/pytorch/pytorch/pull/180118,PR blocked,65,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-05-16T04:18:07Z
issue,183842,docs.pytorch.org/memory_viz has Unhandled version 5 error,https://github.com/pytorch/pytorch/issues/183842,needs triage,65,PyTorch issue lacks triaged label; component/oncall labels: oncall: profiler,categorize and decide next maintainer owner/action,2026-05-15T03:50:31Z
pr,183087,[profiler] Synchronize all active devices on profiler exit,https://github.com/pytorch/pytorch/pull/183087,PR blocked,65,failed checks or blocked/unstable merge state; PR closes or references open issue numbers; PR has not updated for at least 45 days,"identify whether block is CI, merge conflict, or review gate",2026-05-14T17:08:37Z
pr,182406,Fix misleading error when caching_allocator_alloc size overflows ssize_t,https://github.com/pytorch/pytorch/pull/182406,PR blocked,65,failed checks or blocked/unstable merge state; PR closes or references open issue numbers; PR has not updated for at least 45 days,"identify whether block is CI, merge conflict, or review gate",2026-05-13T12:47:59Z
pr,177511,[export] preserve original traceback when augmenting export error messages,https://github.com/pytorch/pytorch/pull/177511,PR blocked,65,failed checks or blocked/unstable merge state; PR closes or references open issue numbers; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2026-05-10T05:35:42Z
issue,182382,[RFC] [DTensor] Support for One-Sided Communication in DTensor,https://github.com/pytorch/pytorch/issues/182382,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: dtensor, module: symm_mem, oncall: distributed, oncall: distributed parallelisms",categorize and decide next maintainer owner/action,2026-05-08T00:09:11Z
issue,182661,[v2.11.0] Cannot jvp through DDP,https://github.com/pytorch/pytorch/issues/182661,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: ddp, oncall: distributed, oncall: distributed parallelisms",categorize and decide next maintainer owner/action,2026-05-07T17:12:08Z
issue,181181,FP8 Per-Head Quantized Scaled Dot Product Attention,https://github.com/pytorch/pytorch/issues/181181,needs triage,65,PyTorch issue lacks triaged label; component/oncall labels: oncall: quantization,categorize and decide next maintainer owner/action,2026-05-06T17:36:53Z
issue,182653,CPU FPE crash in `torch.nn.quantized.functional.conv2d` (qnnpack),https://github.com/pytorch/pytorch/issues/182653,needs triage,65,PyTorch issue lacks triaged label; component/oncall labels: oncall: quantization,categorize and decide next maintainer owner/action,2026-05-06T14:35:23Z
issue,125171,datettime.now() is not supported by Dynamo,https://github.com/pytorch/pytorch/issues/125171,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: dynamo, oncall: export",categorize and decide next maintainer owner/action,2026-05-01T23:42:47Z
issue,181865,CUDA trace callbacks leak across tests in a shared process,https://github.com/pytorch/pytorch/issues/181865,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: cuda, oncall: profiler",categorize and decide next maintainer owner/action,2026-04-30T17:35:39Z
issue,178044,Compilation fails with Apple Clang 17 due to ambiguous py::make_tuple return types,https://github.com/pytorch/pytorch/issues/178044,needs triage,65,PyTorch issue lacks triaged label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2026-04-29T23:50:06Z
issue,177258,AArch64 Unit Test Failure - test/jit/test_freezing.py TestFreezing.test_module_with_shared_type_instances,https://github.com/pytorch/pytorch/issues/177258,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: arm, module: tests, oncall: jit",categorize and decide next maintainer owner/action,2026-04-29T19:47:25Z
issue,173988,[DTensor] Implement decomposed sharding propagation for SDPA operations,https://github.com/pytorch/pytorch/issues/173988,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: dtensor, oncall: distributed",categorize and decide next maintainer owner/action,2026-04-29T16:37:26Z
issue,173990,[DTensor] Propagate 'grad_dtype' attribute in 'to_local' and 'from_local' operations,https://github.com/pytorch/pytorch/issues/173990,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: dtensor, oncall: distributed",categorize and decide next maintainer owner/action,2026-04-29T16:37:21Z
issue,174732,[DTensor][aot_eager] compiling sharded embedding with dynamic seqlen fails,https://github.com/pytorch/pytorch/issues/174732,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: dtensor, oncall: distributed",categorize and decide next maintainer owner/action,2026-04-29T16:37:15Z
issue,177959,"dist.new_group(..., use_local_synchronization=True) hangs with Gloo for overlapping groups, while NCCL succeeds on the same participant-consistent creation order",https://github.com/pytorch/pytorch/issues/177959,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: c10d, oncall: distributed",categorize and decide next maintainer owner/action,2026-04-29T16:36:56Z
issue,144663,Support loading and executing a ExportedProgram from torch.export in C++ environment,https://github.com/pytorch/pytorch/issues/144663,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: oncall: export, oncall: pt2",categorize and decide next maintainer owner/action,2026-04-28T18:44:32Z
issue,181473,`torch.jit.ScriptModule.register_full_backward_hook` raises RuntimeError due to `_is_full_backward_hook` attribute type cast failure,https://github.com/pytorch/pytorch/issues/181473,needs triage,65,PyTorch issue lacks triaged label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2026-04-27T01:57:51Z
issue,181480,DISABLED LiteInterpreterUpgraderTest.DivTensorV2,https://github.com/pytorch/pytorch/issues/181480,needs triage,65,PyTorch issue lacks triaged label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2026-04-25T05:23:14Z
issue,181231,`jit.script` sequence unpacking breaks when using square brackets,https://github.com/pytorch/pytorch/issues/181231,needs triage,65,PyTorch issue lacks triaged label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2026-04-23T08:57:25Z
issue,176288,memory profiler data dump streaming-friendly format,https://github.com/pytorch/pytorch/issues/176288,needs triage,65,PyTorch issue lacks triaged label; component/oncall labels: oncall: profiler,categorize and decide next maintainer owner/action,2026-04-22T02:57:05Z
issue,177239,[Question] JIT Path Deprecation — Maintenance Timeline and Removal Plans,https://github.com/pytorch/pytorch/issues/177239,needs triage,65,PyTorch issue lacks triaged label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2026-04-21T05:33:51Z
issue,180307,PackageImporter.load_pickle() race can silently deserialize the wrong tensor across concurrent loads,https://github.com/pytorch/pytorch/issues/180307,needs triage,65,PyTorch issue lacks triaged label; component/oncall labels: oncall: package/deploy,categorize and decide next maintainer owner/action,2026-04-14T02:17:25Z
issue,153949,Profiler does not capture steps on Blackwell,https://github.com/pytorch/pytorch/issues/153949,needs triage,65,PyTorch issue lacks triaged label; component/oncall labels: oncall: profiler,categorize and decide next maintainer owner/action,2026-04-13T17:23:00Z
issue,141668,[inductor][cpu] AMP CPP wrapper performance regression in 2024-11-25 nightly release,https://github.com/pytorch/pytorch/issues/141668,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: oncall: cpu inductor, oncall: pt2",categorize and decide next maintainer owner/action,2026-04-09T21:54:46Z
issue,167631,`jit.export` analoge for `torch.export`,https://github.com/pytorch/pytorch/issues/167631,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: oncall: export, oncall: pt2",categorize and decide next maintainer owner/action,2026-04-09T20:30:46Z
issue,179111,The smallest values of E4M3 for two-level quantization is incorrect,https://github.com/pytorch/pytorch/issues/179111,needs triage,65,PyTorch issue lacks triaged label; component/oncall labels: oncall: quantization,categorize and decide next maintainer owner/action,2026-04-09T11:46:04Z
issue,179176,Support unbacked in export Dim API.,https://github.com/pytorch/pytorch/issues/179176,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: oncall: export, oncall: pt2",categorize and decide next maintainer owner/action,2026-04-02T19:50:24Z
issue,179109,Request to support fp8 x4 packed dtypes in PT,https://github.com/pytorch/pytorch/issues/179109,needs triage,65,PyTorch issue lacks triaged label; component/oncall labels: oncall: quantization,categorize and decide next maintainer owner/action,2026-04-02T03:51:02Z
issue,177825,Internal Assert Failure in `torch.ao.nn.quantized.GroupNorm` with incorrectly sized bias tensor,https://github.com/pytorch/pytorch/issues/177825,needs triage,65,PyTorch issue lacks triaged label; component/oncall labels: oncall: quantization,categorize and decide next maintainer owner/action,2026-03-26T00:38:43Z
issue,177508,[inductor][cpu] torch_multimodal_clip AMP multiple thread performance regression in 2026-03-16 nightly release,https://github.com/pytorch/pytorch/issues/177508,needs triage,65,PyTorch issue lacks triaged label; component/oncall labels: oncall: cpu inductor,categorize and decide next maintainer owner/action,2026-03-24T06:31:55Z
issue,178087,"[Profiler] Expose moduleHierarchy, activityType, and other KinetoEvent methods to Python",https://github.com/pytorch/pytorch/issues/178087,needs triage,65,PyTorch issue lacks triaged label; component/oncall labels: oncall: profiler,categorize and decide next maintainer owner/action,2026-03-23T22:17:21Z
issue,177476,Crash in torch.ao.nn.quantized.functional.interpolate when input tensor has a 0-sized dimension,https://github.com/pytorch/pytorch/issues/177476,needs triage,65,PyTorch issue lacks triaged label; component/oncall labels: oncall: quantization,categorize and decide next maintainer owner/action,2026-03-16T15:47:31Z
issue,124502,[export] Llama3 export with dynamic shapes fails with constraint violations,https://github.com/pytorch/pytorch/issues/124502,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: dynamic shapes, oncall: export",categorize and decide next maintainer owner/action,2026-03-15T21:02:11Z
issue,177249,AArch64 Unit Test Failure - test_ops_jit.py test_variant_consistency_jit_linalg_lstsq_cpu_complex64,https://github.com/pytorch/pytorch/issues/177249,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: arm, oncall: jit",categorize and decide next maintainer owner/action,2026-03-12T11:35:05Z
issue,177247,AArch64 Unit Test Failure - test_jit_autocast,https://github.com/pytorch/pytorch/issues/177247,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: arm, oncall: jit",categorize and decide next maintainer owner/action,2026-03-12T11:32:47Z
issue,175708,O(N * B) loop in _C._jit_pass_lint (potentially O(N^2)),https://github.com/pytorch/pytorch/issues/175708,needs triage,65,PyTorch issue lacks triaged label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2026-03-11T15:46:38Z
issue,175293,TL-parse does not work for export.,https://github.com/pytorch/pytorch/issues/175293,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: oncall: export, oncall: pt2",categorize and decide next maintainer owner/action,2026-03-09T23:28:36Z
issue,176148,"export runtime error: Tensor specified that the operator does not mutate the argument, but this seems to be empirically wrong.",https://github.com/pytorch/pytorch/issues/176148,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: oncall: export, oncall: pt2",categorize and decide next maintainer owner/action,2026-03-09T23:28:22Z
issue,124328,"PreDispatch tracing doesn't produce correct ""val"" for Sparse Fake tensors.",https://github.com/pytorch/pytorch/issues/124328,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: pt2-dispatcher, oncall: export",categorize and decide next maintainer owner/action,2026-03-09T23:28:12Z
issue,166572,[Regression] torch.export.ExportedProgram.run_decompositions Starts to Break in Torch 2.8 with Mutable Buffer,https://github.com/pytorch/pytorch/issues/166572,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: oncall: export, oncall: pt2",categorize and decide next maintainer owner/action,2026-03-09T23:28:06Z
issue,174608,AOTInductor compile crashes with CUDA illegal memory access on exported SAM3 model with math SDPA,https://github.com/pytorch/pytorch/issues/174608,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: oncall: export, oncall: pt2",categorize and decide next maintainer owner/action,2026-03-09T23:24:39Z
issue,175520,Question about torch._dynamo.export and nn.GRU support,https://github.com/pytorch/pytorch/issues/175520,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: dynamo, oncall: export",categorize and decide next maintainer owner/action,2026-03-02T23:38:19Z
issue,33692,JIT tracing check fails with boolean tensor modifications,https://github.com/pytorch/pytorch/issues/33692,needs triage,65,triage signal present; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2026-03-02T01:23:15Z
issue,175790,Only first CUDA profile of the process is correct?,https://github.com/pytorch/pytorch/issues/175790,needs triage,65,PyTorch issue lacks triaged label; component/oncall labels: oncall: profiler,categorize and decide next maintainer owner/action,2026-02-25T21:08:29Z
issue,175493,FX GraphModule saving with TorchScript submodules,https://github.com/pytorch/pytorch/issues/175493,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: fx, oncall: jit",categorize and decide next maintainer owner/action,2026-02-23T00:49:39Z
issue,174555,Profiler chrome traces express record_function and profiler step incorrectly when using with_stack=True,https://github.com/pytorch/pytorch/issues/174555,needs triage,65,PyTorch issue lacks triaged label; component/oncall labels: oncall: profiler,categorize and decide next maintainer owner/action,2026-02-19T14:12:26Z
issue,162878,CUDA 13.0 Segmentation Fault in test/test_numba_integration.py::TestNumbaIntegration::test_array_adaptor,https://github.com/pytorch/pytorch/issues/162878,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: ci, module: cuda, oncall: releng",categorize and decide next maintainer owner/action,2026-02-16T09:00:23Z
issue,175045,Consolidate or retire pytorch/almalinux-builder Docker images after MAGMA deprecation,https://github.com/pytorch/pytorch/issues/175045,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: binaries, module: ci, oncall: releng",categorize and decide next maintainer owner/action,2026-02-15T18:20:03Z
issue,174488,[cd][testing] Add almalinux/9-base test to CD pipeline,https://github.com/pytorch/pytorch/issues/174488,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: binaries, module: cuda, oncall: releng",categorize and decide next maintainer owner/action,2026-02-06T20:54:51Z
issue,167623,torch.export: support external weights (.pt2 format),https://github.com/pytorch/pytorch/issues/167623,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: oncall: export, oncall: pt2",categorize and decide next maintainer owner/action,2026-02-05T14:52:00Z
issue,174375,[CUDA] flamegraph.pl download fails on cross-device filesystems,https://github.com/pytorch/pytorch/issues/174375,needs triage,65,PyTorch issue lacks triaged label; component/oncall labels: oncall: visualization,categorize and decide next maintainer owner/action,2026-02-05T13:09:33Z
issue,166205,Enable Graph Capture & Profiler Integration for PrivateUse1 Backends,https://github.com/pytorch/pytorch/issues/166205,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: privateuse1, oncall: profiler",categorize and decide next maintainer owner/action,2026-02-05T03:07:29Z
issue,174273,using profiler on jvp function with record_shapes=True causes memory leaks.,https://github.com/pytorch/pytorch/issues/174273,needs triage,65,PyTorch issue lacks triaged label; component/oncall labels: oncall: profiler,categorize and decide next maintainer owner/action,2026-02-04T10:55:27Z
issue,173546,[vllm] [release 2.10] V1 Test e2e + engine : test_draft_model_correctness[False-args0],https://github.com/pytorch/pytorch/issues/173546,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: vllm, oncall: releng",categorize and decide next maintainer owner/action,2026-02-02T19:29:22Z
issue,173492,torch.jit.script failed for inplace bitwise_and,https://github.com/pytorch/pytorch/issues/173492,needs triage,65,PyTorch issue lacks triaged label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2026-01-27T15:55:51Z
issue,173311,Issue: Remove unnecessary restriction on num_thresholds in TensorBoard PR curve summary,https://github.com/pytorch/pytorch/issues/173311,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: tensorboard, oncall: profiler",categorize and decide next maintainer owner/action,2026-01-26T18:27:19Z
issue,146152,[torch.export] Cannot export TorchVision fasterrcnn_mobilenet_v3_large_fpn,https://github.com/pytorch/pytorch/issues/146152,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: oncall: export, oncall: pt2",categorize and decide next maintainer owner/action,2026-01-23T19:06:23Z
issue,173075,UniformQuantizationObserverBase computes wrong scale and zero_point,https://github.com/pytorch/pytorch/issues/173075,needs triage,65,PyTorch issue lacks triaged label; component/oncall labels: oncall: quantization,categorize and decide next maintainer owner/action,2026-01-22T17:25:30Z
issue,171689,"After enabling XPU Graph, I cannot collect profile data during the replay phase.",https://github.com/pytorch/pytorch/issues/171689,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: xpu, oncall: profiler",categorize and decide next maintainer owner/action,2026-01-21T06:23:27Z
issue,172082,torch.ao.quantization.observer work on CPU (very slow!) due to using registered buffers on CPU,https://github.com/pytorch/pytorch/issues/172082,needs triage,65,PyTorch issue lacks triaged label; component/oncall labels: oncall: quantization,categorize and decide next maintainer owner/action,2026-01-13T18:01:15Z
pr,160883,Fallback to eager AC when backend=eager for TagActivationCheckpoint,https://github.com/pytorch/pytorch/pull/160883,PR blocked,65,failed checks or blocked/unstable merge state; PR closes or references open issue numbers; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2025-11-11T15:21:08Z
pr,155672,Fixed NLLLoss 1D input crash with torch.compile,https://github.com/pytorch/pytorch/pull/155672,PR blocked,65,failed checks or blocked/unstable merge state; PR closes or references open issue numbers; PR has not updated for at least 45 days,"identify whether block is CI, merge conflict, or review gate",2025-08-17T08:53:54Z
issue,154354,integer convolution,https://github.com/pytorch/pytorch/issues/154354,needs triage,65,"PyTorch issue lacks triaged label; needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: cuda, oncall: quantization",route to design/research decision before patch planning,2025-05-26T20:02:11Z
issue,61210,Jitted function cannot be pickled,https://github.com/pytorch/pytorch/issues/61210,needs triage,65,PyTorch issue lacks triaged label; label says issue is suitable for a scoped external contribution; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-04-03T19:30:59Z
issue,49961,rfc: automating the switching of inputs to the device of the params,https://github.com/pytorch/pytorch/issues/49961,needs triage,65,"PyTorch issue lacks triaged label; needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: rpc, module: ux, oncall: distributed",route to design/research decision before patch planning,2022-07-06T10:49:56Z
issue,55577,Source location range issue for dictionary,https://github.com/pytorch/pytorch/issues/55577,needs triage,65,"PyTorch issue lacks triaged label; label says issue is suitable for a scoped external contribution; old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, oncall: jit",categorize and decide next maintainer owner/action,2021-10-20T01:50:08Z
issue,51138,[TorchScript Usability] tensor printout format differs from eager mode,https://github.com/pytorch/pytorch/issues/51138,needs triage,65,PyTorch issue lacks triaged label; label says issue is suitable for a scoped external contribution; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-07-17T04:41:20Z
issue,45745,Native CUDA-Aware MPI support for MVAPICH2 and other MPI libraries,https://github.com/pytorch/pytorch/issues/45745,needs triage,65,"PyTorch issue lacks triaged label; needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda, oncall: distributed",route to design/research decision before patch planning,2020-10-08T20:26:16Z
issue,111901,Multiprocess. DataLoader worker is killed by signal: Segmentation fault.,https://github.com/pytorch/pytorch/issues/111901,high-priority/blocker,64,"high-priority/blocker/regression label present; needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: dataloader",confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-10-30T09:49:01Z
issue,159454,libtorch_cpu.so link libm.so issue:GLIBC_2.29 not found,https://github.com/pytorch/pytorch/issues/159454,high-priority/blocker,64,"high-priority/blocker/regression label present; needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: build",confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-10-29T03:50:24Z
issue,45198,DataLoader with cv2 and some numpy/cv2 import order causes workers to not work,https://github.com/pytorch/pytorch/issues/45198,high-priority/blocker,64,"high-priority/blocker/regression label present; needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: dataloader, module: multithreading, module: regression",confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-09-29T23:25:36Z
issue,66247,Floating point exception in mkl_vml_serv_GetMinN () on a specific computer,https://github.com/pytorch/pytorch/issues/66247,high-priority/blocker,64,"high-priority/blocker/regression label present; needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: crash, module: mkl",confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-09-29T23:24:38Z
issue,31758,DataLoader: Segmentation Fault (core dumped),https://github.com/pytorch/pytorch/issues/31758,high-priority/blocker,64,"high-priority/blocker/regression label present; needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: dataloader",confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-09-29T23:16:25Z
issue,162343,[BUG] `Aborted (core dumped)` on RTX 5090 with Nightly Build when running TTS inference,https://github.com/pytorch/pytorch/issues/162343,high-priority/blocker,64,"high-priority/blocker/regression label present; needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: wsl",confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-09-17T17:20:57Z
issue,162797,Segmentation fault in mkl_blas_avx512_dgemv_t_intrinsics,https://github.com/pytorch/pytorch/issues/162797,high-priority/blocker,64,"high-priority/blocker/regression label present; needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: intel, module: mkl",confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-09-13T07:48:28Z
issue,161865,SEGFAULT in libomp.dylib after torch.matmul on M4 Max with MPS backend,https://github.com/pytorch/pytorch/issues/161865,high-priority/blocker,64,"high-priority/blocker/regression label present; needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: crash, module: macos, module: mps",confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-09-10T13:28:19Z
issue,160876,F.silu(torch.tensor(float('-inf'))) returns nan instead of 0 in PyTorch 2.7.0+cu128,https://github.com/pytorch/pytorch/issues/160876,high-priority/blocker,64,"high-priority/blocker/regression label present; needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: nans and infs, module: regression",confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-08-19T19:24:32Z
issue,143780,Looking for valid compiling option for extension based on torch-2.1.0+cpu.cxx11.abi,https://github.com/pytorch/pytorch/issues/143780,high-priority/blocker,64,"high-priority/blocker/regression label present; needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: cpp-extensions, module: crash",confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-07-09T16:53:07Z
issue,157704,Seg faults on macos / OSX,https://github.com/pytorch/pytorch/issues/157704,high-priority/blocker,64,"high-priority/blocker/regression label present; needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: macos",confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-07-07T15:05:05Z
issue,156723,Segmentation fault on Apple Macbook Pro M4 24GB RAM and 1TB flash dasd during forward() with AutoModelForSequenceClassification using CPU,https://github.com/pytorch/pytorch/issues/156723,high-priority/blocker,64,high-priority/blocker/regression label present; needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: crash,confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-06-30T02:48:07Z
issue,156613,[CUDA 12.9 | sm_120] Build fails on RowwiseScaledMM.cu with PTXAS error (PyTorch 2.8.0a0),https://github.com/pytorch/pytorch/issues/156613,high-priority/blocker,64,"high-priority/blocker/regression label present; needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: build, module: crash, module: cuda, module: third_party",confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-06-23T16:21:46Z
issue,155693,torch typing: register_buffer -> device,https://github.com/pytorch/pytorch/issues/155693,high-priority/blocker,64,"high-priority/blocker/regression label present; needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: regression, module: typing",confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-06-13T03:52:05Z
issue,154014,Floating point exception when autocast is enabled,https://github.com/pytorch/pytorch/issues/154014,high-priority/blocker,64,"high-priority/blocker/regression label present; needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision), module: crash",confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-06-02T17:31:31Z
issue,153937,Dynamic compilation fails with torch 2.7,https://github.com/pytorch/pytorch/issues/153937,high-priority/blocker,64,"high-priority/blocker/regression label present; needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: regression, oncall: pt2",confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-05-23T17:27:53Z
issue,152102,Segmentation fault with # FIXME: copy.deepcopy() is not defined on nn.module,https://github.com/pytorch/pytorch/issues/152102,high-priority/blocker,64,"high-priority/blocker/regression label present; needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: nn",confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-04-24T15:52:02Z
issue,53895,"terminate called after throwing an instance of 'c10::Error' what(): isTuple() INTERNAL ASSERT FAILED at ""/home/wenda/libtorch/include/ATen/core/ivalue_inl.h"":927, please report a bug to PyTorch. Expected Tuple but got GenericList",https://github.com/pytorch/pytorch/issues/53895,high-priority/blocker,64,"high-priority/blocker/regression label present; needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: crash",confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-04-17T11:59:34Z
issue,149941,SEGV in static destructors when Vulkan is enabled,https://github.com/pytorch/pytorch/issues/149941,high-priority/blocker,64,"high-priority/blocker/regression label present; needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: vulkan",confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-03-25T15:08:16Z
issue,22259,using multi thread lead to gpu stuck with GPU-util 100%,https://github.com/pytorch/pytorch/issues/22259,high-priority/blocker,64,high-priority/blocker/regression label present; needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda,confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-02-13T07:27:42Z
issue,140230,Inconsistent Model Loading with `weights_only=True` for Custom Models,https://github.com/pytorch/pytorch/issues/140230,high-priority/blocker,64,high-priority/blocker/regression label present; needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: serialization,confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-11-19T21:28:23Z
issue,134198,[Intel CPU] Same CPU operations on Torch 2.3+ many times slower those on Torch 2.1,https://github.com/pytorch/pytorch/issues/134198,high-priority/blocker,64,"high-priority/blocker/regression label present; needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: intel, module: performance, module: regression",confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-08-28T02:05:32Z
issue,126963,Segmentation fault (core dumped) when using pytorch Conv layers,https://github.com/pytorch/pytorch/issues/126963,high-priority/blocker,64,"high-priority/blocker/regression label present; needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: cudnn",confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-08-07T01:45:42Z
issue,129304,`python3 setup.py bdist_wheel` tries to write to /usr/local/... during build,https://github.com/pytorch/pytorch/issues/129304,high-priority/blocker,64,"high-priority/blocker/regression label present; needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: build, module: regression",confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-07-30T05:33:01Z
issue,26165,Memory leak in multithreading environment when loading checkpoint,https://github.com/pytorch/pytorch/issues/26165,high-priority/blocker,64,"high-priority/blocker/regression label present; needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: memory usage, module: multiprocessing, module: multithreading",confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-07-24T20:48:35Z
issue,122478,6470 segmentation fault (core dumped),https://github.com/pytorch/pytorch/issues/122478,high-priority/blocker,64,"high-priority/blocker/regression label present; needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: crash",confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-07-18T15:31:20Z
issue,130708,"Converting a numpy array of size larger than 32,768 to a tensor causes a segmentation fault",https://github.com/pytorch/pytorch/issues/130708,high-priority/blocker,64,"high-priority/blocker/regression label present; needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: macos, module: numpy",confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-07-17T03:50:47Z
issue,127275,"Python sometimes crashes inexplicably, with ""/var/log/kernel. log"" displaying index errors",https://github.com/pytorch/pytorch/issues/127275,high-priority/blocker,64,"high-priority/blocker/regression label present; needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: cuda",confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-05-29T09:05:45Z
issue,122308,Segmentation fault about torch::nn::Conv2d in `libtorch2.0.1+cuda11.8`,https://github.com/pytorch/pytorch/issues/122308,high-priority/blocker,64,"high-priority/blocker/regression label present; needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: crash",confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-05-24T12:33:45Z
issue,125664,Aborted (core dumped) in torch.fft.irfftn/hfftn/ihfftn with malloc(): corrupted top size,https://github.com/pytorch/pytorch/issues/125664,high-priority/blocker,64,"high-priority/blocker/regression label present; needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: crash, module: fft",confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-05-09T03:11:03Z
issue,124656,Importing polars before torch causes a segfault,https://github.com/pytorch/pytorch/issues/124656,high-priority/blocker,64,"high-priority/blocker/regression label present; needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: crash",confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-04-23T18:01:07Z
issue,122935,Pytorch 2.1 / 2.2 uses 4x more memory during data loading than Pytorch 2.4 or 1.13,https://github.com/pytorch/pytorch/issues/122935,high-priority/blocker,64,"high-priority/blocker/regression label present; needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: dataloader, module: memory usage, module: regression",confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-04-02T21:31:32Z
issue,121235,Issue with sentence_transformers pointing to torch tensors manipulations,https://github.com/pytorch/pytorch/issues/121235,high-priority/blocker,64,"high-priority/blocker/regression label present; needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: macos",confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-04-01T13:53:34Z
issue,118088,Segmentation fault (SIGSEGV) in c10::detail::getNonDeterministicRandom(bool),https://github.com/pytorch/pytorch/issues/118088,high-priority/blocker,64,"high-priority/blocker/regression label present; needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: cuda, module: random",confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-02-08T07:54:27Z
issue,117494,Kernel crashed when importing torch,https://github.com/pytorch/pytorch/issues/117494,high-priority/blocker,64,"high-priority/blocker/regression label present; needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: crash",confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-01-16T14:48:12Z
issue,26120,Process fails with assertion error in magma-cuda100,https://github.com/pytorch/pytorch/issues/26120,high-priority/blocker,64,"high-priority/blocker/regression label present; needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: dependency bug, module: linear algebra",confirm whether existing evidence is enough; otherwise wait for a reproducer,2023-12-27T05:43:28Z
issue,109819,ValueError: only one element tensors can be converted to Python scalars,https://github.com/pytorch/pytorch/issues/109819,high-priority/blocker,64,high-priority/blocker/regression label present; needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: regression,confirm whether existing evidence is enough; otherwise wait for a reproducer,2023-10-06T19:13:40Z
issue,91633,Segmentation fault when running torch.nn.functional.fractional_max_pool3d on torch 1.13.1,https://github.com/pytorch/pytorch/issues/91633,high-priority/blocker,64,high-priority/blocker/regression label present; needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: crash,confirm whether existing evidence is enough; otherwise wait for a reproducer,2023-08-21T06:19:03Z
issue,105872,PyTorch 2.0.x `CUDA error: operation not supported` when `Tensor.to` a different device,https://github.com/pytorch/pytorch/issues/105872,high-priority/blocker,64,"high-priority/blocker/regression label present; needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: regression",confirm whether existing evidence is enough; otherwise wait for a reproducer,2023-07-31T17:14:56Z
issue,99200,Unused `import torch` followed by `cuml.NearestNeighbors` leads to nondeterministic segfault (during Python process exit?),https://github.com/pytorch/pytorch/issues/99200,high-priority/blocker,64,high-priority/blocker/regression label present; needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: crash,confirm whether existing evidence is enough; otherwise wait for a reproducer,2023-04-21T13:31:06Z
issue,89817,"Basic math operations produce a ""floating point exception""",https://github.com/pytorch/pytorch/issues/89817,high-priority/blocker,64,"high-priority/blocker/regression label present; needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: crash",confirm whether existing evidence is enough; otherwise wait for a reproducer,2023-03-28T15:09:37Z
issue,96779,Segmentation fault (core dumped) during Torch finetuning (at random step),https://github.com/pytorch/pytorch/issues/96779,high-priority/blocker,64,high-priority/blocker/regression label present; needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: crash,confirm whether existing evidence is enough; otherwise wait for a reproducer,2023-03-16T00:59:51Z
issue,70245,"SEGFAULT on ""import torch""",https://github.com/pytorch/pytorch/issues/70245,high-priority/blocker,64,high-priority/blocker/regression label present; needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: crash,confirm whether existing evidence is enough; otherwise wait for a reproducer,2023-01-23T17:10:01Z
issue,66033,Crashed with `terminating with uncaught exception of type std::__1::system_error: condition_variable wait failed: Invalid argument`,https://github.com/pytorch/pytorch/issues/66033,high-priority/blocker,64,"high-priority/blocker/regression label present; needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: crash, module: macos, module: multithreading",confirm whether existing evidence is enough; otherwise wait for a reproducer,2023-01-13T09:32:06Z
issue,91533,Cannot cast float64 to float32,https://github.com/pytorch/pytorch/issues/91533,high-priority/blocker,64,"high-priority/blocker/regression label present; needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: macos, module: numpy",confirm whether existing evidence is enough; otherwise wait for a reproducer,2023-01-03T14:31:18Z
issue,89714,Segfault on torch.nn.functional.one_hot with large tensor on Python 3.9,https://github.com/pytorch/pytorch/issues/89714,high-priority/blocker,64,high-priority/blocker/regression label present; needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: crash,confirm whether existing evidence is enough; otherwise wait for a reproducer,2022-11-29T10:26:30Z
issue,87209,"Reproducible ""CUDA error: an illegal memory access was encountered""",https://github.com/pytorch/pytorch/issues/87209,high-priority/blocker,64,"high-priority/blocker/regression label present; needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: cuda",confirm whether existing evidence is enough; otherwise wait for a reproducer,2022-11-02T16:57:55Z
issue,87446,ipykernel crash importing torch after scipy in .ipynb file,https://github.com/pytorch/pytorch/issues/87446,high-priority/blocker,64,high-priority/blocker/regression label present; needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: crash,confirm whether existing evidence is enough; otherwise wait for a reproducer,2022-10-21T14:59:30Z
issue,76543,Segfault in ~PyFunctionPreHook,https://github.com/pytorch/pytorch/issues/76543,high-priority/blocker,64,"high-priority/blocker/regression label present; needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: crash",confirm whether existing evidence is enough; otherwise wait for a reproducer,2022-08-15T21:19:49Z
issue,81532,forward program terminated from __cxa_pure_virtual,https://github.com/pytorch/pytorch/issues/81532,high-priority/blocker,64,"high-priority/blocker/regression label present; needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: cpp, module: crash",confirm whether existing evidence is enough; otherwise wait for a reproducer,2022-07-25T13:32:27Z
issue,71149,Memory leak in distributions.multivariate_normal.MultivariateNormal,https://github.com/pytorch/pytorch/issues/71149,high-priority/blocker,64,"high-priority/blocker/regression label present; needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: distributions, module: memory usage",confirm whether existing evidence is enough; otherwise wait for a reproducer,2022-02-10T03:16:46Z
issue,68920,Breakes with -OO flag in script,https://github.com/pytorch/pytorch/issues/68920,high-priority/blocker,64,high-priority/blocker/regression label present; needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: regression,confirm whether existing evidence is enough; otherwise wait for a reproducer,2021-12-02T19:57:59Z
issue,62308,Error while running FINETUNING TORCHVISION MODELS,https://github.com/pytorch/pytorch/issues/62308,high-priority/blocker,64,"high-priority/blocker/regression label present; needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: vision",confirm whether existing evidence is enough; otherwise wait for a reproducer,2021-08-04T07:49:55Z
issue,62331,broken pipe error,https://github.com/pytorch/pytorch/issues/62331,high-priority/blocker,64,"high-priority/blocker/regression label present; needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: cuda",confirm whether existing evidence is enough; otherwise wait for a reproducer,2021-07-28T15:40:15Z
issue,60835,"Pytorch V1.1 runs OK, but Pytorch v1.5 to 1.9 run wrong",https://github.com/pytorch/pytorch/issues/60835,high-priority/blocker,64,"high-priority/blocker/regression label present; needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: regression",confirm whether existing evidence is enough; otherwise wait for a reproducer,2021-06-28T13:47:35Z
issue,48555,segment fault for Image model training by four GPUs,https://github.com/pytorch/pytorch/issues/48555,high-priority/blocker,64,high-priority/blocker/regression label present; needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: crash,confirm whether existing evidence is enough; otherwise wait for a reproducer,2021-04-06T10:04:17Z
issue,54578,cublasSgemmStridedBatched failure when calling grad of grad,https://github.com/pytorch/pytorch/issues/54578,high-priority/blocker,64,"high-priority/blocker/regression label present; needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: cuda",confirm whether existing evidence is enough; otherwise wait for a reproducer,2021-03-26T16:24:49Z
issue,40425,Segment Fault after use cusolverDnDestroy() with torch1.5,https://github.com/pytorch/pytorch/issues/40425,high-priority/blocker,64,"high-priority/blocker/regression label present; needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: cuda, module: dependency bug",confirm whether existing evidence is enough; otherwise wait for a reproducer,2020-06-29T16:14:00Z
issue,36871,Stochasticity for DistributedDataParallel on CPU but not on GPU,https://github.com/pytorch/pytorch/issues/36871,high-priority/blocker,64,"high-priority/blocker/regression label present; needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: data parallel, module: determinism, oncall: distributed",confirm whether existing evidence is enough; otherwise wait for a reproducer,2020-06-25T12:12:17Z
issue,35792,Python 3 build PyTorch got core dump,https://github.com/pytorch/pytorch/issues/35792,high-priority/blocker,64,high-priority/blocker/regression label present; needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: crash,confirm whether existing evidence is enough; otherwise wait for a reproducer,2020-04-01T13:34:28Z
issue,31362,sklearn and pytorch incompatibility issue,https://github.com/pytorch/pytorch/issues/31362,high-priority/blocker,64,high-priority/blocker/regression label present; needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: crash,confirm whether existing evidence is enough; otherwise wait for a reproducer,2020-02-25T08:42:27Z
issue,28319,Seg-fault in LayerNormKernelImpl,https://github.com/pytorch/pytorch/issues/28319,high-priority/blocker,64,"high-priority/blocker/regression label present; needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: cuda",confirm whether existing evidence is enough; otherwise wait for a reproducer,2019-12-13T15:22:28Z
issue,29498,"""malloc(): memory corruption (fast)"", action=3) at malloc.c",https://github.com/pytorch/pytorch/issues/29498,high-priority/blocker,64,high-priority/blocker/regression label present; needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: crash,confirm whether existing evidence is enough; otherwise wait for a reproducer,2019-11-18T18:19:19Z
pr,189024,[Dynamo] Support exception attribute access (StopIteration/AttributeError/NameError),https://github.com/pytorch/pytorch/pull/189024,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-09T02:17:19Z
pr,189318,Bump pip from 26.0.1 to 26.1.2 in /.ci/docker,https://github.com/pytorch/pytorch/pull/189318,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/docker,"identify whether block is CI, merge conflict, or review gate",2026-07-09T02:16:34Z
pr,188597,[ROCm] Add initial support for gfx1250,https://github.com/pytorch/pytorch/pull/188597,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/b200, ciflow/binaries, ciflow/docker, ciflow/h100","identify whether block is CI, merge conflict, or review gate",2026-07-09T02:14:29Z
pr,189173,torchfuzz: detach batch_norm running stats,https://github.com/pytorch/pytorch/pull/189173,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-07-09T02:13:41Z
pr,189111,[torchcomms hash update] update the pinned torchcomms hash,https://github.com/pytorch/pytorch/pull/189111,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-09T02:13:38Z
pr,184218,Support decorator/curried calling convention for torch.utils.checkpoint.checkpoint,https://github.com/pytorch/pytorch/pull/184218,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/dtensor, ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-09T02:13:23Z
pr,189328,Add retry with backoff to download_mnist.py,https://github.com/pytorch/pytorch/pull/189328,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-07-09T02:05:04Z
pr,181680,[triton hash update] update the pinned triton hash,https://github.com/pytorch/pytorch/pull/181680,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/docker, ciflow/inductor, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-09T01:54:30Z
pr,189217,[vision hash update] update the pinned vision hash,https://github.com/pytorch/pytorch/pull/189217,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-09T01:54:24Z
pr,187288,[vllm hash update] update the pinned vllm hash,https://github.com/pytorch/pytorch/pull/187288,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/trunk, ciflow/vllm","identify whether block is CI, merge conflict, or review gate",2026-07-09T01:54:21Z
pr,188262,[CI] fix OpenBLAS build for non-AArch64,https://github.com/pytorch/pytorch/pull/188262,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/docker, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-09T01:54:09Z
pr,179651,[torchtitan hash update] update the pinned torchtitan hash,https://github.com/pytorch/pytorch/pull/179651,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-09T01:54:03Z
pr,189083,[profiler][cupti] Drive periodic cuptiActivityFlushAll from the decode thread,https://github.com/pytorch/pytorch/pull/189083,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-07-09T01:53:56Z
pr,189185,[profiler][cupti] Make CuptiMonitor a singleton configured via a global configure(),https://github.com/pytorch/pytorch/pull/189185,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-07-09T01:53:53Z
pr,188621,"[profiler][cupti] Generate CUPTI field-id catalogs from the ABI as Field(id, ctype)",https://github.com/pytorch/pytorch/pull/188621,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-07-09T01:53:52Z
pr,188100,Support callable msg lambda in TestCase.assert* methods beyond assertEqual(),https://github.com/pytorch/pytorch/pull/188100,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-07-09T01:53:44Z
pr,115316,Automated submodule update: FBGEMM,https://github.com/pytorch/pytorch/pull/115316,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/b200, ciflow/h100, ciflow/rocm-mi300","identify whether block is CI, merge conflict, or review gate",2026-07-09T01:53:38Z
pr,189175,torchfuzz: fix clamp_min/max emitting None bound,https://github.com/pytorch/pytorch/pull/189175,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-07-09T01:53:33Z
pr,188299,Lazily build error messages for assert*() calls across tests to reduce overhead,https://github.com/pytorch/pytorch/pull/188299,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/b200, ciflow/dtensor, ciflow/h100, ciflow/h100-symm-mem","identify whether block is CI, merge conflict, or review gate",2026-07-09T01:53:28Z
pr,189192,[MPS] Add `geqrf` and refactor `linalg_qr`,https://github.com/pytorch/pytorch/pull/189192,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/mps","identify whether block is CI, merge conflict, or review gate",2026-07-09T01:53:27Z
pr,188573,Cleanup dead code across tests,https://github.com/pytorch/pytorch/pull/188573,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/b200, ciflow/dtensor, ciflow/h100, ciflow/inductor","identify whether block is CI, merge conflict, or review gate",2026-07-09T01:53:25Z
pr,189264,"[ROCm][CI] Make rocm-nightly workflow fully mi300-specific (name, ciflow tag, filename)",https://github.com/pytorch/pytorch/pull/189264,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/binaries, ciflow/rocm-mi300","identify whether block is CI, merge conflict, or review gate",2026-07-09T01:53:23Z
pr,187465,[Symm_mem] Support expandable segements inside PyTorch Symmetric memory for NCCL backend,https://github.com/pytorch/pytorch/pull/187465,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/h100-symm-mem, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-09T01:53:15Z
pr,189302,[ROCm][CI] Make ROCM_SO_FILES a single shared source of truth for wheel bundling + preload,https://github.com/pytorch/pytorch/pull/189302,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/binaries_wheel, ciflow/rocm-mi300","identify whether block is CI, merge conflict, or review gate",2026-07-09T01:53:12Z
pr,189314,flex_gemm: recognize 2-D block local-reduce contracts,https://github.com/pytorch/pytorch/pull/189314,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-09T01:53:11Z
pr,189315,flex_gemm: store 2-D block local reductions,https://github.com/pytorch/pytorch/pull/189315,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-09T01:53:10Z
pr,187699,Add a debug log when we skip sync_decision_cross_ranks (#187699),https://github.com/pytorch/pytorch/pull/187699,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-09T01:53:09Z
pr,189316,flex_gemm: support mx block-scale stores,https://github.com/pytorch/pytorch/pull/189316,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-09T01:53:08Z
pr,189266,[ROCm][CI] Fix py3.15 ROCm manywheel numpy source build (pkg-config python3),https://github.com/pytorch/pytorch/pull/189266,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/binaries_wheel,"identify whether block is CI, merge conflict, or review gate",2026-07-09T01:53:02Z
pr,189291,Add an accurate Metal erfc for MPS (eager + inductor),https://github.com/pytorch/pytorch/pull/189291,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/mps, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-09T01:53:01Z
pr,188862,[inductor] Fix bool minimum/maximum crash in index propagation (#188862) (#188862),https://github.com/pytorch/pytorch/pull/188862,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-09T01:53:01Z
pr,188643,[AOTI] Add external constants API for zero-copy weight sharing (#188643),https://github.com/pytorch/pytorch/pull/188643,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-09T01:52:59Z
pr,181726,[xpu][1/4]Implement scaled_mm_v2 for MXFP8/MXFP4/NVFP4 on XPU,https://github.com/pytorch/pytorch/pull/181726,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-07-09T01:51:44Z
pr,189051,[dynamo] Reject keyword args to set()/frozenset() constructors,https://github.com/pytorch/pytorch/pull/189051,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-09T01:51:36Z
pr,189325,[Dynamo] Fix self-referential bw_compiler on backend re-entry,https://github.com/pytorch/pytorch/pull/189325,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-09T01:32:21Z
pr,179286,Use C++20 concepts where it improves readability,https://github.com/pytorch/pytorch/pull/179286,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/trunk, ciflow/vllm","identify whether block is CI, merge conflict, or review gate",2026-07-09T01:05:38Z
pr,189249,[Dynamo] Remove dead code in tvm and common backends,https://github.com/pytorch/pytorch/pull/189249,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-09T01:04:57Z
pr,189311,[wip] Reduce hackery for Dynamo appeasement,https://github.com/pytorch/pytorch/pull/189311,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-09T01:04:10Z
pr,186055,Add CUDA graph support for torch.while_loop,https://github.com/pytorch/pytorch/pull/186055,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-07-09T01:04:09Z
pr,189088,[SymmMem] Place the signal pad at the front of symmetric memory allocations,https://github.com/pytorch/pytorch/pull/189088,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/b200-symm-mem, ciflow/h100-symm-mem","identify whether block is CI, merge conflict, or review gate",2026-07-09T01:04:08Z
pr,189305,"Reapply ""Remove CUDA sync in torch.combinations to improve `torch.compile` support (#186595)""",https://github.com/pytorch/pytorch/pull/189305,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/dtensor, ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-09T01:04:00Z
pr,183328,[special] Add arbitrary-order modified Bessel functions,https://github.com/pytorch/pytorch/pull/183328,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-09T00:43:43Z
pr,189319,Enable flake8-bugbear B007 (unused loop control variable),https://github.com/pytorch/pytorch/pull/189319,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/dtensor, ciflow/h100-symm-mem, ciflow/inductor, ciflow/inductor-pallas","identify whether block is CI, merge conflict, or review gate",2026-07-09T00:09:57Z
pr,181781,[dynamo] turn canonicalize_output_graph_node_order on by default,https://github.com/pytorch/pytorch/pull/181781,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/dtensor, ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T23:50:03Z
pr,188137,[inductor] Decompose shard_dim_alltoall via regular decomposition,https://github.com/pytorch/pytorch/pull/188137,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/dtensor, ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-08T23:49:43Z
pr,187468,[dynamo] add break_graph_if_unsupported to LOAD_ATTR,https://github.com/pytorch/pytorch/pull/187468,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/dynamo, ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T23:23:28Z
pr,187469,[dynamo] extract UDOV hooks: lookup_instance_dict and call_getattr_fallback,https://github.com/pytorch/pytorch/pull/187469,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T23:23:14Z
pr,181780,[dynamo] set canonicalize_output_graph_node_order=True for export tests,https://github.com/pytorch/pytorch/pull/181780,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/dtensor, ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T23:22:50Z
pr,189190,flex_gemm: extend feed-main to trailing pointwise and axis-1 fragments,https://github.com/pytorch/pytorch/pull/189190,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T23:12:53Z
pr,185359,Add CE multicast low-contention all-gather,https://github.com/pytorch/pytorch/pull/185359,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/h100-symm-mem, ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-08T22:47:54Z
pr,187458,[MPS] Fuse cross_entropy into a native Metal kernel,https://github.com/pytorch/pytorch/pull/187458,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/mps, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T22:47:52Z
pr,189295,Pin flash-attention checkout in FA3 ABI-stable H100 smoke test,https://github.com/pytorch/pytorch/pull/189295,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/h100,"identify whether block is CI, merge conflict, or review gate",2026-07-08T22:36:47Z
pr,189257,[rocm] Re-enable conv2d backward parametrized test on ROCm by using fp16 on NAVI,https://github.com/pytorch/pytorch/pull/189257,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/inductor-rocm-mi200, ciflow/inductor-rocm-mi300, ciflow/inductor-rocm-mi355","identify whether block is CI, merge conflict, or review gate",2026-07-08T22:19:46Z
pr,189080,[inductor] Generalize reinplacing to functional ATen ops with an in-place variant,https://github.com/pytorch/pytorch/pull/189080,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T22:03:33Z
pr,186790,Remove c10/util/Array.h,https://github.com/pytorch/pytorch/pull/186790,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/trunk, ciflow/vllm","identify whether block is CI, merge conflict, or review gate",2026-07-08T21:50:46Z
pr,189277,Skip redundant contiguity rescan in empty_tensor_restride (robust) (#189277),https://github.com/pytorch/pytorch/pull/189277,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-07-08T21:50:21Z
pr,175900,[Inductor] Fix Enum serialization in user-defined Triton kernel codegen (#175900),https://github.com/pytorch/pytorch/pull/175900,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-08T21:50:16Z
pr,189220,Strip the arc experiment from workflow check_experiments,https://github.com/pytorch/pytorch/pull/189220,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor-pallas,"identify whether block is CI, merge conflict, or review gate",2026-07-08T21:36:08Z
pr,189202,[CuteDSL][RMSNorm] use const_data_ptr for alignment check,https://github.com/pytorch/pytorch/pull/189202,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-07-08T21:23:13Z
pr,189153,Bump onnx from 1.21.0 to 1.22.0 in /.ci/docker,https://github.com/pytorch/pytorch/pull/189153,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/docker,"identify whether block is CI, merge conflict, or review gate",2026-07-08T21:09:36Z
pr,177757,[MPS] support for SparseCSR tensor and its unary operators,https://github.com/pytorch/pytorch/pull/177757,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/mps,"identify whether block is CI, merge conflict, or review gate",2026-07-08T21:06:41Z
pr,188824,Port doctest tests from cpython test_generators.py,https://github.com/pytorch/pytorch/pull/188824,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2026-07-08T21:06:06Z
pr,157149,[generator] Close all open generators in compile_subgraph,https://github.com/pytorch/pytorch/pull/157149,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-08T21:06:05Z
pr,187744,[Dynamo] Add a per-generator exception stack,https://github.com/pytorch/pytorch/pull/187744,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-08T21:06:02Z
pr,187690,[Dynamo] Refactor generator methods,https://github.com/pytorch/pytorch/pull/187690,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T21:06:01Z
pr,189131,Skip building unused XNNPACK microkernels-all; guard GCC14 workaround,https://github.com/pytorch/pytorch/pull/189131,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-07-08T20:53:11Z
pr,184447,[inductor][autotuning] add incremental autotune foundation,https://github.com/pytorch/pytorch/pull/184447,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T20:50:06Z
pr,183684,[inductor] extract _bench_launchers and _finalize_autotune_winner helpers,https://github.com/pytorch/pytorch/pull/183684,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T20:49:59Z
pr,183683,[inductor] extract _iter_compile_results generator with optional thread pool,https://github.com/pytorch/pytorch/pull/183683,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T20:49:54Z
pr,183682,"[inductor] extract _bundle_compile_result, _all_failed_fallback, _maybe_put_static_autotuner",https://github.com/pytorch/pytorch/pull/183682,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T20:49:49Z
pr,181826,[inductor][autotuning] extract _should_coordesc_tune cached property (#181826),https://github.com/pytorch/pytorch/pull/181826,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-08T20:49:37Z
pr,189289,[TEST] Refactor test_c10d_pypg.py with hw_classification,https://github.com/pytorch/pytorch/pull/189289,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-07-08T20:49:21Z
pr,189263,Remove the Miniconda base env from the CD builder images,https://github.com/pytorch/pytorch/pull/189263,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/binaries, ciflow/docker","identify whether block is CI, merge conflict, or review gate",2026-07-08T20:48:54Z
pr,189206,[Inductor] register `target_code` in `input_codes` tracker,https://github.com/pytorch/pytorch/pull/189206,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T20:35:27Z
pr,189186,"Extend ""Fix positional arg overflow in TritonBenchmarkRequest when template inputs are deduplicated"" to CUTLASS backend",https://github.com/pytorch/pytorch/pull/189186,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/b200, ciflow/h100, ciflow/inductor, ciflow/rocm-mi300","identify whether block is CI, merge conflict, or review gate",2026-07-08T20:27:22Z
pr,189247,Register _foreach_clone XPU dispatch,https://github.com/pytorch/pytorch/pull/189247,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-07-08T20:26:40Z
pr,186056,Support input mutations in CUDA graph control flow.,https://github.com/pytorch/pytorch/pull/186056,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-07-08T20:26:01Z
pr,189044,[inductor] Handle symbolic float scalars in FX wrapper metadata hook and reference analysis (#189044),https://github.com/pytorch/pytorch/pull/189044,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2026-07-08T20:25:58Z
pr,106149,Automated submodule update: kineto,https://github.com/pytorch/pytorch/pull/106149,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-07-08T20:25:54Z
pr,189197,make post-processing in codegen mandatory,https://github.com/pytorch/pytorch/pull/189197,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T20:25:29Z
pr,189246,[DO NOT MERGE] Test L3 CRCR,https://github.com/pytorch/pytorch/pull/189246,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/crcr/crcr-test,"identify whether block is CI, merge conflict, or review gate",2026-07-08T20:23:09Z
pr,189042,[dynamo] Add specialize_args option to assume_constant_result,https://github.com/pytorch/pytorch/pull/189042,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T20:16:02Z
pr,183910,[Test] Make test/nn/test_embedding.py device-agnostic for out-of-tree backends,https://github.com/pytorch/pytorch/pull/183910,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-07-08T19:50:46Z
pr,185251,[Test] Move device-agnostic grouped_gemm tests to new test_matmul.py,https://github.com/pytorch/pytorch/pull/185251,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/b200, ciflow/h100, ciflow/rocm-mi300","identify whether block is CI, merge conflict, or review gate",2026-07-08T19:50:28Z
pr,188440,Simplify atomic_add_float using atomic_ref,https://github.com/pytorch/pytorch/pull/188440,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/android, ciflow/inductor, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-08T19:22:53Z
pr,187307,Fix the potential global range overflow issue,https://github.com/pytorch/pytorch/pull/187307,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/trunk, ciflow/xpu","identify whether block is CI, merge conflict, or review gate",2026-07-08T19:13:28Z
pr,189076,[inductor] null-guard size/alignment asserts for optional outputs,https://github.com/pytorch/pytorch/pull/189076,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-08T19:13:05Z
pr,180250,Add setup_.py shim for legacy build commands,https://github.com/pytorch/pytorch/pull/180250,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/binaries, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-08T19:12:48Z
pr,174699,Add CuTeDSL-based scaled grouped MM kernel,https://github.com/pytorch/pytorch/pull/174699,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/b200, ciflow/docker, ciflow/h100, ciflow/rocm-mi300","identify whether block is CI, merge conflict, or review gate",2026-07-08T18:45:25Z
pr,189276,Update third_party/kineto submodule to cb6e5f7,https://github.com/pytorch/pytorch/pull/189276,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-07-08T18:10:06Z
pr,184365,Memory aware partition selection for partitioned scatter fx pass,https://github.com/pytorch/pytorch/pull/184365,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/inductor-perf-test-nightly, ciflow/inductor-perf-test-nightly-rocm-mi350, ciflow/inductor-rocm-mi355","identify whether block is CI, merge conflict, or review gate",2026-07-08T18:04:47Z
pr,187508,c10d: register nccl2 backend,https://github.com/pytorch/pytorch/pull/187508,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/torchtitan,"identify whether block is CI, merge conflict, or review gate",2026-07-08T17:00:06Z
pr,186918,[Test] Add test hardware classification for test classes,https://github.com/pytorch/pytorch/pull/186918,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-07-08T16:17:52Z
pr,188181,Move __dict__ view (get_dict_vt) onto base VariableTracker,https://github.com/pytorch/pytorch/pull/188181,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T16:17:37Z
pr,189178,Add PyTorch AI policy,https://github.com/pytorch/pytorch/pull/189178,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-07-08T16:15:18Z
pr,185918,[Test] Make autocast op-list tests device-agnostic via registry,https://github.com/pytorch/pytorch/pull/185918,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-07-08T15:57:15Z
pr,189103,Support dist.split_group on CPU-only (gloo) process groups,https://github.com/pytorch/pytorch/pull/189103,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-08T15:35:21Z
pr,188650,[SymmMem] Allocate symmetric memory through the CUDACachingAllocator v2,https://github.com/pytorch/pytorch/pull/188650,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/b200-symm-mem, ciflow/h100-symm-mem, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T15:26:33Z
pr,185220,[Test] Make test_wrap_inductor_compiled_regions.py device-agnostic for out-of-tree backends,https://github.com/pytorch/pytorch/pull/185220,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2026-07-08T14:08:40Z
pr,188736,[xpu][inductor] update Intel Triton commit pin,https://github.com/pytorch/pytorch/pull/188736,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/docker, ciflow/inductor, ciflow/xpu","identify whether block is CI, merge conflict, or review gate",2026-07-08T10:34:30Z
pr,178336,[Distributed] Make DDP tests and tensor parallel dependencies backend agnostic,https://github.com/pytorch/pytorch/pull/178336,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/torchtitan,"identify whether block is CI, merge conflict, or review gate",2026-07-08T10:17:37Z
pr,184355,[3/N][Test] Migrate TestBwdGradients device-agnostic skips to `@skipOps`,https://github.com/pytorch/pytorch/pull/184355,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-07-08T09:32:56Z
pr,184356,[4/N][Test] Migrate TestFwdGradients device-agnostic skips to `@skipOps`,https://github.com/pytorch/pytorch/pull/184356,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-07-08T08:41:27Z
pr,176574,Increase dynamo coverage of the `operator` module by adding polyfills,https://github.com/pytorch/pytorch/pull/176574,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T08:08:34Z
pr,186915,[WIP][XPU]Enable inductor cases when issues are fixed,https://github.com/pytorch/pytorch/pull/186915,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/b200, ciflow/dtensor, ciflow/h100, ciflow/inductor","identify whether block is CI, merge conflict, or review gate",2026-07-08T07:45:36Z
pr,181000,[inductor] Dump Python stacks on CI test subprocess timeout,https://github.com/pytorch/pytorch/pull/181000,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/rocm-mi200, ciflow/rocm-mi300, ciflow/rocm-mi355","identify whether block is CI, merge conflict, or review gate",2026-07-08T06:51:27Z
pr,182376,[CUDA][CPU] Unify CPU/GPU relu/max behavior,https://github.com/pytorch/pytorch/pull/182376,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-07-08T05:34:02Z
pr,188043,[Test] Make test/test_sort_and_select.py device-agnostic for out-of-tree backends,https://github.com/pytorch/pytorch/pull/188043,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-07-08T03:59:54Z
pr,189195,Cuda Graph Default Enable (#189195),https://github.com/pytorch/pytorch/pull/189195,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T02:23:24Z
pr,181357,[dynamo] Add polyfill for operator.indexOf,https://github.com/pytorch/pytorch/pull/181357,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T02:18:56Z
pr,176876,[inductor] Add device-type benchmarking dispatch registry,https://github.com/pytorch/pytorch/pull/176876,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T02:18:46Z
pr,188276,[FSDP] Add MORI SDMA all-gather backend and zero-copy output capability,https://github.com/pytorch/pytorch/pull/188276,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T00:54:36Z
pr,183620,Port 2 tests under test/distributed folder to XPU,https://github.com/pytorch/pytorch/pull/183620,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/torchtitan,"identify whether block is CI, merge conflict, or review gate",2026-07-08T00:49:47Z
pr,189027,[dynamo] Guard os.environ reads with an ambient env-var guard,https://github.com/pytorch/pytorch/pull/189027,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T00:38:03Z
pr,179550,Enable eager tests for inductor-unittest - Enable TestViewOps and TestDevicePrecision on Intel GPU,https://github.com/pytorch/pytorch/pull/179550,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/mps, ciflow/xpu","identify whether block is CI, merge conflict, or review gate",2026-07-07T23:20:53Z
pr,179549,Enable eager tests for inductor-unittest - define xfails for xpu in op_db for test_ops.py using device_type list in DecorateInfo,https://github.com/pytorch/pytorch/pull/179549,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/xpu,"identify whether block is CI, merge conflict, or review gate",2026-07-07T23:05:37Z
pr,176689,Enable eager tests for inductor-unittest - enable test_ops_fwd_gradients.py and test_ops_gradients.py on Intel GPU,https://github.com/pytorch/pytorch/pull/176689,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/xpu,"identify whether block is CI, merge conflict, or review gate",2026-07-07T22:44:17Z
pr,176688,Enable eager tests for inductor-unittest - enable test_dtypes and align XPU op dyptes with OpInfo,https://github.com/pytorch/pytorch/pull/176688,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/xpu,"identify whether block is CI, merge conflict, or review gate",2026-07-07T22:41:50Z
pr,186100,"[dynamo, nested graph breaks] enable nested_graph_breaks by default on OSS",https://github.com/pytorch/pytorch/pull/186100,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/dynamo, ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T22:27:42Z
pr,185524,"[dynamo, nested graph breaks] add strong nested graph break test infrastructure",https://github.com/pytorch/pytorch/pull/185524,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T22:27:31Z
pr,189095,[inductor] Preserve stream annotations across pattern replacements,https://github.com/pytorch/pytorch/pull/189095,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-07T21:59:07Z
pr,187131,[muon] add multi-tensor (foreach) standard Muon path (#187131),https://github.com/pytorch/pytorch/pull/187131,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-07-07T21:29:05Z
pr,187133,[muon] add CUDA graph capture for the NS computation (#187133),https://github.com/pytorch/pytorch/pull/187133,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-07-07T20:51:41Z
pr,187132,[muon] add Gram Newton-Schulz algorithm + batched standard NS for square groups (#187132),https://github.com/pytorch/pytorch/pull/187132,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-07-07T20:51:40Z
pr,188979,remaining HOPs register_fake,https://github.com/pytorch/pytorch/pull/188979,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-07-07T20:51:23Z
pr,182145,[ROCm] Reland #181955: Redirect index_add_ to scatter_add_ for large tensors,https://github.com/pytorch/pytorch/pull/182145,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/rocm-mi300,"identify whether block is CI, merge conflict, or review gate",2026-07-07T20:12:10Z
pr,165766,Enable kineto XPU scope profiler extending existing XPU profiler plugin,https://github.com/pytorch/pytorch/pull/165766,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/xpu,"identify whether block is CI, merge conflict, or review gate",2026-07-07T19:13:14Z
pr,188130,[Test] Make test_autocast device-agnostic for Out-of-Tree backends,https://github.com/pytorch/pytorch/pull/188130,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-07-07T17:33:31Z
pr,184946,[Inductor] Fix split reduction codegen and skip redundant masking for block ptr backends (#184946),https://github.com/pytorch/pytorch/pull/184946,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T17:14:10Z
pr,188766,[CP] Fix FlexCP + load balancing enabled + aot_fx_trace issue,https://github.com/pytorch/pytorch/pull/188766,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/dtensor, ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-07T17:06:29Z
pr,187164,[ROCm] unskip export tests on ROCm,https://github.com/pytorch/pytorch/pull/187164,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/rocm-mi200, ciflow/rocm-mi300, ciflow/rocm-mi355","identify whether block is CI, merge conflict, or review gate",2026-07-07T14:16:38Z
pr,187315,[xpu][3/4] inductor: route MX scaled_mm_v2 fallback through v2 aten kernel,https://github.com/pytorch/pytorch/pull/187315,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/b200, ciflow/h100, ciflow/inductor, ciflow/rocm-mi300","identify whether block is CI, merge conflict, or review gate",2026-07-07T10:42:24Z
pr,187931,[XPU Graph] Refine XPU graph capture detection in XPUCachingAllocator,https://github.com/pytorch/pytorch/pull/187931,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-07-07T06:47:39Z
pr,187549,[xpu][test]testing: enable tf32_on_and_off() for XPU devices,https://github.com/pytorch/pytorch/pull/187549,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/xpu,"identify whether block is CI, merge conflict, or review gate",2026-07-07T05:13:31Z
pr,187876,Fix mark dynamic half bounded range,https://github.com/pytorch/pytorch/pull/187876,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-07T03:07:05Z
pr,188050,Add FX graph cache custom guard replay test,https://github.com/pytorch/pytorch/pull/188050,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2026-07-07T02:04:37Z
pr,189107,Introduce Inductor lowering for _philox_uniform,https://github.com/pytorch/pytorch/pull/189107,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T01:54:01Z
pr,188000,Vectorize LayerNormSecondPass null affine cases,https://github.com/pytorch/pytorch/pull/188000,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-07-07T01:53:56Z
pr,188312,[cpu] Address clang-tidy warnings in vec(256/512)_qint.h,https://github.com/pytorch/pytorch/pull/188312,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-07-07T01:53:50Z
pr,182199,[xpu][test] Port distributed _shard tests cases on Intel GPUs,https://github.com/pytorch/pytorch/pull/182199,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/xpu,"identify whether block is CI, merge conflict, or review gate",2026-07-07T01:48:19Z
pr,180608,[5.5/11][aoti] Add V2 runtime test via ctypes,https://github.com/pytorch/pytorch/pull/180608,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2026-07-07T01:08:42Z
pr,180885,CUDA: unrolled kernels for upsample_linear1d forward/backward when grid is small,https://github.com/pytorch/pytorch/pull/180885,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-07-07T01:08:40Z
pr,180283,[cuBLAS][cuBLASLt] Thread-local cuBLAS(Lt) workspace maps,https://github.com/pytorch/pytorch/pull/180283,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/b200, ciflow/h100, ciflow/h100-symm-mem, ciflow/inductor","identify whether block is CI, merge conflict, or review gate",2026-07-07T00:07:43Z
pr,187891,device_mesh: fall back to new_group when parent backend lacks split s…,https://github.com/pytorch/pytorch/pull/187891,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/torchtitan,"identify whether block is CI, merge conflict, or review gate",2026-07-06T23:07:52Z
pr,182183,[PT2][export] Add unsafe_allow_callable_serialization for scoped callable serialization,https://github.com/pytorch/pytorch/pull/182183,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-06T22:51:30Z
pr,183759,[Testing] Expansion of non-contiguous tensor ops testing to cover more complicated and real-world applications,https://github.com/pytorch/pytorch/pull/183759,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/mps, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-06T22:34:52Z
pr,188451,[cpu] Address clang-tidy warnings in vec128_float_neon.h,https://github.com/pytorch/pytorch/pull/188451,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-07-06T20:53:05Z
pr,188311,[cpu] Address clang-tidy warnings in vec(256/512)_int.h,https://github.com/pytorch/pytorch/pull/188311,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-07-06T20:52:58Z
pr,187675,"[inductor] Add fuse_region to set fusions,memory reuse boundaries",https://github.com/pytorch/pytorch/pull/187675,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-06T20:29:48Z
pr,188571,C++ meta kernel for torch.cat (symint aware),https://github.com/pytorch/pytorch/pull/188571,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2026-07-06T20:27:56Z
pr,152806,[invoke_subgraph] Force the output stride to be same as eager,https://github.com/pytorch/pytorch/pull/152806,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-06T20:13:35Z
pr,187778,Flatten all_to_all_nd copy to fix narrow-row throughput,https://github.com/pytorch/pytorch/pull/187778,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/h100-symm-mem, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-06T20:13:04Z
pr,183705,[SymmMem] Normalize device index in empty_strided_p2p (#183705),https://github.com/pytorch/pytorch/pull/183705,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/h100-symm-mem, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-06T20:12:51Z
pr,186370,inductor: validate square mix-order reductions by common reads,https://github.com/pytorch/pytorch/pull/186370,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-06T20:12:50Z
pr,185109,Remove custom FindCUDAToolkit.cmake module,https://github.com/pytorch/pytorch/pull/185109,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/binaries, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-06T20:12:49Z
pr,187015,remove spmd_types dependency from nightly wheel,https://github.com/pytorch/pytorch/pull/187015,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-07-06T20:12:18Z
pr,189062,[profiler][cupti] Make CUPTI monitor buffer_size / flush_period_s config-only,https://github.com/pytorch/pytorch/pull/189062,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-07-06T20:11:27Z
pr,187007,Use distinct cache variables for linker option checks,https://github.com/pytorch/pytorch/pull/187007,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-07-06T20:10:52Z
pr,187009,Apply libnop clang warning suppression to TensorPipe subtargets,https://github.com/pytorch/pytorch/pull/187009,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-07-06T20:10:50Z
pr,187961,[inductor] Add compile-time per-subkernel autotuning for combo kernels,https://github.com/pytorch/pytorch/pull/187961,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-06T17:18:21Z
pr,176293,[CI] aarch64 manywheel and cuda image build fixes,https://github.com/pytorch/pytorch/pull/176293,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/binaries, ciflow/docker, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-06T15:59:45Z
pr,188822,[Inductor][Host-side TMA] Support tensordesc<> in the static Triton launcher,https://github.com/pytorch/pytorch/pull/188822,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-06T15:48:29Z
pr,180208,Update slow tests,https://github.com/pytorch/pytorch/pull/180208,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/slow, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-06T14:46:19Z
pr,189005,fix(fx): filter internal frames in record_stack_traces to resolve broken stack traces (#130861),https://github.com/pytorch/pytorch/pull/189005,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-07-06T14:32:35Z
pr,183276,[Inductor][Refactor] 2/N Introduce shared heuristic registry with codegen_heuristic support,https://github.com/pytorch/pytorch/pull/183276,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/b200, ciflow/h100, ciflow/inductor, ciflow/rocm-mi300","identify whether block is CI, merge conflict, or review gate",2026-07-06T12:16:15Z
pr,183277,[Inductor][Refactor] 3/N Move pointwise heuristics from triton_heuristics.py into triton_codegen/,https://github.com/pytorch/pytorch/pull/183277,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/b200, ciflow/h100, ciflow/inductor, ciflow/rocm-mi300","identify whether block is CI, merge conflict, or review gate",2026-07-06T12:16:14Z
pr,182577,[Inductor] Enable combo_kernel_per_subkernel_blocks by default,https://github.com/pytorch/pytorch/pull/182577,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-06T09:15:13Z
pr,184955,[Inductor] Exclude indirect-indexing nodes from combo kernels,https://github.com/pytorch/pytorch/pull/184955,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-06T09:15:06Z
pr,185797,[Test] Refactor test/test_bmm_outer_product.py,https://github.com/pytorch/pytorch/pull/185797,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-07-06T08:27:23Z
pr,182729,[DTensor] Preserve placement for aten.clone.default,https://github.com/pytorch/pytorch/pull/182729,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/dtensor, ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-06T01:09:44Z
pr,182727,[shard prop] P(*).clone()->P(*),https://github.com/pytorch/pytorch/pull/182727,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/dtensor, ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-06T01:09:43Z
pr,188067,[ROCm] fix wait instructions,https://github.com/pytorch/pytorch/pull/188067,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-07-05T23:05:01Z
pr,182138,Update torchcomms.txt commit hash,https://github.com/pytorch/pytorch/pull/182138,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/h100-distributed, ciflow/inductor, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-05T20:49:09Z
pr,188517,Fix/180346 dtensor clip grad mixed mesh,https://github.com/pytorch/pytorch/pull/188517,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-07-05T20:36:01Z
pr,188516,Fix/187177 localresponsenorm validation,https://github.com/pytorch/pytorch/pull/188516,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-07-05T20:36:00Z
pr,182530,index_add_/index_reduce_: add env-var kill switch for VecSize=4 vectorized path,https://github.com/pytorch/pytorch/pull/182530,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-07-05T18:51:23Z
pr,170571,[CUDA][cuBLASLt] addmm - Lt path improvements,https://github.com/pytorch/pytorch/pull/170571,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/b200, ciflow/h100, ciflow/inductor, ciflow/nightly","identify whether block is CI, merge conflict, or review gate",2026-07-05T15:53:02Z
pr,181743,Remove skipXPU from test_binary_ufuncs,https://github.com/pytorch/pytorch/pull/181743,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/trunk, ciflow/xpu","identify whether block is CI, merge conflict, or review gate",2026-07-05T14:59:27Z
pr,187897,fix: Add explicit null-buffer guards and shared-memory fallback for Intel Macs with discrete GPU (MPS),https://github.com/pytorch/pytorch/pull/187897,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/mps, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-05T14:29:19Z
pr,187550,[test][inductor] Fix stale `xfail` and add SM12 skip.,https://github.com/pytorch/pytorch/pull/187550,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2026-07-05T09:39:14Z
pr,186906,Update trace_rules for torch.accelerator,https://github.com/pytorch/pytorch/pull/186906,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/dynamo, ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-04T23:59:10Z
pr,187061,[cuda graphs] Automatic nn.Module FQN annotation in Inductor cudagraph trees,https://github.com/pytorch/pytorch/pull/187061,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-04T23:34:08Z
pr,182529,Use NCCLCommProvider for TorchComm + NCCL symmetric memory,https://github.com/pytorch/pytorch/pull/182529,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/h100-symm-mem, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-04T20:49:04Z
pr,183776,C++ FakeTensor,https://github.com/pytorch/pytorch/pull/183776,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/docker, ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-04T13:07:15Z
pr,187732,[DDP] Fix static_graph=True + no_sync() gradient accumulation regression,https://github.com/pytorch/pytorch/pull/187732,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-07-04T11:57:08Z
pr,178500,[autograd] Support Mapping and OrderedDict inputs for autograd.grad and autograd.backward,https://github.com/pytorch/pytorch/pull/178500,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-04T08:36:05Z
pr,182134,[dynamo_wrapped] Remove stale dynamo_expected_failures for torch_np error-path tests,https://github.com/pytorch/pytorch/pull/182134,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-07-04T07:27:38Z
pr,182362,[PyTorch][AOTI] Add timing instrumentation to AOTI model loading pipeline,https://github.com/pytorch/pytorch/pull/182362,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2026-07-03T23:50:44Z
pr,182310,Add Triton scaled-mm v2 blockwise recipes,https://github.com/pytorch/pytorch/pull/182310,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/b200, ciflow/h100, ciflow/inductor, ciflow/rocm-mi300","identify whether block is CI, merge conflict, or review gate",2026-07-03T22:50:18Z
pr,182146,[caffe2] Add defensive fallback in get_logging_handler for unknown destinations (#182146),https://github.com/pytorch/pytorch/pull/182146,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/torchtitan,"identify whether block is CI, merge conflict, or review gate",2026-07-03T20:59:18Z
pr,176239,[dtensor][partial] just_knob error message for fixing .item() on partials,https://github.com/pytorch/pytorch/pull/176239,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-03T20:59:09Z
pr,188646,[inductor] Add tile_n=192 and missing cluster configs to NVGEMM supplement set,https://github.com/pytorch/pytorch/pull/188646,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-03T20:15:05Z
pr,180083,[FSDP] Make FSDP v1 tests device-agnostic for PrivateUse1 backends,https://github.com/pytorch/pytorch/pull/180083,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/h100-symm-mem,"identify whether block is CI, merge conflict, or review gate",2026-07-03T16:59:22Z
pr,187691,[BE]: Update fmtlib to 12.2.0,https://github.com/pytorch/pytorch/pull/187691,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-07-03T16:42:16Z
pr,188789,[xpu] Add IPC memory handle sharing support to XPUCachingAllocator,https://github.com/pytorch/pytorch/pull/188789,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/trunk, ciflow/xpu","identify whether block is CI, merge conflict, or review gate",2026-07-03T11:56:08Z
pr,183323,Adding conversion from PyObject to torch::stable::tensor,https://github.com/pytorch/pytorch/pull/183323,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/docker,"identify whether block is CI, merge conflict, or review gate",2026-07-03T11:00:36Z
pr,181727,[xpu][2/4]Implement scaled_mm_v1 for MXFP8/MXFP4/NVFP4 on XPU,https://github.com/pytorch/pytorch/pull/181727,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-07-03T07:59:30Z
pr,186902,Generalize dynamo RNG-related part to be device-agnostic,https://github.com/pytorch/pytorch/pull/186902,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-03T07:06:47Z
pr,185115,[Test] Enable TestSDPAGpuOnly on XPU and skip cases unsupported by torch-xpu-ops,https://github.com/pytorch/pytorch/pull/185115,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/xpu,"identify whether block is CI, merge conflict, or review gate",2026-07-03T02:42:08Z
pr,175472,Add FA4 fp8 implementation to SDPA,https://github.com/pytorch/pytorch/pull/175472,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-07-03T01:01:47Z
pr,182920,Add backward pass for cuDNN FP8 per-tensor attention,https://github.com/pytorch/pytorch/pull/182920,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2026-07-03T01:01:35Z
pr,179224,spaceship operator for comparisons in Dict.h,https://github.com/pytorch/pytorch/pull/179224,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-07-03T00:20:37Z
pr,188439,[CUDA] [CI] fix potential unit test failure after cuda graph xfail (resource leak),https://github.com/pytorch/pytorch/pull/188439,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-07-02T23:15:38Z
pr,188827,[AI Codemod][PerfAICT-General] Use SymDimVector to avoid heap allocs in select_symint,https://github.com/pytorch/pytorch/pull/188827,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-07-02T20:29:49Z
pr,188826,[AI Codemod][PerfAICT-General] Use SymDimVector to avoid heap allocs in select_symint,https://github.com/pytorch/pytorch/pull/188826,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-07-02T19:51:26Z
pr,188502,Fix/188034 autograd setup context leak,https://github.com/pytorch/pytorch/pull/188502,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-07-02T18:49:55Z
pr,179436,Run clang-tidy with readability-container-contains,https://github.com/pytorch/pytorch/pull/179436,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/h100-symm-mem, ciflow/inductor, ciflow/trunk, ciflow/vllm","identify whether block is CI, merge conflict, or review gate",2026-07-02T18:37:45Z
pr,173390,Add interactive glossary with hover tooltips,https://github.com/pytorch/pytorch/pull/173390,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-07-02T18:18:30Z
pr,188755,cudagraph_trees: honor mark_static_address outputs,https://github.com/pytorch/pytorch/pull/188755,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-02T16:46:48Z
pr,187860,fix DDE in meta__transformer_encoder_layer_fwd,https://github.com/pytorch/pytorch/pull/187860,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-07-02T16:39:26Z
pr,188771,[inductor] Rewind block_ptr across outer loop in nested reductions (#188771),https://github.com/pytorch/pytorch/pull/188771,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-02T16:35:35Z
pr,186599,Introduce isInitialized&set_rng_state in torch.accelerator,https://github.com/pytorch/pytorch/pull/186599,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-07-02T14:23:58Z
pr,181041,[hammer] Make HSTU manual recompute policies compatible with PT2 compile,https://github.com/pytorch/pytorch/pull/181041,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-02T14:16:47Z
pr,174323,[effectful ops] get tokens working with AC,https://github.com/pytorch/pytorch/pull/174323,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2026-07-02T14:16:33Z
pr,170396,[typing] Fix return type annotations for various functions,https://github.com/pytorch/pytorch/pull/170396,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-02T11:35:51Z
pr,185114,[Test] Refractor TestSDPACudaOnly into TestSDPAGpuOnly with CUDA-only assumptions removed,https://github.com/pytorch/pytorch/pull/185114,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-07-02T07:39:39Z
pr,185113,[Test] Decouple test_transformers.py from specific hardware backends,https://github.com/pytorch/pytorch/pull/185113,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-07-02T07:39:38Z
pr,187938,[Inductor] Support non-Triton fused kernels in the FX backend via a high-order-operator,https://github.com/pytorch/pytorch/pull/187938,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-02T07:07:50Z
pr,180050,[FSDPV2] Make FSDP v2 and composable tests device-agnostic for PrivateUse1 backends,https://github.com/pytorch/pytorch/pull/180050,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/h100-symm-mem,"identify whether block is CI, merge conflict, or review gate",2026-07-02T05:49:23Z
pr,183306,[AI Codemod][ClaudeCodeFbandroidWarningsAsErrors] [3] Bulk run of CodemodConfigClaudeCodeFbandroidWarningsAsErrors (10 units) (#183306),https://github.com/pytorch/pytorch/pull/183306,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-07-02T04:50:02Z
pr,186122,[ROCm] Enable test_inverse_errors_large & test_lu_solve_batched_many_batches linalg tests,https://github.com/pytorch/pytorch/pull/186122,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/rocm-mi200, ciflow/rocm-mi300, ciflow/rocm-nightly, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-02T03:29:51Z
pr,186771,Introduce _isInBadFork and refine torch.random,https://github.com/pytorch/pytorch/pull/186771,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-02T01:50:27Z
pr,186816,Remove deprecated torch.chain_matmul,https://github.com/pytorch/pytorch/pull/186816,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-02T01:47:46Z
pr,186899,[2/N] Replace module-level RNG-related APIs with torch.accelerator,https://github.com/pytorch/pytorch/pull/186899,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/dtensor, ciflow/inductor, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-02T01:39:12Z
pr,186968,dynamo: enforce target_values is None|Sequence.,https://github.com/pytorch/pytorch/pull/186968,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-02T01:26:05Z
pr,188744,[dynamo] fix lazy module initialization with symbolic shapes,https://github.com/pytorch/pytorch/pull/188744,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-02T01:18:24Z
pr,188745,"[dynamo, nested graph breaks] fix empty nn.Module hook dict reconstruction",https://github.com/pytorch/pytorch/pull/188745,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-02T01:17:42Z
pr,188581,[dynamo] Add polyfills for operator.contains and operator.call,https://github.com/pytorch/pytorch/pull/188581,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-02T00:52:13Z
pr,185700,[OpenReg] Add compiler integration doc for OOT accelerators,https://github.com/pytorch/pytorch/pull/185700,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-07-02T00:42:37Z
pr,175011,[S615805][RecSysCheckpointing][PyTorch] Add strides to TensorProperties (#175011),https://github.com/pytorch/pytorch/pull/175011,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-02T00:10:07Z
pr,188616,[AI Codemod][PerfAICT-General] Skip per-operand has_storage scan for non-PrivateUse1 devices in build,https://github.com/pytorch/pytorch/pull/188616,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-07-02T00:09:49Z
pr,188563,[AI Codemod][PerfAICT-General] Enable auto-vectorization of at::vec::convert via __restrict and indexed access,https://github.com/pytorch/pytorch/pull/188563,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-07-01T21:46:16Z
pr,188165,"[cutedsl] Pass current compute capability e.g., SM 10.3 -> `sm_103a` to CuTeDSL compile subprocess",https://github.com/pytorch/pytorch/pull/188165,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-01T05:14:17Z
pr,187224,Duplicate concurrent alignment fixups,https://github.com/pytorch/pytorch/pull/187224,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-01T05:13:59Z
pr,178710,Add native torch.cuda.gds.save/load for GPUDirect Storage,https://github.com/pytorch/pytorch/pull/178710,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-07-01T03:50:58Z
pr,188316,[ROCm] Cap HIP stream pool per priority; add reservable streams,https://github.com/pytorch/pytorch/pull/188316,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/torchtitan,"identify whether block is CI, merge conflict, or review gate",2026-06-30T21:58:01Z
pr,187205,Add get_device_name and memory snapshot accelerator APIs,https://github.com/pytorch/pytorch/pull/187205,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-06-30T21:57:59Z
pr,176894,[cuDNN][Convolution] Expose cuDNN `allow_reduced_precision_reduction` numerical note to fix `float16` tests,https://github.com/pytorch/pytorch/pull/176894,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-06-30T20:13:21Z
pr,188501,Fix/147170 deterministic partition ordering,https://github.com/pytorch/pytorch/pull/188501,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-06-30T19:03:15Z
pr,188498,Fix/grammar with the same parameters,https://github.com/pytorch/pytorch/pull/188498,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-06-30T19:02:00Z
pr,188532,Fix/threshold inplace validation,https://github.com/pytorch/pytorch/pull/188532,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-06-30T19:01:00Z
pr,188499,Fix/typo paramater,https://github.com/pytorch/pytorch/pull/188499,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/dtensor, ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-30T18:54:09Z
pr,188339,Fix/typo overridden applier,https://github.com/pytorch/pytorch/pull/188339,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-30T17:56:08Z
pr,188465,[Doc] Document GIL and allocation constraints for PU1 custom allocators,https://github.com/pytorch/pytorch/pull/188465,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-06-30T15:36:37Z
pr,182051,[docs] Link CUDA stream semantics from Tensor.to/copy_ non_blocking docstrings,https://github.com/pytorch/pytorch/pull/182051,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-06-30T15:20:48Z
pr,182986,[Eager] [Bitwise-Equivalence] Implement an INNER_TREE sum reduction (#182986),https://github.com/pytorch/pytorch/pull/182986,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-06-30T08:48:01Z
pr,188535,[test stats] Sanitize empty-file pytest rows at the upload boundary,https://github.com/pytorch/pytorch/pull/188535,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-06-30T08:47:24Z
pr,180907,[shard prop] tag more ops,https://github.com/pytorch/pytorch/pull/180907,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/dtensor, ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-30T05:55:38Z
pr,187150,Remove unused CUDA 12.4 docker build paths,https://github.com/pytorch/pytorch/pull/187150,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/docker, ciflow/inductor-pallas","identify whether block is CI, merge conflict, or review gate",2026-06-30T01:41:04Z
pr,181944,[cuBLAS][cuBLASLt] Shared default workspace size across cuBLAS and Lt,https://github.com/pytorch/pytorch/pull/181944,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/b200, ciflow/h100, ciflow/trunk, ciflow/vllm","identify whether block is CI, merge conflict, or review gate",2026-06-29T21:04:40Z
pr,188444,[Test] Refactor tensor_creation_ops and foreach.py to be device-agnostic.,https://github.com/pytorch/pytorch/pull/188444,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-06-29T21:04:11Z
pr,188340,Fix/indices or section typo,https://github.com/pytorch/pytorch/pull/188340,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-06-29T21:02:55Z
pr,188296,linear_cross_entropy: support probability targets with reduction='none',https://github.com/pytorch/pytorch/pull/188296,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/mps, ciflow/rocm-mi300, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-29T21:02:34Z
pr,188341,Fix/184845 pad docs,https://github.com/pytorch/pytorch/pull/188341,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-06-29T21:02:09Z
pr,188385,[fx] _iterate_exprs: treat device/dtype/layout/memory_format as symbol-free,https://github.com/pytorch/pytorch/pull/188385,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2026-06-29T20:24:36Z
pr,182011,[WIP][PoC] Enable Arm Optimized Routines for AArch64+Linux,https://github.com/pytorch/pytorch/pull/182011,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/pull, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-29T19:13:21Z
pr,188294,Fix GradientEdge None-grad error message formatting,https://github.com/pytorch/pytorch/pull/188294,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-06-29T16:27:59Z
pr,182851,[dynamo] Fix free-threading race in dict recursive-tag watcher callback,https://github.com/pytorch/pytorch/pull/182851,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/tsan","identify whether block is CI, merge conflict, or review gate",2026-06-29T16:27:23Z
pr,188295,Validate queue_callback input is callable,https://github.com/pytorch/pytorch/pull/188295,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-06-29T16:27:19Z
pr,187922,[TEST] Refactor test_embedding.py with hw_classification,https://github.com/pytorch/pytorch/pull/187922,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-06-29T03:35:32Z
pr,175809,[cuDNN] Migrate to new frontend API for batchnorm,https://github.com/pytorch/pytorch/pull/175809,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-06-29T01:29:15Z
pr,177924,[triton hash update] update the pinned triton hash,https://github.com/pytorch/pytorch/pull/177924,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-28T23:52:32Z
pr,178283,Bump torch_tpu pin,https://github.com/pytorch/pytorch/pull/178283,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/inductor-pallas","identify whether block is CI, merge conflict, or review gate",2026-06-28T20:49:52Z
pr,173914,[CI] [CUDA] Add CUDA ARM CI image,https://github.com/pytorch/pytorch/pull/173914,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-06-28T20:49:51Z
pr,179820,[ATen][Native][CUDA] Use dynamic loading in CUDNN FrontEnd,https://github.com/pytorch/pytorch/pull/179820,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/binaries,"identify whether block is CI, merge conflict, or review gate",2026-06-28T19:57:15Z
pr,187711,Strict pyrefly checking: round 1 (2/N),https://github.com/pytorch/pytorch/pull/187711,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2026-06-28T19:22:51Z
pr,181767,"[shard prop] autogen infra for inplace,out variants",https://github.com/pytorch/pytorch/pull/181767,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/dtensor, ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-27T22:50:21Z
pr,181724,S390X: update binaries labels,https://github.com/pytorch/pytorch/pull/181724,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-06-27T18:54:05Z
pr,181623,[torch.library] Add pallas_op for JAX/Pallas kernel registration,https://github.com/pytorch/pytorch/pull/181623,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-06-27T18:54:01Z
pr,174771,Fix `sample_kwargs` in `polygamma` op info,https://github.com/pytorch/pytorch/pull/174771,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-06-27T14:08:20Z
pr,187390,Widen aot_autograd annotations to match runtime values,https://github.com/pytorch/pytorch/pull/187390,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2026-06-27T09:18:37Z
pr,187692,Skip mobilenetv2_100 eager-nondeterministic accuracy check,https://github.com/pytorch/pytorch/pull/187692,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-27T09:18:31Z
pr,181530,[caffe2] Fix HIP build: cccl_counting_iterator needs explicit type,https://github.com/pytorch/pytorch/pull/181530,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-06-27T07:33:50Z
pr,178906,Fix batch_isend_irecv to support ProcessGroup subclasses,https://github.com/pytorch/pytorch/pull/178906,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-27T07:33:45Z
pr,188161,Demote changing CUDAGraph-managed inputs to copied inputs,https://github.com/pytorch/pytorch/pull/188161,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-27T02:16:25Z
pr,188103,Enable recursive dict-tag guard optimization by default on 3.12+ non-free-threaded,https://github.com/pytorch/pytorch/pull/188103,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-27T02:15:46Z
pr,188101,Require an installed dict watcher for the recursive dict-tag fast path on 3.12+,https://github.com/pytorch/pytorch/pull/188101,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2026-06-27T02:15:44Z
pr,181414,Update FBGEMM submodule to v1.6.0,https://github.com/pytorch/pytorch/pull/181414,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/b200, ciflow/h100, ciflow/rocm-mi300","identify whether block is CI, merge conflict, or review gate",2026-06-27T01:21:39Z
pr,188298,[aot_autograd] Add PySourceBuilder for runtime-wrapper codegen,https://github.com/pytorch/pytorch/pull/188298,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2026-06-26T23:49:39Z
pr,180990,[pytorch] Fix symmetric quantization scale to match C++ ChooseQuantizationParams (#180990),https://github.com/pytorch/pytorch/pull/180990,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-06-26T22:53:26Z
pr,188289,[mtia] Make torch.distributions test suite portable across acelerators,https://github.com/pytorch/pytorch/pull/188289,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-06-26T22:23:50Z
pr,188215,[dtensor] Resolve ProcessGroup from mesh in-graph under compile_on_one_rank,https://github.com/pytorch/pytorch/pull/188215,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/dtensor, ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-26T19:03:04Z
pr,188093,Fix record_stack_traces: filter fx-internal frames from recorded trac…,https://github.com/pytorch/pytorch/pull/188093,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-06-26T18:32:23Z
pr,188180,Reuse full-range splits for persistent CSE,https://github.com/pytorch/pytorch/pull/188180,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-26T18:09:09Z
pr,187875,[AOTI] Refactor model container checkout ownership (#187875),https://github.com/pytorch/pytorch/pull/187875,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-26T17:30:50Z
pr,188063,Enable Blackwell FP8 grouped GEMM,https://github.com/pytorch/pytorch/pull/188063,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/b200, ciflow/h100, ciflow/rocm-mi300","identify whether block is CI, merge conflict, or review gate",2026-06-26T16:56:20Z
pr,182948,[CUDA] Fix reentrant deadlock in torch.cuda._lazy_call,https://github.com/pytorch/pytorch/pull/182948,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/inductor-rocm, ciflow/rocm-mi300, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-26T15:20:31Z
pr,178751,Delay common_utils import to pytest_sessionstart to prevent faulthandler race condition,https://github.com/pytorch/pytorch/pull/178751,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/trunk, ciflow/xpu","identify whether block is CI, merge conflict, or review gate",2026-06-26T14:18:26Z
pr,186243,"[dynamo, nested graph breaks] enable NGB on dynamo-wrapped tests",https://github.com/pytorch/pytorch/pull/186243,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/dynamo, ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-26T03:29:55Z
pr,181398,[spmd_types] local_map,https://github.com/pytorch/pytorch/pull/181398,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/dtensor, ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-26T03:29:20Z
pr,188188,[DTensor] local_map normalizes shard dim checking in_placements,https://github.com/pytorch/pytorch/pull/188188,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/dtensor, ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-25T23:48:01Z
pr,186926,Remove sleef from test_aoti_abi_check linkage,https://github.com/pytorch/pytorch/pull/186926,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2026-06-25T22:47:19Z
pr,186947,[functorch] Add native_group_norm_backward.multiple_grads plumbing,https://github.com/pytorch/pytorch/pull/186947,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-25T22:08:41Z
pr,188179,Use tiling-aware hints for persistent reductions,https://github.com/pytorch/pytorch/pull/188179,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-25T22:05:36Z
pr,186946,[mps] Add kernel for native_group_norm_backward.multiple_grads,https://github.com/pytorch/pytorch/pull/186946,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/mps, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-25T21:30:26Z
pr,186414,"native_group_norm: Handle non-contiguous tensors, rather than throwing",https://github.com/pytorch/pytorch/pull/186414,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-25T20:54:28Z
pr,184568,Always reset Dynamo if it was imported,https://github.com/pytorch/pytorch/pull/184568,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-25T20:14:25Z
pr,175242,Improve FX partitioner complexity,https://github.com/pytorch/pytorch/pull/175242,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-06-25T18:44:25Z
pr,178318,[Inductor] Add xmask unswitch for dynamic pointwise kernels,https://github.com/pytorch/pytorch/pull/178318,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-25T13:15:40Z
pr,181307,Set device in forward input tests,https://github.com/pytorch/pytorch/pull/181307,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-06-25T11:51:05Z
pr,187784,[TokenSwitch] Add expert-major dispatch layout,https://github.com/pytorch/pytorch/pull/187784,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/h100-symm-mem, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-25T10:06:26Z
pr,186129,[dtensor] fix conv1d output placement and group scaling,https://github.com/pytorch/pytorch/pull/186129,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/dtensor, ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-25T08:14:44Z
pr,186877,Use C++20 std::numbers in c10 MathConstants,https://github.com/pytorch/pytorch/pull/186877,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/trunk, ciflow/vllm","identify whether block is CI, merge conflict, or review gate",2026-06-25T07:44:16Z
pr,186211,[Inductor] Skip nonfinite preservation for safe scaled softmax,https://github.com/pytorch/pytorch/pull/186211,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-25T03:16:55Z
pr,176920,Add distributed runtime autotuning for Triton kernels,https://github.com/pytorch/pytorch/pull/176920,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-25T02:16:53Z
pr,184924,[Test] [CUDA] Skip test_variant_consistency_jit for grid_sampler_2d,https://github.com/pytorch/pytorch/pull/184924,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-06-24T21:45:19Z
pr,184824,[Reland] Port D104346887/PR 182675 for index_add fast path (#184824),https://github.com/pytorch/pytorch/pull/184824,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-06-24T21:08:00Z
pr,188094,[profiler][cupti] Remove unused Python reference decoder (CuptiMonitorBuffer),https://github.com/pytorch/pytorch/pull/188094,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-06-24T21:07:35Z
pr,187797,Add linalg SVD operator benchmarks,https://github.com/pytorch/pytorch/pull/187797,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/mps,"identify whether block is CI, merge conflict, or review gate",2026-06-24T20:34:10Z
pr,186777,[1/N] Replace module-level RNG-related APIs with torch.accelerator,https://github.com/pytorch/pytorch/pull/186777,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/dtensor, ciflow/inductor, ciflow/mps, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-24T19:04:55Z
pr,187208,[BE][Ez]: Use CPP20 rvalue overload for ostringstream,https://github.com/pytorch/pytorch/pull/187208,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/h100-symm-mem, ciflow/inductor, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-24T18:46:10Z
pr,187823,Fix regex for extracting gist URLs from logs,https://github.com/pytorch/pytorch/pull/187823,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2026-06-24T18:35:16Z
pr,186600,Introduce manual_seed&seed to torch.accelerator,https://github.com/pytorch/pytorch/pull/186600,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-06-24T18:20:51Z
pr,185213,[Test] Make dynamo test_streams.py device-agnostic for out-of-tree backends,https://github.com/pytorch/pytorch/pull/185213,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2026-06-24T16:23:50Z
pr,188058,[inductor] Avoid disjoint horizontal fusion when allow_peak_memory_increasing_fusion is off,https://github.com/pytorch/pytorch/pull/188058,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-24T16:15:51Z
pr,188056,[inductor] Avoid disjoint horizontal fusion when allow_peak_memory_increasing_fusion is off,https://github.com/pytorch/pytorch/pull/188056,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-24T16:15:45Z
pr,188059,[inductor] Block only peak-crossing fusions in the peak-memory guards,https://github.com/pytorch/pytorch/pull/188059,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-24T16:15:42Z
pr,187842,[inductor] Avoid cross-extern fusion when allow_peak_memory_increasing_fusion is off,https://github.com/pytorch/pytorch/pull/187842,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-24T15:24:18Z
pr,188055,[inductor] Evaluate compile-time ProcessGroup attrs in overlap estimates,https://github.com/pytorch/pytorch/pull/188055,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-24T15:24:11Z
pr,188057,[inductor] Evaluate compile-time ProcessGroup attrs in overlap estimates,https://github.com/pytorch/pytorch/pull/188057,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-24T15:24:08Z
pr,187832,[FX][Dynamo] Add tests for nn_module_stack and source_fn_stack metadata contracts,https://github.com/pytorch/pytorch/pull/187832,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2026-06-24T15:23:57Z
pr,184867,[symbolic shapes] Avoid direct replacement cycles,https://github.com/pytorch/pytorch/pull/184867,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2026-06-24T14:47:23Z
pr,187619,Remove Dynamo guard from custom op kernels,https://github.com/pytorch/pytorch/pull/187619,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-06-24T14:39:30Z
pr,181396,Print and validate full CUDA version in smoke test,https://github.com/pytorch/pytorch/pull/181396,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/binaries,"identify whether block is CI, merge conflict, or review gate",2026-06-24T12:19:52Z
pr,180455,"[AARCH64] Enable MKLDNN backend for AArch64 for INT8 Matmul, using torch op _int_mm.dtype",https://github.com/pytorch/pytorch/pull/180455,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/linux-aarch64, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-24T06:45:05Z
pr,187471,[inductor][mtia] Strip AMD-only autotune kwargs from MTIA Triton compile configs (#187471),https://github.com/pytorch/pytorch/pull/187471,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-24T04:33:18Z
pr,180644,Relax align_random_eager test tolerances for Philox float precision,https://github.com/pytorch/pytorch/pull/180644,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2026-06-24T03:54:04Z
pr,182193,Tighten any_index_expr_overflows_int32 overflow check,https://github.com/pytorch/pytorch/pull/182193,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-24T03:01:44Z
pr,187520,[profiler][cupti] Comms monitoring tests,https://github.com/pytorch/pytorch/pull/187520,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-06-24T02:55:15Z
pr,187519,[profiler][cupti] Comms monitoring: CommsObserver + comm hook + plugins (CollTrace replacement),https://github.com/pytorch/pytorch/pull/187519,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-06-24T02:55:14Z
pr,187517,[profiler][cupti] CUDA-graph correlation utils: node-id remap + side-stream events,https://github.com/pytorch/pytorch/pull/187517,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-06-24T02:55:14Z
pr,187518,[profiler][cupti] In-process NCCL metadata plugin (ncclProfiler_v6),https://github.com/pytorch/pytorch/pull/187518,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-06-24T02:55:13Z
pr,186655,[profiler][cupti] Benchmarks: columnar-decode roofline + monitor distortion,https://github.com/pytorch/pytorch/pull/186655,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-06-24T02:55:11Z
pr,187328,Add fusion provenance to profiler timelines,https://github.com/pytorch/pytorch/pull/187328,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-23T22:47:00Z
pr,187367,Fix typos in comments and docstrings,https://github.com/pytorch/pytorch/pull/187367,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-23T22:41:56Z
pr,183468,[Inductor] Fix linter from tf32 fix,https://github.com/pytorch/pytorch/pull/183468,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-23T22:01:56Z
pr,187965,[inductor] Avoid disjoint horizontal fusion when allow_peak_memory_increasing_fusion is off,https://github.com/pytorch/pytorch/pull/187965,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-23T21:34:11Z
pr,187962,[inductor] Evaluate compile-time ProcessGroup attrs in overlap estimates,https://github.com/pytorch/pytorch/pull/187962,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-23T21:34:05Z
pr,187964,[inductor] Evaluate compile-time ProcessGroup attrs in overlap estimates,https://github.com/pytorch/pytorch/pull/187964,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-23T21:33:50Z
pr,187548,Update ck aiter gfx1250,https://github.com/pytorch/pytorch/pull/187548,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/rocm-mi300, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-23T20:56:50Z
pr,185491,Cleanup unused tensor variables,https://github.com/pytorch/pytorch/pull/185491,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/trunk, ciflow/xpu","identify whether block is CI, merge conflict, or review gate",2026-06-23T18:36:01Z
pr,187809,Make DTensor logging and export tests accelerator-agnostic,https://github.com/pytorch/pytorch/pull/187809,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/dtensor,"identify whether block is CI, merge conflict, or review gate",2026-06-23T14:25:23Z
pr,187357,Convert _scaled_grouped_mm_v2 to a structured operator,https://github.com/pytorch/pytorch/pull/187357,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/b200, ciflow/h100, ciflow/rocm-mi300","identify whether block is CI, merge conflict, or review gate",2026-06-23T13:29:24Z
pr,187529,perf(dynamo): add guard lookup token plan,https://github.com/pytorch/pytorch/pull/187529,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2026-06-23T08:56:22Z
pr,187642,Add parameter-contiguous all-gather (symm_mem.all_gather_offset),https://github.com/pytorch/pytorch/pull/187642,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/h100-symm-mem, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-23T03:28:37Z
pr,186880,Add FlexAttention shared memory compile hints,https://github.com/pytorch/pytorch/pull/186880,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-23T01:09:16Z
pr,187750,Tune all_gather_offset for skewed buckets,https://github.com/pytorch/pytorch/pull/187750,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/h100-symm-mem, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-22T23:54:08Z
pr,187776,Speed up all_gather_offset writes with a shared unrolled copy helper,https://github.com/pytorch/pytorch/pull/187776,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/h100-symm-mem, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-22T23:28:03Z
pr,176754,[CUDA Graph] Thread-safe RNG state for concurrent graph captures,https://github.com/pytorch/pytorch/pull/176754,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-06-22T23:27:13Z
pr,181518,[Inductor] enable MSVC precompiled headers for cpp-wrapper,https://github.com/pytorch/pytorch/pull/181518,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-22T23:14:35Z
pr,174395,Add optional out argument to F.scaled_mm,https://github.com/pytorch/pytorch/pull/174395,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/b200, ciflow/h100, ciflow/rocm-mi300, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-22T22:12:29Z
pr,181265,TEMP,https://github.com/pytorch/pytorch/pull/181265,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-22T21:25:13Z
pr,187134,[dynamo] Reuse ACT-traced graphs for resolved tensor inputs,https://github.com/pytorch/pytorch/pull/187134,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-22T19:50:06Z
pr,187785,[TokenSwitch] Add symmetric-memory zero-copy dispatch/combine,https://github.com/pytorch/pytorch/pull/187785,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/h100-symm-mem,"identify whether block is CI, merge conflict, or review gate",2026-06-22T15:16:43Z
pr,181221,[DCP] Fix flaky test_async_process_executor by resetting global state between tests (#181221),https://github.com/pytorch/pytorch/pull/181221,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-06-22T11:23:21Z
pr,181178,[Inductor] Allow max_autotune_gemm on Jetson Orin AGX,https://github.com/pytorch/pytorch/pull/181178,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-22T03:48:42Z
pr,186332,Improved common device test module for xpu,https://github.com/pytorch/pytorch/pull/186332,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/xpu,"identify whether block is CI, merge conflict, or review gate",2026-06-22T02:08:33Z
pr,181150,update copy_misaligned to copy_if_misaligned,https://github.com/pytorch/pytorch/pull/181150,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-21T22:52:20Z
pr,180529,[inductor] Add TORCHINDUCTOR_PERSISTENT_AUTOTUNE_DIR for persistent autotune cache (#180529),https://github.com/pytorch/pytorch/pull/180529,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-21T22:52:19Z
pr,178800,debug server tutorial,https://github.com/pytorch/pytorch/pull/178800,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-21T21:58:30Z
pr,181023,[PP] Support extra loss_fn kwargs in pipeline schedules,https://github.com/pytorch/pytorch/pull/181023,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/torchtitan,"identify whether block is CI, merge conflict, or review gate",2026-06-21T20:56:46Z
pr,181010,[inductor][comms] allow compute to borrow memory from pg_alloc mempool,https://github.com/pytorch/pytorch/pull/181010,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-21T09:48:13Z
pr,181009,[inductor][comms] pg_alloc config for strategies,https://github.com/pytorch/pytorch/pull/181009,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-21T09:48:12Z
pr,181008,[inductor][comms] memory budget for pg allocs,https://github.com/pytorch/pytorch/pull/181008,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-21T09:48:11Z
pr,186137,[dynamo] Swap @disable / @allow_in_graph order on invoke_subgraph_wrapper_unboxed,https://github.com/pytorch/pytorch/pull/186137,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-21T08:51:50Z
pr,187712,Strict pyrefly checking: round 1 (3/N),https://github.com/pytorch/pytorch/pull/187712,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-06-21T07:29:31Z
pr,180535,"[AC remat] Per-region independent recomputation, remove user phase annotations",https://github.com/pytorch/pytorch/pull/180535,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-21T06:14:17Z
pr,180947,[CUDACachingAllocator] Add test for expandable segment OOM stat,https://github.com/pytorch/pytorch/pull/180947,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-06-20T23:52:44Z
pr,180858,[inductor] symm memm ag/rs with cuStreamWrite/WaitValue32 sync,https://github.com/pytorch/pytorch/pull/180858,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/h100-symm-mem, ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-20T14:16:47Z
pr,180930,[Inductor] Lower MAX_COMPLEX_POINTWISE_CAT from 8 to 5,https://github.com/pytorch/pytorch/pull/180930,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-20T01:26:37Z
pr,180665,"[FSDP2] Fix HSDP + CPUOffloadPolicy grad corruption race, add stream-behavior tests",https://github.com/pytorch/pytorch/pull/180665,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-19T23:52:43Z
pr,178338,[inductor] add option to autotune at fusion time for mm,https://github.com/pytorch/pytorch/pull/178338,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-19T08:21:12Z
pr,187062,[inductor] Expose compile_worker_malloc_conf to tune the compile-worker allocator (#187062),https://github.com/pytorch/pytorch/pull/187062,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-19T05:59:00Z
pr,185173,Run cpp/test_jit and cpp/test_lazy sequentially in test_libtorch_jit,https://github.com/pytorch/pytorch/pull/185173,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-06-19T02:38:39Z
pr,174940,[BE] Delete old Metal ops,https://github.com/pytorch/pytorch/pull/174940,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-06-19T01:02:01Z
pr,178804,Merge clang18 Python-version Docker images into one multi-env container,https://github.com/pytorch/pytorch/pull/178804,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-06-19T01:01:58Z
pr,178812,[BE][MPS] Delete stale skips from common_modules.py,https://github.com/pytorch/pytorch/pull/178812,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/mps,"identify whether block is CI, merge conflict, or review gate",2026-06-19T01:01:57Z
pr,180181,[CI] Split benchmark docker builds into three-phase build,https://github.com/pytorch/pytorch/pull/180181,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2026-06-19T01:01:55Z
pr,182038,[MPS] Migrate backward to MPP on MacOS-26,https://github.com/pytorch/pytorch/pull/182038,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/mps,"identify whether block is CI, merge conflict, or review gate",2026-06-19T01:01:51Z
pr,184404,[BE] Exclude `c10/metal` from HEADER_ONLY wraps,https://github.com/pytorch/pytorch/pull/184404,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-06-19T01:01:48Z
pr,186844,Add owner headers to docs/lint/benchmark/vllm workflows,https://github.com/pytorch/pytorch/pull/186844,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-06-19T01:01:41Z
pr,187545,[BE] Use `std::numbers` in `c10/util/MathConstants.h`,https://github.com/pytorch/pytorch/pull/187545,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/docker,"identify whether block is CI, merge conflict, or review gate",2026-06-19T01:01:33Z
pr,187419,Add numerics-debugging skill,https://github.com/pytorch/pytorch/pull/187419,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-06-18T22:27:01Z
pr,182074,[TEST][CUDA] Bump reflection_pad_2d decomp test to use `bfloat16` threshold,https://github.com/pytorch/pytorch/pull/182074,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-06-18T22:06:23Z
pr,181906,"[lint][RAWTHROW][3/5] aten, c10/util, distributed, inductor",https://github.com/pytorch/pytorch/pull/181906,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/docker, ciflow/inductor","identify whether block is CI, merge conflict, or review gate",2026-06-18T19:35:00Z
pr,181904,[lint][RAWTHROW][1/5] Small patterns sweep: annotate + fix 22 exclude groups,https://github.com/pytorch/pytorch/pull/181904,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/docker, ciflow/inductor","identify whether block is CI, merge conflict, or review gate",2026-06-18T19:34:59Z
pr,181907,[lint][RAWTHROW][4/5] Live JIT: python + runtime,https://github.com/pytorch/pytorch/pull/181907,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/docker,"identify whether block is CI, merge conflict, or review gate",2026-06-18T19:28:04Z
pr,181905,"[lint][RAWTHROW][2/5] torch/csrc autograd, utils, profiler, fx",https://github.com/pytorch/pytorch/pull/181905,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/docker,"identify whether block is CI, merge conflict, or review gate",2026-06-18T19:25:32Z
pr,180641,Prune binary size from flash,https://github.com/pytorch/pytorch/pull/180641,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-06-18T17:36:11Z
pr,170679,Fix FileBaton leaving dangling lock files that cause compilation hang…,https://github.com/pytorch/pytorch/pull/170679,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-06-18T12:58:30Z
pr,178150,[12/12] Enforce C++20 minimum in header guards (#178150),https://github.com/pytorch/pytorch/pull/178150,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-18T09:57:27Z
pr,175914,[Export] Fix unflatten crash when mutation intermediate has list args,https://github.com/pytorch/pytorch/pull/175914,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-06-17T21:17:39Z
pr,180442,fix(mps): skip nograph linear on AMD RDNA2 GPUs when input_features>=256,https://github.com/pytorch/pytorch/pull/180442,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/mps,"identify whether block is CI, merge conflict, or review gate",2026-06-17T19:19:22Z
pr,187391,[export][serde] Serialize SymBoolArgument in graph input/output specs (#187391),https://github.com/pytorch/pytorch/pull/187391,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-17T05:10:33Z
pr,187430,Add Kaiser-Bessel Derived window,https://github.com/pytorch/pytorch/pull/187430,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/dtensor, ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-17T05:09:59Z
pr,181167,remove size oblivious reasoning core,https://github.com/pytorch/pytorch/pull/181167,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-17T05:09:46Z
pr,180722,Friendler error messages,https://github.com/pytorch/pytorch/pull/180722,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/b200, ciflow/h100, ciflow/rocm-mi300","identify whether block is CI, merge conflict, or review gate",2026-06-17T03:48:46Z
pr,180528,Add dynamo_timed wrappers for AOT autograd compilation stages (#180528),https://github.com/pytorch/pytorch/pull/180528,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-16T20:31:38Z
pr,184487,dynamo: round-trip torch.cuda.stream ctx mgr across graph breaks,https://github.com/pytorch/pytorch/pull/184487,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-16T19:42:01Z
pr,187369,[Test] Add test hardware classification for test_complex.py and test_kernel_launch_checks.py,https://github.com/pytorch/pytorch/pull/187369,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-06-16T18:51:59Z
pr,183638,[inductor] Fuse NVFP4 nested-reduction packing,https://github.com/pytorch/pytorch/pull/183638,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-16T18:50:10Z
pr,178004,[ROCm] enable compiler-driven loop unrolling,https://github.com/pytorch/pytorch/pull/178004,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/rocm-mi300, ciflow/rocm-mi355, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-16T17:27:05Z
pr,176657,[CUDAGraph] Detect and error on freeing parent-capture tensors inside CUDA graph conditional nodes,https://github.com/pytorch/pytorch/pull/176657,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-06-16T16:35:47Z
pr,179702,[torch][inductor] Add dump_python_module config for AOT single-pass mode (#179702),https://github.com/pytorch/pytorch/pull/179702,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-16T15:36:46Z
pr,178167,triangular_solve_stub: expand leading dim support beyond dense and non-overlapping inputs,https://github.com/pytorch/pytorch/pull/178167,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-06-16T12:23:46Z
pr,181570,Fix batched LU factorization grain_size heuristic for large matrices (#181570),https://github.com/pytorch/pytorch/pull/181570,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-06-16T08:18:26Z
pr,177087,[torch][inductor] Get shape information for extern/Aten kernels with enable_kernel_profile (#177087),https://github.com/pytorch/pytorch/pull/177087,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-16T03:48:30Z
pr,179699,[torch][inductor] Add KernelContextGuard profiling to all kernel codegen paths (#179699),https://github.com/pytorch/pytorch/pull/179699,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-16T00:30:54Z
pr,179587,[pytorch] Fix resource loading for PAR compatibility using importlib.resources (#179587),https://github.com/pytorch/pytorch/pull/179587,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-16T00:05:33Z
pr,186102,[inductor] Fast paths in sizevars for concrete integer comparisons,https://github.com/pytorch/pytorch/pull/186102,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-15T22:52:58Z
pr,186101,[inductor] Avoid O(n) set union in decide_inplace_update,https://github.com/pytorch/pytorch/pull/186101,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-15T22:52:28Z
pr,186097,[inductor] Add early exits to can_fusion_increase_peak_memory,https://github.com/pytorch/pytorch/pull/186097,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-15T22:51:36Z
pr,186096,[inductor] Skip combined tiling computation for <=2D pointwise fusion,https://github.com/pytorch/pytorch/pull/186096,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-15T22:51:32Z
pr,187002,Use C++20 <bit> instead of c10::llvm bit helpers in c10,https://github.com/pytorch/pytorch/pull/187002,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/vllm,"identify whether block is CI, merge conflict, or review gate",2026-06-15T15:31:05Z
pr,187168,Fix accuracy degradation for baddbmm on CPU for float tensors,https://github.com/pytorch/pytorch/pull/187168,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-06-15T11:37:59Z
pr,171219,[DTensor] Support Partial to _StridedShard,https://github.com/pytorch/pytorch/pull/171219,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/dtensor, ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-15T10:36:36Z
pr,171338,[DTensor] Fully support uneven _StridedShard redistribution with device order,https://github.com/pytorch/pytorch/pull/171338,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/dtensor, ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-15T10:36:34Z
pr,171023,[DTensor] Support Shard to _StridedShard through all-to-all,https://github.com/pytorch/pytorch/pull/171023,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/dtensor, ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-15T10:36:33Z
pr,171178,[DTensor] Refactor padding/unpadding with CollectivePaddingContext,https://github.com/pytorch/pytorch/pull/171178,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/dtensor, ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-15T10:36:28Z
pr,186762,Fix typo in quantization documentation,https://github.com/pytorch/pytorch/pull/186762,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-06-15T10:35:42Z
pr,179127,Skip FBGEMM on Arm64,https://github.com/pytorch/pytorch/pull/179127,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-06-15T04:36:48Z
pr,179642,Cache cuSPARSELt matmul plan/descriptors in _cslt_sparse_mm (#179642),https://github.com/pytorch/pytorch/pull/179642,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-06-14T18:55:58Z
pr,178014,add missing to() operator which is called in benchmark (#178014),https://github.com/pytorch/pytorch/pull/178014,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-06-14T17:59:57Z
pr,180183,autograd: handle kMkldnn and kJagged layouts in record_stream_any_impl,https://github.com/pytorch/pytorch/pull/180183,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-06-14T16:59:38Z
pr,180167,Clarify that spectral_norm does not compute the true operator norm for conv layers,https://github.com/pytorch/pytorch/pull/180167,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-06-14T16:59:36Z
pr,187069,[torch][autograd] Fix Python refcount leaks in autograd C++ glue,https://github.com/pytorch/pytorch/pull/187069,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-06-14T16:30:46Z
pr,185958,torch/headeronly/util/TypeList.h: simplify contains and find_if with C++20 fold expressions and if constexpr,https://github.com/pytorch/pytorch/pull/185958,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/trunk, ciflow/vllm","identify whether block is CI, merge conflict, or review gate",2026-06-14T01:33:13Z
pr,185859,ATen/core/TensorBase.h: replace symint enable_if aliases with C++20 requires clauses,https://github.com/pytorch/pytorch/pull/185859,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/trunk, ciflow/vllm","identify whether block is CI, merge conflict, or review gate",2026-06-14T01:33:09Z
pr,174409,[inductor] Add logging to autotune output showing cache hit,https://github.com/pytorch/pytorch/pull/174409,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-13T21:52:52Z
pr,185875,c10/core/DispatchKeySet.h: add noexcept to non-constexpr query methods,https://github.com/pytorch/pytorch/pull/185875,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/trunk, ciflow/vllm","identify whether block is CI, merge conflict, or review gate",2026-06-13T20:33:14Z
pr,187142,Fully type-annotate torch/utils/data/graph_settings.py,https://github.com/pytorch/pytorch/pull/187142,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-06-13T18:40:59Z
pr,180274,[WIP][Triton] update triton pin,https://github.com/pytorch/pytorch/pull/180274,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-13T13:03:15Z
pr,178922,Add CUDA support for Lanczos interpolation mode,https://github.com/pytorch/pytorch/pull/178922,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/dtensor, ciflow/inductor","identify whether block is CI, merge conflict, or review gate",2026-06-13T13:03:11Z
pr,185261,add meta registration for scaled_grouped_mm_v2,https://github.com/pytorch/pytorch/pull/185261,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/b200, ciflow/h100, ciflow/rocm-mi300","identify whether block is CI, merge conflict, or review gate",2026-06-13T01:01:30Z
pr,186190,[spmd_types] typecheck FlexAttention,https://github.com/pytorch/pytorch/pull/186190,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-12T22:48:39Z
pr,186653,Make topk deterministic under deterministic algorithms,https://github.com/pytorch/pytorch/pull/186653,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-12T22:15:52Z
pr,177619,more ops,https://github.com/pytorch/pytorch/pull/177619,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/dtensor, ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-12T22:03:57Z
pr,177618,replicate-only ops,https://github.com/pytorch/pytorch/pull/177618,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/dtensor, ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-12T22:03:55Z
pr,186848,Use C++20 concepts (1 of N),https://github.com/pytorch/pytorch/pull/186848,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/trunk, ciflow/vllm","identify whether block is CI, merge conflict, or review gate",2026-06-12T21:00:38Z
pr,183617,[fsdp2] fix mem accumulation of all reduce input buffers for hsdp with mixed precision,https://github.com/pytorch/pytorch/pull/183617,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/h100-symm-mem, ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-12T20:07:44Z
pr,187189,[DO NOT MERGE][ROCm] Bump default trunk ROCm test shards to 20,https://github.com/pytorch/pytorch/pull/187189,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-06-12T18:47:15Z
pr,186815,Remove deprecated torch.qr,https://github.com/pytorch/pytorch/pull/186815,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-12T18:00:49Z
pr,185366,[ROCm][Distributed] Add synthesized test for sac_milp to ensure numerical correctness,https://github.com/pytorch/pytorch/pull/185366,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/rocm-mi200, ciflow/rocm-mi300, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-12T12:56:54Z
pr,185236,Detect fp32 precision flag leaks in TestCase and fix offenders,https://github.com/pytorch/pytorch/pull/185236,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/b200, ciflow/h100, ciflow/inductor, ciflow/rocm-mi300","identify whether block is CI, merge conflict, or review gate",2026-06-11T18:53:33Z
pr,179257,[c10] Fix off-by-one issue in ApproximateClock.cpp,https://github.com/pytorch/pytorch/pull/179257,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-06-11T17:43:17Z
pr,182964,[inductor][overlap] Wire _low_contention_all_reduce into LC FX pass,https://github.com/pytorch/pytorch/pull/182964,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-11T16:54:07Z
pr,182963,[symm_mem] Add _low_contention_all_reduce to symmetric memory,https://github.com/pytorch/pytorch/pull/182963,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/h100-symm-mem, ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-11T16:53:59Z
pr,181748,[Inductor][Triton] Enable decompose_k and use Triton for inner BMM on HIP (#181748) (#181748),https://github.com/pytorch/pytorch/pull/181748,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/b200, ciflow/h100, ciflow/inductor, ciflow/rocm-mi300","identify whether block is CI, merge conflict, or review gate",2026-06-11T16:51:30Z
pr,178300,[Inductor] Retry Triton compilation without pointer_range_32 on failure,https://github.com/pytorch/pytorch/pull/178300,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-11T16:51:23Z
pr,175450,[inductor] CUDAGraph P2P pool handling for symm_mem,https://github.com/pytorch/pytorch/pull/175450,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/h100-symm-mem, ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-11T16:50:40Z
pr,181007,[inductor][comms] Optional use of process_group alloc_tensor for collectives,https://github.com/pytorch/pytorch/pull/181007,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-11T16:48:36Z
pr,179548,[inductor] Add dedicated thread pool for nogil Triton compilation (#179548) (#179548),https://github.com/pytorch/pytorch/pull/179548,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-11T16:40:50Z
pr,179142,[Inductor] Add runtime assertion support to FX wrapper backend (#179142),https://github.com/pytorch/pytorch/pull/179142,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-11T16:35:11Z
pr,178083,Update inductor expected accuracy files,https://github.com/pytorch/pytorch/pull/178083,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2026-06-11T16:27:40Z
pr,177854,[Inductor] Run first cudagraph warmup in eager pool (#177854),https://github.com/pytorch/pytorch/pull/177854,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-11T16:26:02Z
pr,175584,fix aten.remainder in inductor,https://github.com/pytorch/pytorch/pull/175584,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-11T16:20:54Z
pr,171445,Intorduce torch.accelerator.is_graph_available to check if Graph is available,https://github.com/pytorch/pytorch/pull/171445,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-06-11T16:17:54Z
pr,170609,Add auto-generated cuDNN depthwise conv heuristic,https://github.com/pytorch/pytorch/pull/170609,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-06-11T14:58:57Z
pr,185053,[Test]Refactor test_custom_ops.py to be device agnostic,https://github.com/pytorch/pytorch/pull/185053,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-06-11T14:54:37Z
pr,185932,[MPS] Add `lcm` Metal kernel,https://github.com/pytorch/pytorch/pull/185932,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/mps, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-11T06:58:02Z
pr,182526,[DTensor] register missing pooling backward sharding strategies,https://github.com/pytorch/pytorch/pull/182526,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/dtensor, ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-11T05:16:27Z
pr,184825,[DTensor] add sharding strategy for group norm backward,https://github.com/pytorch/pytorch/pull/184825,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/dtensor, ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-11T02:25:56Z
pr,186949,Fix for scan autograd issue,https://github.com/pytorch/pytorch/pull/186949,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-06-11T00:25:57Z
pr,186608,[user-streams][aoti] Add minimal AOTI user stream support (#186608),https://github.com/pytorch/pytorch/pull/186608,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-10T21:53:51Z
pr,175468,[DO NOT MERGE] Test AMD Capacity.,https://github.com/pytorch/pytorch/pull/175468,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor-rocm-mi300, ciflow/periodic-rocm-mi300, ciflow/rocm-mi300","identify whether block is CI, merge conflict, or review gate",2026-06-10T19:23:01Z
pr,113258,[pytree] support PyStructSequence types for Python pytree,https://github.com/pytorch/pytorch/pull/113258,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-10T16:06:12Z
pr,181188,[AOTInductor] Outline per-kernel-call teardown into AOTI_NOINLINE helpers,https://github.com/pytorch/pytorch/pull/181188,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-10T16:02:28Z
pr,186139,Support closed-over data-dependent indexing in scan (forward and autograd),https://github.com/pytorch/pytorch/pull/186139,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-06-10T15:56:56Z
pr,186882,RFC Change error order,https://github.com/pytorch/pytorch/pull/186882,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-10T06:44:40Z
pr,186010,Surface ROCm GPU device faults in TorchElastic signal-failure messages (#186010),https://github.com/pytorch/pytorch/pull/186010,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/torchtitan,"identify whether block is CI, merge conflict, or review gate",2026-06-10T06:43:26Z
pr,179400,Use std::in_range for integer range checking,https://github.com/pytorch/pytorch/pull/179400,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/binaries, ciflow/trunk, ciflow/vllm","identify whether block is CI, merge conflict, or review gate",2026-06-10T05:02:37Z
pr,180081,[DTensor] Fix GIL-unsafe thread-local py::object in DTensor dispatch logging,https://github.com/pytorch/pytorch/pull/180081,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-06-10T01:27:12Z
pr,186831,Improve cudagraph stack trace coverage and assertions,https://github.com/pytorch/pytorch/pull/186831,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-10T00:10:48Z
pr,186829,Auto-promote CUDA graph static input copies,https://github.com/pytorch/pytorch/pull/186829,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-09T23:44:06Z
pr,186830,Add CUDA graph disable regions,https://github.com/pytorch/pytorch/pull/186830,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-09T23:44:04Z
pr,186828,Add CUDA graph output and input copy annotations,https://github.com/pytorch/pytorch/pull/186828,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-09T23:44:00Z
pr,186636,c10/core/impl/SizesAndStrides.h: add [[nodiscard]] to size() and isInline(),https://github.com/pytorch/pytorch/pull/186636,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-06-09T23:43:28Z
pr,178242,[XPU] Add grouped_mm dispatch for XPU via sycl-tla,https://github.com/pytorch/pytorch/pull/178242,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/trunk, ciflow/xpu","identify whether block is CI, merge conflict, or review gate",2026-06-09T13:01:20Z
pr,182013,Rename vec256_tests to vec_tests and fix executable search,https://github.com/pytorch/pytorch/pull/182013,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-06-09T10:11:09Z
pr,176600,Fix 1C1W tensor's shape and stride,https://github.com/pytorch/pytorch/pull/176600,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-06-09T07:32:32Z
pr,186634,c10/util/intrusive_ptr.h: replace hand-written comparison operators with operator<=>,https://github.com/pytorch/pytorch/pull/186634,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/trunk, ciflow/vllm","identify whether block is CI, merge conflict, or review gate",2026-06-09T06:20:33Z
pr,185101,Drop redundant tuple/list/int wrappers around tensor shape and numel,https://github.com/pytorch/pytorch/pull/185101,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/dtensor, ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-09T05:23:49Z
pr,185668,Fix typos in comments and docstrings across torch,https://github.com/pytorch/pytorch/pull/185668,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-09T04:14:09Z
pr,186609,[user-streams][aoti] Add minimal AOTI user stream support,https://github.com/pytorch/pytorch/pull/186609,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-08T18:33:12Z
pr,186458,Fix SDPA backend detection for autograd cache,https://github.com/pytorch/pytorch/pull/186458,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2026-06-08T14:10:03Z
pr,186389,[BE][Ez]: Update string_util APIs to use std::string_view more,https://github.com/pytorch/pytorch/pull/186389,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-06-08T14:04:31Z
pr,186586,Propagate override_cudagraphs annotation across CM-lifecycle gaps,https://github.com/pytorch/pytorch/pull/186586,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-08T10:29:44Z
pr,186334,Fix torch.kron crash on non-contiguous inputs,https://github.com/pytorch/pytorch/pull/186334,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-06-06T19:08:10Z
pr,185649,[Test] Make test_multihead_attention device generic for out-of-tree accelerators,https://github.com/pytorch/pytorch/pull/185649,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-06-05T19:09:58Z
pr,177263,Eliminate the unnecessary logic,https://github.com/pytorch/pytorch/pull/177263,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-06-05T10:44:50Z
pr,186304,"c10/core/DispatchKeySet: promote Full, FullAfter, Raw to enum class",https://github.com/pytorch/pytorch/pull/186304,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2026-06-05T06:35:30Z
pr,186218,Limit expandable IPC receiver VA reservation,https://github.com/pytorch/pytorch/pull/186218,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-06-05T03:16:09Z
pr,185956,torch/headeronly/core/TensorAccessor.h: replace typedef with using,https://github.com/pytorch/pytorch/pull/185956,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-06-05T01:46:11Z
pr,186253,[fsdp2] Add SPMD no-typecheck scopes to hooks,https://github.com/pytorch/pytorch/pull/186253,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-04T22:28:07Z
pr,183847,[distributed][docs] Fix group_src docstring in irecv,https://github.com/pytorch/pytorch/pull/183847,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-04T18:45:08Z
pr,185830,"Fix typos in comments, docstrings, and error messages",https://github.com/pytorch/pytorch/pull/185830,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/dtensor, ciflow/inductor, ciflow/inductor-pallas, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-04T18:40:31Z
pr,184315,[Test] Make autograd/test_functional.py device-agnostic for out-of-tree backends.,https://github.com/pytorch/pytorch/pull/184315,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-06-04T11:47:37Z
pr,186048,Dedupe namespaced codegen and annotation parsing in torchgen,https://github.com/pytorch/pytorch/pull/186048,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-06-04T09:37:25Z
pr,186130,Add _single ProcessGroup methods and deprecate the _base aliases,https://github.com/pytorch/pytorch/pull/186130,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/h100-symm-mem, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-04T00:09:31Z
pr,170644,[dynamo] Support dict assignment with lazy constant keys,https://github.com/pytorch/pytorch/pull/170644,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk, ciflow/vllm","identify whether block is CI, merge conflict, or review gate",2026-06-03T23:37:38Z
pr,170567,[dynamo] Support dict mutation without a guard,https://github.com/pytorch/pytorch/pull/170567,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-03T23:37:29Z
pr,170092,[dynamo] Support constant ops on unrealized LazyConstantVariables,https://github.com/pytorch/pytorch/pull/170092,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/dynamo, ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-03T23:37:16Z
pr,186131,Migrate ProcessGroup method usages to the _single names,https://github.com/pytorch/pytorch/pull/186131,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-06-03T23:30:37Z
pr,186095,[inductor] Cache per-node tiling in fusion candidate evaluation,https://github.com/pytorch/pytorch/pull/186095,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-03T21:13:17Z
pr,186086,[PyTorch] Add missing meta kernel for aten::_sample_dirichlet,https://github.com/pytorch/pytorch/pull/186086,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-03T21:13:12Z
pr,186106,[inductor] Fast paths in sizevars for concrete integer comparisons,https://github.com/pytorch/pytorch/pull/186106,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-03T21:13:11Z
pr,184965,[Inductor] Cap combo-kernel seed autotune configs by size,https://github.com/pytorch/pytorch/pull/184965,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-03T18:27:25Z
pr,186020,[WIP] Reintroduce ptxas bundling for CUDA 13.0 binaries (revert #174716),https://github.com/pytorch/pytorch/pull/186020,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-03T16:34:21Z
pr,179547,[inductor] Add thread-based compilation support for free-threaded Python (#179547) (#179547),https://github.com/pytorch/pytorch/pull/179547,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-03T00:14:38Z
pr,185844,[CuTeDSL] Just use tensor strides directly in `_to_fake_cute_tensor`,https://github.com/pytorch/pytorch/pull/185844,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/b200, ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-02T22:56:35Z
pr,185592,Remove android directory,https://github.com/pytorch/pytorch/pull/185592,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-06-02T20:51:03Z
pr,184226,[Inductor] Allow mixed pointwise/reduction combo kernels,https://github.com/pytorch/pytorch/pull/184226,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-02T05:19:36Z
pr,185183,[Inductor] Fix combo-kernel num_warps regression on mixed reductions,https://github.com/pytorch/pytorch/pull/185183,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-02T05:19:20Z
pr,185606,Enable TF32 fpmath mode for XPU deconvolution,https://github.com/pytorch/pytorch/pull/185606,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/xpu,"identify whether block is CI, merge conflict, or review gate",2026-06-02T05:19:06Z
pr,183390,[SymmMem] Add NCCL implementation of all_to_all_vdev,https://github.com/pytorch/pytorch/pull/183390,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/h100-symm-mem, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-01T23:47:50Z
pr,185689,Upload docs preview directly to S3 from the OSDC pod when authenticated,https://github.com/pytorch/pytorch/pull/185689,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-06-01T20:23:56Z
pr,185646,[AI Codemod][AsyncioGetEventLoopMigration] [asyncio-codemod] xplat/caffe2/test,https://github.com/pytorch/pytorch/pull/185646,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-06-01T18:51:45Z
pr,185432,Upgrade pyrefly 1.0.0,https://github.com/pytorch/pytorch/pull/185432,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/b200, ciflow/dtensor, ciflow/h100, ciflow/inductor","identify whether block is CI, merge conflict, or review gate",2026-06-01T15:07:27Z
pr,181087,[MPS] RDMA for distributed MPS,https://github.com/pytorch/pytorch/pull/181087,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/mps, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-01T14:59:43Z
pr,185666,Add ROCm CI workflows for mi210,https://github.com/pytorch/pytorch/pull/185666,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-05-30T06:58:45Z
pr,185570,tmp,https://github.com/pytorch/pytorch/pull/185570,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-05-29T04:59:41Z
pr,184253,[DO NOT MERGE][ROCm][CI] Test k8s dpx,https://github.com/pytorch/pytorch/pull/184253,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-05-29T04:59:10Z
pr,182655,Update ACL/OpenBLAS/manywheel build scripts,https://github.com/pytorch/pytorch/pull/182655,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-05-29T02:18:09Z
pr,178314,Separate tagging path for AC when non-strict tracing,https://github.com/pytorch/pytorch/pull/178314,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-05-29T01:57:49Z
pr,178416,[dynamo] Migrate allow_in_graph → nonstrict_trace in torch/ production code,https://github.com/pytorch/pytorch/pull/178416,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-05-29T01:57:47Z
pr,177283,RFC : WIP warmup compile with fake tensor with dynamic dims then run with real,https://github.com/pytorch/pytorch/pull/177283,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-05-29T01:57:45Z
pr,176989,Add supports for itertools.batched and itertools.combinations_with_re…,https://github.com/pytorch/pytorch/pull/176989,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-05-28T16:30:16Z
pr,182256,[MPS] enable metal performance primitives attention,https://github.com/pytorch/pytorch/pull/182256,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/mps,"identify whether block is CI, merge conflict, or review gate",2026-05-28T03:00:04Z
pr,177568,[MTIA] Add dilation support for max_pool2d_with_indices_backward lowering (#177568),https://github.com/pytorch/pytorch/pull/177568,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/inductor-pallas, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-05-28T01:35:09Z
pr,184192,[Test]Refactor test/nn/test_lazy_modules.py to be device- generic by replacing hardcoded device selection with torch.accelerator API,https://github.com/pytorch/pytorch/pull/184192,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-05-27T14:51:08Z
pr,175746,RISC-V Vector Extension (RVV) support for ATen,https://github.com/pytorch/pytorch/pull/175746,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-05-27T12:59:11Z
pr,175758,Create a DeviceModule protocol,https://github.com/pytorch/pytorch/pull/175758,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-05-27T08:01:58Z
pr,184607,"Speed up inductor test infrastructure (~4x collection, ~1.7x execution)",https://github.com/pytorch/pytorch/pull/184607,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-05-27T04:01:30Z
pr,185097,Migrate deprecated torch.norm calls to torch.linalg.*,https://github.com/pytorch/pytorch/pull/185097,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/torchtitan,"identify whether block is CI, merge conflict, or review gate",2026-05-26T20:05:15Z
pr,185091,Prefer .view() and drop redundant .contiguous() in tensor-creator chains,https://github.com/pytorch/pytorch/pull/185091,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-05-26T13:51:40Z
pr,184509,[torch/csrc] Add bulk storage helpers; remove per-byte loops,https://github.com/pytorch/pytorch/pull/184509,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-05-26T13:31:38Z
pr,184657,[MPS] Tolerate lenient mixed-dtype batch_norm in InstanceNorm test,https://github.com/pytorch/pytorch/pull/184657,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/mps,"identify whether block is CI, merge conflict, or review gate",2026-05-26T13:29:59Z
pr,165837,[OpenReg][Random] Enrich OpenReg random number generator module implementation add focused documentation,https://github.com/pytorch/pytorch/pull/165837,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-05-26T13:28:08Z
pr,184254,"Fix `replication_pad{2d,3d}` meta accepting negative output dims",https://github.com/pytorch/pytorch/pull/184254,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-05-25T12:12:57Z
issue,187781,[RFC] Topology-aware nn.Module primitives for memory optimization in torch.compile,https://github.com/pytorch/pytorch/issues/187781,has linked PR,62,"open PR references this issue; needs design/research label present; component/oncall labels: module: compile ux, module: memory usage, module: nn, oncall: pt2",route to design/research decision before patch planning,2026-07-07T16:34:40Z
issue,108744,switch more test cases to use MultithreadTestCase,https://github.com/pytorch/pytorch/issues/108744,has linked PR,62,open PR references this issue; label says issue is suitable for a scoped external contribution; component/oncall labels: module: dtensor,review linked PR state before touching issue,2026-07-02T18:54:27Z
issue,113040,[PT2] [Hardening] Track recompiles alongside graph breaks in our actual/expected comparison CI runs,https://github.com/pytorch/pytorch/issues/113040,has linked PR,62,"open PR references this issue; label says issue is suitable for a scoped external contribution; component/oncall labels: module: dynamo, oncall: pt2",review linked PR state before touching issue,2026-06-28T20:14:13Z
issue,115016,Enable ruff rule PLW1510 codebase wide,https://github.com/pytorch/pytorch/issues/115016,has linked PR,62,open PR references this issue; label says issue is suitable for a scoped external contribution; component/oncall labels: module: lint,review linked PR state before touching issue,2026-06-27T21:33:27Z
issue,102870,Inductor: delete code that extracts out sizevars by inspecting tensor inputs to find a size that handled it,https://github.com/pytorch/pytorch/issues/102870,has linked PR,62,open PR references this issue; label says issue is suitable for a scoped external contribution; component/oncall labels: module: inductor,review linked PR state before touching issue,2026-06-25T02:16:35Z
issue,108406,enhance documentation around the developer build,https://github.com/pytorch/pytorch/issues/108406,has linked PR,62,open PR references this issue; label says issue is suitable for a scoped external contribution,review linked PR state before touching issue,2026-06-18T01:02:39Z
issue,137285,Docs are little bit outdated for torch logs,https://github.com/pytorch/pytorch/issues/137285,has linked PR,62,"open PR references this issue; label says issue is suitable for a scoped external contribution; component/oncall labels: module: logging, oncall: pt2",review linked PR state before touching issue,2026-06-17T11:05:49Z
issue,149534,Addind RoPE to pytorch core,https://github.com/pytorch/pytorch/issues/149534,has linked PR,62,open PR references this issue; needs design/research label present; component/oncall labels: module: nn,route to design/research decision before patch planning,2026-06-01T01:02:28Z
pr,161044,Thread based dataloading workers,https://github.com/pytorch/pytorch/pull/161044,ready for maintainer decision,62,approved with no detected failing/pending checks; PR has not updated for at least 45 days; CI lane labels: ciflow/trunk,final maintainer merge/release decision,2026-05-21T17:05:17Z
pr,183335,Internal,https://github.com/pytorch/pytorch/pull/183335,ready for maintainer decision,62,approved with no detected failing/pending checks; PR has not updated for at least 45 days; CI lane labels: ciflow/trunk,final maintainer merge/release decision,2026-05-18T18:46:39Z
issue,133443,[BE] Deduplicate auto_functionalized and triton_kernel_wrapper_functional,https://github.com/pytorch/pytorch/issues/133443,has linked PR,62,"open PR references this issue; label says issue is suitable for a scoped external contribution; component/oncall labels: module: pt2-dispatcher, module: user triton, oncall: pt2",review linked PR state before touching issue,2026-05-15T17:45:27Z
issue,122129,Obscure error: Expected a value of type 'List[int]' for argument 'sizes' but instead found type 'immutable_list',https://github.com/pytorch/pytorch/issues/122129,has linked PR,62,"open PR references this issue; label says issue is suitable for a scoped external contribution; component/oncall labels: module: dynamic shapes, module: pt2-dispatcher, oncall: pt2",review linked PR state before touching issue,2026-05-12T21:07:07Z
pr,176001,[test] Resolve undefined GRAPH_EXECUTOR in test_jit_profiling.py,https://github.com/pytorch/pytorch/pull/176001,ready for maintainer decision,62,approved with no detected failing/pending checks; PR has not updated for at least 45 days; CI lane labels: ciflow/trunk,final maintainer merge/release decision,2026-05-11T14:19:27Z
issue,48251,torch.eye(d) is slow and hogs cpu for d >= 182,https://github.com/pytorch/pytorch/issues/48251,has linked PR,62,"open PR references this issue; label says issue is suitable for a scoped external contribution; component/oncall labels: module: cpu, module: multithreading, module: performance",review linked PR state before touching issue,2026-04-15T01:15:15Z
issue,179250,Muon: support of dynamic coefficients and normalization,https://github.com/pytorch/pytorch/issues/179250,has linked PR,62,open PR references this issue; needs design/research label present; component/oncall labels: module: optimizer,route to design/research decision before patch planning,2026-04-03T16:58:12Z
issue,135859,"bmm, topk, cholesky, linalg.norm, max with out variants set causing recompilations in torch.compile",https://github.com/pytorch/pytorch/issues/135859,has linked PR,62,"open PR references this issue; label says issue is suitable for a scoped external contribution; component/oncall labels: module: dynamic shapes, oncall: pt2",review linked PR state before touching issue,2026-03-10T16:14:33Z
issue,116396,[dynamo] Missing support for many trivial builtin functions in operator,https://github.com/pytorch/pytorch/issues/116396,has linked PR,62,"open PR references this issue; label says issue is suitable for a scoped external contribution; component/oncall labels: module: dynamo, oncall: pt2",review linked PR state before touching issue,2026-03-08T15:46:12Z
issue,92141,Improve the error message in `linalg.eigh` when using large matrices,https://github.com/pytorch/pytorch/issues/92141,has linked PR,62,"open PR references this issue; label says issue is suitable for a scoped external contribution; component/oncall labels: module: 64-bit, module: linear algebra, module: mkl",review linked PR state before touching issue,2026-02-08T19:05:21Z
pr,166813,Add min/max support for barebones uint types,https://github.com/pytorch/pytorch/pull/166813,ready for maintainer decision,62,approved with no detected failing/pending checks; PR has not updated for at least 45 days; CI lane labels: ciflow/trunk,final maintainer merge/release decision,2025-12-29T04:45:53Z
pr,146172,Factory function support for NestedTensor,https://github.com/pytorch/pytorch/pull/146172,ready for maintainer decision,62,approved with no detected failing/pending checks; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor,final maintainer merge/release decision,2025-04-08T21:21:14Z
pr,138626,Fix global namespace pollution in ATen/Dispatch.h,https://github.com/pytorch/pytorch/pull/138626,ready for maintainer decision,62,"approved with no detected failing/pending checks; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/trunk",final maintainer merge/release decision,2025-04-05T08:00:58Z
pr,138436,"Replace use of PyTorch 2.0 with torch.compile, and minor edits",https://github.com/pytorch/pytorch/pull/138436,ready for maintainer decision,62,approved with no detected failing/pending checks; PR has not updated for at least 45 days; CI lane labels: ciflow/trunk,final maintainer merge/release decision,2025-01-30T18:36:45Z
pr,138471,Add torch.compiler.config.cache,https://github.com/pytorch/pytorch/pull/138471,ready for maintainer decision,62,approved with no detected failing/pending checks; PR has not updated for at least 45 days,final maintainer merge/release decision,2025-01-30T15:10:38Z
pr,132814,Update TwoTensor to no longer inherit requires_grad from inner tensor,https://github.com/pytorch/pytorch/pull/132814,ready for maintainer decision,62,approved with no detected failing/pending checks; PR has not updated for at least 45 days,final maintainer merge/release decision,2024-10-15T13:50:45Z
pr,124490,[dynamo] Support ndarray.dtype attribute access,https://github.com/pytorch/pytorch/pull/124490,ready for maintainer decision,62,"approved with no detected failing/pending checks; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/trunk",final maintainer merge/release decision,2024-08-06T21:20:16Z
pr,120076,Change ATEN generator argument type to const std::optional<Generator>&,https://github.com/pytorch/pytorch/pull/120076,ready for maintainer decision,62,"approved with no detected failing/pending checks; PR has not updated for at least 45 days; CI lane labels: ciflow/binaries, ciflow/inductor, ciflow/mps, ciflow/trunk",final maintainer merge/release decision,2024-05-25T23:18:28Z
pr,111673,[quant][bc-breaking] Remove deprecated QConfigDynamic,https://github.com/pytorch/pytorch/pull/111673,ready for maintainer decision,62,approved with no detected failing/pending checks; PR has not updated for at least 45 days,final maintainer merge/release decision,2024-01-03T07:00:57Z
pr,189313,Support aten::cumprod ONNX export via CumProd (opset 26),https://github.com/pytorch/pytorch/pull/189313,draft/noise,61,PR is draft; failed checks or blocked/unstable merge state; PR closes or references open issue numbers,"identify whether block is CI, merge conflict, or review gate",2026-07-09T00:21:13Z
pr,186082,Improve SymIntArrayRef concrete-int diagnostics,https://github.com/pytorch/pytorch/pull/186082,draft/noise,61,PR is draft; failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2026-07-08T21:23:11Z
pr,189273,Fix nn.functional.threshold OpInfo reference for strict NumPy uint8 casting,https://github.com/pytorch/pytorch/pull/189273,draft/noise,61,PR is draft; failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-07-08T19:12:52Z
pr,186189,Report dynamo expected-failure XPASSes,https://github.com/pytorch/pytorch/pull/186189,draft/noise,61,"PR is draft; failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T16:17:51Z
pr,189147,Fix unload utility for newer Triton pyds,https://github.com/pytorch/pytorch/pull/189147,draft/noise,61,"PR is draft; failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/xpu","identify whether block is CI, merge conflict, or review gate",2026-07-08T16:17:35Z
pr,186373,Disable duck sizing by default,https://github.com/pytorch/pytorch/pull/186373,draft/noise,61,"PR is draft; failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/b200, ciflow/dtensor, ciflow/h100, ciflow/inductor","identify whether block is CI, merge conflict, or review gate",2026-07-08T15:35:48Z
pr,186283,Trace data_ptr through tensor constructors,https://github.com/pytorch/pytorch/pull/186283,draft/noise,61,"PR is draft; failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T15:35:09Z
pr,187609,Match CUDA threshold scalar dtype in decomp,https://github.com/pytorch/pytorch/pull/187609,draft/noise,61,PR is draft; failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2026-07-08T07:46:01Z
pr,185082,Preserve lifted SymInts during strict export capture,https://github.com/pytorch/pytorch/pull/185082,draft/noise,61,"PR is draft; failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T07:45:02Z
pr,186801,Allow swap_tensors with internal TensorWeakRef,https://github.com/pytorch/pytorch/pull/186801,draft/noise,61,"PR is draft; failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/slow, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-08T06:02:14Z
pr,185460,Fix CPU perf smoketest baseline drift checks,https://github.com/pytorch/pytorch/pull/185460,draft/noise,61,PR is draft; failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2026-07-07T19:21:55Z
pr,181860,[c10] Add InlineEventBase to unify event APIs in DeviceGuardImplInter…,https://github.com/pytorch/pytorch/pull/181860,draft/noise,61,PR is draft; failed checks or blocked/unstable merge state; PR closes or references open issue numbers,"identify whether block is CI, merge conflict, or review gate",2026-07-07T19:12:51Z
pr,187452,Polyhedral Optimization Pass for PyTorch Inductor (Phase 1),https://github.com/pytorch/pytorch/pull/187452,draft/noise,61,"PR is draft; failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T13:19:20Z
pr,188894,Fix HAS_XPU_AND_TRITON to require triton XPU backend,https://github.com/pytorch/pytorch/pull/188894,draft/noise,61,PR is draft; failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/xpu,"identify whether block is CI, merge conflict, or review gate",2026-07-07T09:01:03Z
pr,185748,[dynamo] Model RemovableHandle hook registration side effects,https://github.com/pytorch/pytorch/pull/185748,draft/noise,61,PR is draft; failed checks or blocked/unstable merge state; PR closes or references open issue numbers,"identify whether block is CI, merge conflict, or review gate",2026-07-07T08:22:16Z
pr,185749,Fix Dynamo construction for classes with __getattribute__,https://github.com/pytorch/pytorch/pull/185749,draft/noise,61,"PR is draft; failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T07:05:24Z
pr,182768,fix(resize): rollback tensor metadata when storage resize fails,https://github.com/pytorch/pytorch/pull/182768,draft/noise,61,PR is draft; failed checks or blocked/unstable merge state; PR closes or references open issue numbers,"identify whether block is CI, merge conflict, or review gate",2026-07-06T06:01:55Z
pr,188995,[dynamo] Allow raw unbacked SymInts in strict mode (FlexAttention BlockMask) (#187547),https://github.com/pytorch/pytorch/pull/188995,draft/noise,61,"PR is draft; failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-06T04:31:33Z
pr,188993,[dynamo] Preserve Event/Stream subclass identity across graph breaks (#188536),https://github.com/pytorch/pytorch/pull/188993,draft/noise,61,"PR is draft; failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-06T04:16:23Z
pr,188992,[inductor][XPU] Fix AMP autocast round-trip fp16 accuracy (#188261),https://github.com/pytorch/pytorch/pull/188992,draft/noise,61,"PR is draft; failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-06T04:16:23Z
pr,177115,Fix opcheck to detect stride mismatches on CPU tensors,https://github.com/pytorch/pytorch/pull/177115,draft/noise,61,PR is draft; failed checks or blocked/unstable merge state; PR closes or references open issue numbers,"identify whether block is CI, merge conflict, or review gate",2026-07-05T23:50:50Z
pr,188350,Fix Dynamo stream/event variable python_type,https://github.com/pytorch/pytorch/pull/188350,draft/noise,61,"PR is draft; failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-01T10:55:08Z
pr,179443,[FSDP2] Fix HSDP O(n_layers) memory accumulation of reduce-scatter output buffers,https://github.com/pytorch/pytorch/pull/179443,draft/noise,61,"PR is draft; failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/h100-symm-mem, ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-30T22:00:07Z
pr,187461,Do Not Review: Polyhedral Optimization,https://github.com/pytorch/pytorch/pull/187461,draft/noise,61,"PR is draft; failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/inductor-perf-compare, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-29T17:14:26Z
pr,177205,Fix all_gather_into_tensor with stack-shaped output under torch.compile + dynamic shapes,https://github.com/pytorch/pytorch/pull/177205,draft/noise,61,"PR is draft; failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-29T10:04:16Z
pr,179625,Enable ruff rule PLW1510 codebase wide,https://github.com/pytorch/pytorch/pull/179625,draft/noise,61,"PR is draft; failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/dtensor, ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-27T01:01:55Z
pr,188253,Add XPU dispatch for linalg.polar,https://github.com/pytorch/pytorch/pull/188253,draft/noise,61,PR is draft; failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/xpu,"identify whether block is CI, merge conflict, or review gate",2026-06-26T12:36:49Z
pr,188178,Guard score_mod oob indexing,https://github.com/pytorch/pytorch/pull/188178,draft/noise,61,"PR is draft; failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-25T22:46:40Z
pr,187883,Avoid per-symbol tuple allocation in the C++ symbolic shape guard,https://github.com/pytorch/pytorch/pull/187883,draft/noise,61,PR is draft; failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2026-06-25T02:15:49Z
pr,188037,Fix current stream selection,https://github.com/pytorch/pytorch/pull/188037,draft/noise,61,PR is draft; failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2026-06-24T07:31:38Z
pr,180955,[ROCm] Normalize AMDSMI UUID matching,https://github.com/pytorch/pytorch/pull/180955,draft/noise,61,PR is draft; failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/rocm-mi300,"identify whether block is CI, merge conflict, or review gate",2026-06-21T21:58:42Z
pr,180716,Fix torch.compile addmm with beta=0 and mismatched bias,https://github.com/pytorch/pytorch/pull/180716,draft/noise,61,"PR is draft; failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-21T21:58:41Z
pr,133297,Update cmake,https://github.com/pytorch/pytorch/pull/133297,draft/noise,61,"PR is draft; failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/linux-aarch64, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-18T16:40:32Z
pr,186797,"Fix mm_plus_mm XPU: guard broken Triton template, enable all mm_plus_mm tests",https://github.com/pytorch/pytorch/pull/186797,draft/noise,61,"PR is draft; failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/b200, ciflow/h100, ciflow/inductor, ciflow/rocm-mi300","identify whether block is CI, merge conflict, or review gate",2026-06-17T16:02:37Z
pr,187490,[Dynamo] Trace raw unbacked SymInt inputs,https://github.com/pytorch/pytorch/pull/187490,draft/noise,61,"PR is draft; failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-17T03:27:22Z
pr,133289,Update native/mkldnn to native/onednn,https://github.com/pytorch/pytorch/pull/133289,draft/noise,61,"PR is draft; failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/docker, ciflow/inductor, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-17T01:25:48Z
pr,180707,Improve cpp_extension and drop setuptools runtime requirement,https://github.com/pytorch/pytorch/pull/180707,draft/noise,61,PR is draft; failed checks or blocked/unstable merge state; PR closes or references open issue numbers,"identify whether block is CI, merge conflict, or review gate",2026-06-16T23:41:24Z
pr,180379,[Dynamo] Inline root module forward hooks when compiling modules,https://github.com/pytorch/pytorch/pull/180379,draft/noise,61,"PR is draft; failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-14T03:48:23Z
pr,177287,"Default backward_pass_autocast to ""off"" to match eager autocast semantics",https://github.com/pytorch/pytorch/pull/177287,draft/noise,61,PR is draft; failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2026-06-12T17:27:50Z
pr,185989,Bump onnx to 1.22.0rc1 to integrate ONNX 1.22.0 release branch,https://github.com/pytorch/pytorch/pull/185989,draft/noise,61,PR is draft; failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/docker,"identify whether block is CI, merge conflict, or review gate",2026-06-09T23:44:57Z
pr,185308,Align subclass output SymInt unwrapping with metadata,https://github.com/pytorch/pytorch/pull/185308,draft/noise,61,"PR is draft; failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/dtensor, ciflow/inductor","identify whether block is CI, merge conflict, or review gate",2026-06-03T00:16:59Z
pr,183909,changing Sleef_asinhf8_u10 to std impl,https://github.com/pytorch/pytorch/pull/183909,draft/noise,61,PR is draft; failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2026-05-28T14:56:18Z
issue,167605,[Profiler] About adding the line number of calling stack from parent,https://github.com/pytorch/pytorch/issues/167605,needs triage,59,PyTorch issue lacks triaged label; needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: oncall: profiler,confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-11-20T02:13:47Z
issue,158701,"RuntimeError: optimized_plan_ INTERNAL ASSERT FAILED at ""/opt/conda/envs/bld/conda-bld/libtorch_1741644046984/work/torch/csrc/jit/runtime/profiling_graph_executor_impl.cpp"":725, please report a bug to PyTorch.",https://github.com/pytorch/pytorch/issues/158701,needs triage,59,"PyTorch issue lacks triaged label; needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: third_party, oncall: jit",confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-07-21T01:43:34Z
issue,154708,torch.quantized_batch_norm API return different output when input is nan/inf or undefined mathematical calculation method,https://github.com/pytorch/pytorch/issues/154708,needs triage,59,"PyTorch issue lacks triaged label; needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: nans and infs, oncall: quantization",confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-06-26T06:12:05Z
issue,139745,"cmake libtorch_jni.so, bug: 'fp16/psimd.h' file not found",https://github.com/pytorch/pytorch/issues/139745,needs triage,59,"PyTorch issue lacks triaged label; needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: binaries, oncall: mobile",confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-11-07T03:12:44Z
issue,133227,Issue Exporting PyTorch Model with pytorch3d.ops.ball_query to ONNX: Bug or Compatibility Problem?,https://github.com/pytorch/pytorch/issues/133227,needs triage,59,"PyTorch issue lacks triaged label; needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: onnx, oncall: export",confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-10-17T18:22:11Z
issue,106846,untimeError: The following operation failed in the TorchScript interpreter. Traceback of TorchScript (most recent call last): RuntimeError: nvrtc: error: invalid value for --gpu-architecture (-arch),https://github.com/pytorch/pytorch/issues/106846,needs triage,59,PyTorch issue lacks triaged label; needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-03-21T14:50:54Z
issue,97876,Missing torch import in _contextlib.py when using torch.jit._recursive,https://github.com/pytorch/pytorch/issues/97876,needs triage,59,PyTorch issue lacks triaged label; needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-03-18T13:49:08Z
issue,49157,softmax fail,https://github.com/pytorch/pytorch/issues/49157,needs triage,59,"PyTorch issue lacks triaged label; needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: nn, module: viewing and reshaping, oncall: jit",confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-01-12T03:07:10Z
issue,19106,"Performance issue with torch.jit.trace(), slow prediction in C++ (CPU)",https://github.com/pytorch/pytorch/issues/19106,needs triage,59,"triage signal present; needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: performance, oncall: jit",confirm whether existing evidence is enough; otherwise wait for a reproducer,2023-03-20T02:38:22Z
issue,30651,[TensorBoard] The different order of import SummaryWriter may cause Segmentation fault,https://github.com/pytorch/pytorch/issues/30651,needs triage,59,PyTorch issue lacks triaged label; needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: oncall: visualization,confirm whether existing evidence is enough; otherwise wait for a reproducer,2022-12-15T01:48:46Z
issue,30525,[Tensorboard] nothing appears in the projector tab,https://github.com/pytorch/pytorch/issues/30525,needs triage,59,"PyTorch issue lacks triaged label; needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: tensorboard, oncall: visualization",confirm whether existing evidence is enough; otherwise wait for a reproducer,2022-05-16T15:34:09Z
issue,60930,error for using faster rcnn on c++ & GPU,https://github.com/pytorch/pytorch/issues/60930,needs triage,59,PyTorch issue lacks triaged label; needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,confirm whether existing evidence is enough; otherwise wait for a reproducer,2021-06-30T00:03:38Z
issue,16592,[JIT] NVRTC unknown error,https://github.com/pytorch/pytorch/issues/16592,needs triage,59,PyTorch issue lacks triaged label; needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,confirm whether existing evidence is enough; otherwise wait for a reproducer,2020-03-16T20:00:44Z
issue,15342,[JIT] Batch op. batch_sum return a mask which is on cpu,https://github.com/pytorch/pytorch/issues/15342,needs triage,59,PyTorch issue lacks triaged label; needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,confirm whether existing evidence is enough; otherwise wait for a reproducer,2020-02-10T23:50:56Z
issue,188355,[Inductor][CUDA Graphs] max-autotune attempts to capture torch.linalg.eigh and crashes with cudaErrorStreamCaptureUnsupported,https://github.com/pytorch/pytorch/issues/188355,has linked PR,56,"open PR references this issue; needs reproduction label present; component/oncall labels: module: cuda graphs, module: inductor, module: linear algebra, oncall: pt2",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-06-29T17:22:27Z
issue,169688,[CPU] Stride 3 bfloat16 convolutions appear incorrect,https://github.com/pytorch/pytorch/issues/169688,has linked PR,56,"open PR references this issue; needs reproduction label present; component/oncall labels: module: convolution, module: cuda",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-06-27T05:30:44Z
issue,132756,JITerator: Torch.prod produces RuntimeError: CUDA driver error: invalid argument,https://github.com/pytorch/pytorch/issues/132756,has linked PR,56,"open PR references this issue; needs reproduction label present; component/oncall labels: module: jiterator, oncall: pt2",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-06-14T07:20:25Z
issue,181650,"MPS: `add_dense_scalar_cast_float` shader fails with ""Read-only bytes bound to shader argument with write access enabled"" assertion during pyannote.audio inference (M4 Pro, torch 2.11)",https://github.com/pytorch/pytorch/issues/181650,has linked PR,56,open PR references this issue; needs reproduction label present; component/oncall labels: module: mps,confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-05-26T10:07:50Z
issue,176178,[CuTeDSL] grouped_mm illegal address from TMA descriptor init/update shape mismatch,https://github.com/pytorch/pytorch/issues/176178,has linked PR,56,"open PR references this issue; needs reproduction label present; component/oncall labels: module: inductor, oncall: pt2",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-05-19T18:06:43Z
issue,150262,Fake mode mismatch when doing nested compile + tensor subclass,https://github.com/pytorch/pytorch/issues/150262,has linked PR,56,"open PR references this issue; needs reproduction label present; component/oncall labels: module: faketensor, module: pt2-dispatcher, oncall: pt2",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-05-15T02:08:39Z
issue,149963,[ued][wan] nonstrict_trace does not support None inputs,https://github.com/pytorch/pytorch/issues/149963,has linked PR,56,"open PR references this issue; needs reproduction label present; component/oncall labels: module: aotdispatch, module: pt2-dispatcher, oncall: pt2",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-05-15T02:03:50Z
issue,148112,MLA with Learnable RoPE Tensors is Broken with Flex Attention,https://github.com/pytorch/pytorch/issues/148112,has linked PR,56,"open PR references this issue; needs reproduction label present; component/oncall labels: module: flex attention, module: higher order operators, module: pt2-dispatcher, oncall: pt2",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-05-14T22:11:22Z
issue,137009,pre-dispatch tracing can't interpose on detach() calls from the autograd engine.,https://github.com/pytorch/pytorch/issues/137009,has linked PR,56,"open PR references this issue; needs reproduction label present; component/oncall labels: module: proxytensor, module: pt2-dispatcher, oncall: pt2",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-02-22T18:35:23Z
issue,173046,[Segfault] compilion with eager for the rnn_tanh operator leads to a segfault,https://github.com/pytorch/pytorch/issues/173046,has linked PR,56,open PR references this issue; needs reproduction label present; component/oncall labels: oncall: pt2,confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-02-04T03:27:31Z
issue,148003,"Error : torch/utils/_sympy/interp.py:176] [0/2] failed while executing pow_by_natural([VR1, int_oo], VR[-1, -1]])",https://github.com/pytorch/pytorch/issues/148003,has linked PR,56,open PR references this issue; needs reproduction label present; component/oncall labels: oncall: pt2,confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-01-27T14:18:26Z
issue,170696,[torch.compile] `CatchErrorsWrapper' object has no attribute '__closure__` when wrapping a fullgraph'd function in fullgraph'd module forward,https://github.com/pytorch/pytorch/issues/170696,has linked PR,56,"open PR references this issue; needs reproduction label present; component/oncall labels: module: dynamo, oncall: pt2",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-01-16T22:52:32Z
pr,185044,[AI SLOP] Support SymInt list custom op returns,https://github.com/pytorch/pytorch/pull/185044,PR blocked,55,"failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-05-24T07:25:53Z
pr,184756,Fix Dynamo dict equality semantics,https://github.com/pytorch/pytorch/pull/184756,PR blocked,55,"failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-05-22T20:46:16Z
pr,184758,Fix Dynamo defaultdict shallow copy semantics,https://github.com/pytorch/pytorch/pull/184758,PR blocked,55,"failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-05-22T17:43:40Z
pr,184813,Use custom Min Max in Inductor symbolic indexing,https://github.com/pytorch/pytorch/pull/184813,PR blocked,55,"failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/inductor-pallas, ciflow/mps, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-05-22T16:34:35Z
pr,184489,Forward CXX compiler to nvcc as host compiler for GCC too,https://github.com/pytorch/pytorch/pull/184489,PR blocked,55,failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-05-22T15:43:28Z
pr,181077,Enable fbgemm on MacOS,https://github.com/pytorch/pytorch/pull/181077,PR blocked,55,failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-05-22T15:42:12Z
pr,182781,[1/N] Use const pointers and references,https://github.com/pytorch/pytorch/pull/182781,PR blocked,55,"failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/b200, ciflow/h100, ciflow/inductor, ciflow/rocm-mi300","identify whether block is CI, merge conflict, or review gate",2026-05-22T13:34:55Z
pr,179397,Modernize map usage in ATen,https://github.com/pytorch/pytorch/pull/179397,PR blocked,55,"failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/mps, ciflow/trunk, ciflow/vllm","identify whether block is CI, merge conflict, or review gate",2026-05-22T11:55:18Z
pr,184812,Improve Inductor index simplification,https://github.com/pytorch/pytorch/pull/184812,PR blocked,55,"failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-05-22T03:31:02Z
pr,184801,Gate baddbmm expectedFailureDynamic on TEST_Z3,https://github.com/pytorch/pytorch/pull/184801,PR blocked,55,failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2026-05-22T02:16:57Z
pr,178176,[DO NOT MERGE]Test New Landing Pg,https://github.com/pytorch/pytorch/pull/178176,PR blocked,55,failed checks or blocked/unstable merge state; PR has not updated for at least 45 days,"identify whether block is CI, merge conflict, or review gate",2026-05-21T23:38:45Z
pr,184617,Fix Dynamo sequence repeat identity semantics,https://github.com/pytorch/pytorch/pull/184617,PR blocked,55,"failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-05-21T22:15:40Z
pr,184755,Fix Dynamo sequence search semantics,https://github.com/pytorch/pytorch/pull/184755,PR blocked,55,"failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-05-21T21:53:45Z
pr,184754,Fix Dynamo reversed iterator consumption,https://github.com/pytorch/pytorch/pull/184754,PR blocked,55,"failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-05-21T21:50:47Z
pr,184765,Fix Dynamo ordered contains semantics,https://github.com/pytorch/pytorch/pull/184765,PR blocked,55,"failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-05-21T21:49:04Z
pr,184766,Fix Dynamo set discard and remove semantics,https://github.com/pytorch/pytorch/pull/184766,PR blocked,55,"failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-05-21T21:49:02Z
pr,184635,Fix Dynamo cmp_to_key list sort semantics,https://github.com/pytorch/pytorch/pull/184635,PR blocked,55,"failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-05-21T21:48:59Z
pr,184761,Fix Dynamo mutable set dict-key hashing semantics,https://github.com/pytorch/pytorch/pull/184761,PR blocked,55,"failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-05-21T21:48:03Z
pr,184759,Fix Dynamo frozenset dict-key hashing semantics,https://github.com/pytorch/pytorch/pull/184759,PR blocked,55,"failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-05-21T21:48:02Z
pr,184763,Fix Dynamo dict copy GC tracking semantics,https://github.com/pytorch/pytorch/pull/184763,PR blocked,55,"failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-05-21T21:47:57Z
pr,184762,Fix Dynamo list extend iterator semantics,https://github.com/pytorch/pytorch/pull/184762,PR blocked,55,"failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-05-21T21:47:56Z
pr,184757,Fix Dynamo dict fromkeys semantics,https://github.com/pytorch/pytorch/pull/184757,PR blocked,55,"failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-05-21T21:47:49Z
pr,184616,Fix Dynamo list constructor iteration protocol,https://github.com/pytorch/pytorch/pull/184616,PR blocked,55,"failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-05-21T21:47:36Z
pr,184622,Fix Dynamo tuple constructor semantics,https://github.com/pytorch/pytorch/pull/184622,PR blocked,55,"failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-05-21T21:47:24Z
pr,184645,Fix Dynamo defaultdict inplace union semantics,https://github.com/pytorch/pytorch/pull/184645,PR blocked,55,"failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-05-21T21:46:35Z
pr,184627,Fix Dynamo dict update semantics,https://github.com/pytorch/pytorch/pull/184627,PR blocked,55,"failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-05-21T21:46:24Z
pr,184764,Fix Dynamo iterator reduce semantics,https://github.com/pytorch/pytorch/pull/184764,PR blocked,55,"failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-05-21T21:11:34Z
pr,184760,Fix Dynamo frozenset subclass dict-key hashing semantics,https://github.com/pytorch/pytorch/pull/184760,PR blocked,55,"failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-05-21T21:10:39Z
pr,184300,[xplat/caffe2] Add define_aten_lite macros for selective CUDA registration composed on top of pt_operator_registry,https://github.com/pytorch/pytorch/pull/184300,PR blocked,55,failed checks or blocked/unstable merge state; PR has not updated for at least 45 days,"identify whether block is CI, merge conflict, or review gate",2026-05-21T18:20:49Z
pr,184536,Skip `TestFullyShardSymmMem` for NCCL < 2.28,https://github.com/pytorch/pytorch/pull/184536,PR blocked,55,failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-05-21T17:39:56Z
pr,184673,Fix Dynamo dict fromkeys semantics,https://github.com/pytorch/pytorch/pull/184673,PR blocked,55,"failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-05-21T16:02:36Z
pr,184672,Fix Dynamo dict equality semantics,https://github.com/pytorch/pytorch/pull/184672,PR blocked,55,"failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-05-21T11:24:17Z
pr,184671,Fix Dynamo sequence search semantics,https://github.com/pytorch/pytorch/pull/184671,PR blocked,55,"failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-05-21T11:24:14Z
pr,184670,Fix Dynamo reversed iterator consumption,https://github.com/pytorch/pytorch/pull/184670,PR blocked,55,"failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-05-21T11:24:13Z
pr,184669,Fix Dynamo frozenset copy identity semantics,https://github.com/pytorch/pytorch/pull/184669,PR blocked,55,"failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-05-21T11:24:12Z
pr,184674,fixup! Add CPython Dynamo agentic loop plan,https://github.com/pytorch/pytorch/pull/184674,PR blocked,55,failed checks or blocked/unstable merge state; PR has not updated for at least 45 days,"identify whether block is CI, merge conflict, or review gate",2026-05-21T10:14:30Z
pr,88221,Add torch.tensor replacement and int_tensor prim,https://github.com/pytorch/pytorch/pull/88221,PR blocked,55,failed checks or blocked/unstable merge state; PR has not updated for at least 45 days,"identify whether block is CI, merge conflict, or review gate",2026-05-20T21:13:38Z
pr,184257,[scatter_add] Reject misaligned data_ptr / row stride in CuTeDSL override cond,https://github.com/pytorch/pytorch/pull/184257,PR blocked,55,failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-05-20T15:21:15Z
pr,179538,Fix incorrect type annotation in Future.set_exception (#179538),https://github.com/pytorch/pytorch/pull/179538,PR blocked,55,failed checks or blocked/unstable merge state; PR has not updated for at least 45 days,"identify whether block is CI, merge conflict, or review gate",2026-05-20T12:16:47Z
pr,184389,[ROCm] Fix test_origami autotune comparison after PR #181617 cap,https://github.com/pytorch/pytorch/pull/184389,PR blocked,55,"failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/inductor-rocm-mi200, ciflow/inductor-rocm-mi300, ciflow/inductor-rocm-mi355","identify whether block is CI, merge conflict, or review gate",2026-05-20T12:07:31Z
pr,181617,"Speed up inductor test infrastructure (~4x collection, ~1.7x execution)",https://github.com/pytorch/pytorch/pull/181617,PR blocked,55,"failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-05-20T12:07:30Z
pr,184401,Revert D105603961,https://github.com/pytorch/pytorch/pull/184401,PR blocked,55,"failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-05-19T22:13:51Z
pr,183592,[ROCm] op benchmark: use noble nightly image,https://github.com/pytorch/pytorch/pull/183592,PR blocked,55,"failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/nightly, ciflow/op-benchmark","identify whether block is CI, merge conflict, or review gate",2026-05-19T15:50:40Z
pr,184236,[inductor][overlap] Enable simple_overlap by default,https://github.com/pytorch/pytorch/pull/184236,PR blocked,55,"failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-05-19T15:36:52Z
pr,57772,Deprecate torch.svd and change svd -> linalg_svd,https://github.com/pytorch/pytorch/pull/57772,PR blocked,55,failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-05-19T12:26:08Z
pr,184307,Update SDPA error diagnostics,https://github.com/pytorch/pytorch/pull/184307,PR blocked,55,failed checks or blocked/unstable merge state; PR has not updated for at least 45 days,"identify whether block is CI, merge conflict, or review gate",2026-05-19T06:20:24Z
pr,183394,lint: ignore B904 for vendored code,https://github.com/pytorch/pytorch/pull/183394,PR blocked,55,failed checks or blocked/unstable merge state; PR has not updated for at least 45 days,"identify whether block is CI, merge conflict, or review gate",2026-05-19T01:17:01Z
pr,183651,[OSDC] Migrate h100-distributed.yml to OSDC (ARC) via dial-up pattern,https://github.com/pytorch/pytorch/pull/183651,PR blocked,55,failed checks or blocked/unstable merge state; PR has not updated for at least 45 days,"identify whether block is CI, merge conflict, or review gate",2026-05-18T17:35:59Z
pr,184194,[inductor] Add simple_overlap pass for defensive comm/compute overlap,https://github.com/pytorch/pytorch/pull/184194,PR blocked,55,"failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-05-18T15:44:01Z
pr,176857,[XPU] test_indexing test_reductions: Remove obsolete XPU skip guards after verification,https://github.com/pytorch/pytorch/pull/176857,PR blocked,55,"failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/trunk, ciflow/xpu","identify whether block is CI, merge conflict, or review gate",2026-05-18T09:17:54Z
pr,183953,[DO NOT MERGE] Point CI workflows to test-infra IPv6 branch,https://github.com/pytorch/pytorch/pull/183953,PR blocked,55,"failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/binaries, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-05-18T01:16:18Z
pr,183978,"[docs] Add .. deprecated:: directives to torch.{cuda,cpu}.amp public APIs",https://github.com/pytorch/pytorch/pull/183978,PR blocked,55,failed checks or blocked/unstable merge state; PR has not updated for at least 45 days,"identify whether block is CI, merge conflict, or review gate",2026-05-17T23:50:17Z
pr,183455,Port D104346887/PR 182675 for index large select (#183455),https://github.com/pytorch/pytorch/pull/183455,PR blocked,55,failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-05-16T10:01:13Z
pr,183993,"Back out ""Port D104346887/PR 182675 for index large select""",https://github.com/pytorch/pytorch/pull/183993,PR blocked,55,failed checks or blocked/unstable merge state; PR has not updated for at least 45 days,"identify whether block is CI, merge conflict, or review gate",2026-05-16T04:29:38Z
pr,179096,[Inductor] Add _scaled_mm_v2 lowering,https://github.com/pytorch/pytorch/pull/179096,PR blocked,55,"failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/b200, ciflow/h100, ciflow/inductor, ciflow/rocm-mi300","identify whether block is CI, merge conflict, or review gate",2026-05-16T01:31:44Z
pr,174893,[Inductor][NVGEMM] Enable Epilogue Fusions,https://github.com/pytorch/pytorch/pull/174893,PR blocked,55,"failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-05-16T01:31:05Z
pr,181909,[Inductor][Triton] Forward hint_override in fused Triton compile,https://github.com/pytorch/pytorch/pull/181909,PR blocked,55,"failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-05-16T01:26:00Z
pr,182730,[DTensor] Preserve placement for aten.clone.default,https://github.com/pytorch/pytorch/pull/182730,PR blocked,55,"failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/dtensor, ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-05-16T00:03:33Z
pr,183940,[Native DSL] Vendor kernelagent-oink library,https://github.com/pytorch/pytorch/pull/183940,PR blocked,55,failed checks or blocked/unstable merge state; PR has not updated for at least 45 days,"identify whether block is CI, merge conflict, or review gate",2026-05-15T21:47:45Z
pr,183941,[Native DSL] Oink-based cuteDSL RMSNorm,https://github.com/pytorch/pytorch/pull/183941,PR blocked,55,failed checks or blocked/unstable merge state; PR has not updated for at least 45 days,"identify whether block is CI, merge conflict, or review gate",2026-05-15T18:53:58Z
pr,182196,[dynamo][dicts] Fix delitem for dicts,https://github.com/pytorch/pytorch/pull/182196,PR blocked,55,"failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-05-15T07:54:17Z
pr,183487,Update XNNPACK submodule to 0b66c9f,https://github.com/pytorch/pytorch/pull/183487,PR blocked,55,failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-05-15T04:27:59Z
pr,183086,Remove unused 'type: ignore' suppressions in dynamo/inductor/functorch,https://github.com/pytorch/pytorch/pull/183086,PR blocked,55,"failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/b200, ciflow/h100, ciflow/inductor, ciflow/inductor-pallas","identify whether block is CI, merge conflict, or review gate",2026-05-15T03:01:29Z
pr,180545,fix AOTInductor fails to compile dynamic-shape CUDA gather pattern with ValueError: The argument '((0)) + 48' is not comparable,https://github.com/pytorch/pytorch/pull/180545,PR blocked,55,"failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-05-15T03:00:27Z
pr,183747,Adding new test_c10d_ops_privateuse1.py variant for distributed testing,https://github.com/pytorch/pytorch/pull/183747,PR blocked,55,failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/torchtitan,"identify whether block is CI, merge conflict, or review gate",2026-05-14T23:11:21Z
pr,170807,Adding Glossary,https://github.com/pytorch/pytorch/pull/170807,PR blocked,55,failed checks or blocked/unstable merge state; PR has not updated for at least 45 days,"identify whether block is CI, merge conflict, or review gate",2026-05-14T16:11:27Z
pr,183262,[CUDA] Expose cudaHostGetDevicePointer in cudart bindings,https://github.com/pytorch/pytorch/pull/183262,PR blocked,55,failed checks or blocked/unstable merge state; PR has not updated for at least 45 days,"identify whether block is CI, merge conflict, or review gate",2026-05-14T16:06:33Z
pr,182845,[AI Codemod][RlAutotuneFbsourceWearablesMemory] PackedLinearWeightsQnnp::apply_dynamic (qlinear_dynamic.cpp),https://github.com/pytorch/pytorch/pull/182845,PR blocked,55,failed checks or blocked/unstable merge state; PR has not updated for at least 45 days,"identify whether block is CI, merge conflict, or review gate",2026-05-14T16:03:43Z
pr,182846,fix: safely handle non-standard boolean values in the CUDA kernel,https://github.com/pytorch/pytorch/pull/182846,PR blocked,55,failed checks or blocked/unstable merge state; PR has not updated for at least 45 days,"identify whether block is CI, merge conflict, or review gate",2026-05-14T16:03:33Z
pr,182179,[CI] Floor known-slow file times when stale per-config data under-shards,https://github.com/pytorch/pytorch/pull/182179,PR blocked,55,"failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/rocm-mi200, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-05-14T05:59:22Z
pr,181281,[dynamo] Refactor IsInstanceBuiltinVariable,https://github.com/pytorch/pytorch/pull/181281,PR blocked,55,"failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-05-14T05:59:16Z
pr,181338,[dynamo] Refactor TypeBuiltinVariable,https://github.com/pytorch/pytorch/pull/181338,PR blocked,55,"failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-05-14T05:58:47Z
pr,181337,[dynamo] Refactor StrBuiltinVariable,https://github.com/pytorch/pytorch/pull/181337,PR blocked,55,"failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-05-14T05:58:46Z
pr,182598,[dynamo] Refactor builtins init static type check using _fn,https://github.com/pytorch/pytorch/pull/182598,PR blocked,55,"failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-05-14T05:58:45Z
pr,179863,Teach CSAN to detect caching allocator memory reuse races,https://github.com/pytorch/pytorch/pull/179863,PR blocked,55,failed checks or blocked/unstable merge state; PR has not updated for at least 45 days,"identify whether block is CI, merge conflict, or review gate",2026-05-14T05:15:44Z
pr,183611,[inductor] Record output alias for mutates_and_returns_first_arg fallbacks,https://github.com/pytorch/pytorch/pull/183611,PR blocked,55,"failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-05-13T23:28:38Z
pr,178360,Add device/backend agnostic support for distributed c10d/common tests,https://github.com/pytorch/pytorch/pull/178360,PR blocked,55,failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/torchtitan,"identify whether block is CI, merge conflict, or review gate",2026-05-13T23:26:57Z
pr,183384,[inductor] preserve scan carry strides with shape padding,https://github.com/pytorch/pytorch/pull/183384,PR blocked,55,"failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-05-13T21:58:12Z
pr,183550,[AI Codemod][RlAutotuneFbsourceWearablesMemory] PackedLinearWeightsQnnp::apply_dynamic (qlinear_dynamic.cpp),https://github.com/pytorch/pytorch/pull/183550,PR blocked,55,failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-05-13T18:32:00Z
pr,182398,"Revert ""[dynamo-wrapped] Skip all Dynamo wrapped tests for numpy (#182240)""",https://github.com/pytorch/pytorch/pull/182398,PR blocked,55,failed checks or blocked/unstable merge state; PR has not updated for at least 45 days,"identify whether block is CI, merge conflict, or review gate",2026-05-13T07:22:42Z
pr,183418,[Avocado][MI350X] Non-overlay compatibility for new AITER/CK overlays (#183418),https://github.com/pytorch/pytorch/pull/183418,PR blocked,55,failed checks or blocked/unstable merge state; PR has not updated for at least 45 days,"identify whether block is CI, merge conflict, or review gate",2026-05-12T18:10:44Z
pr,183404,test/dynamo: add CPython coverage planning docs,https://github.com/pytorch/pytorch/pull/183404,PR blocked,55,"failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-05-12T14:57:19Z
pr,182274,[docs] Generate doc for values in ScalingType and SwizzleType enum,https://github.com/pytorch/pytorch/pull/182274,PR blocked,55,failed checks or blocked/unstable merge state; PR has not updated for at least 45 days,"identify whether block is CI, merge conflict, or review gate",2026-05-12T13:55:53Z
pr,177726,Introduce test skip markers for Sandcastle,https://github.com/pytorch/pytorch/pull/177726,PR blocked,55,failed checks or blocked/unstable merge state; PR has not updated for at least 45 days,"identify whether block is CI, merge conflict, or review gate",2026-05-12T11:27:16Z
pr,162706,Fix accelerator-related skip-decorators for tests,https://github.com/pytorch/pytorch/pull/162706,PR blocked,55,failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-05-12T11:27:07Z
pr,179139,Report ACL version info,https://github.com/pytorch/pytorch/pull/179139,PR blocked,55,failed checks or blocked/unstable merge state; PR has not updated for at least 45 days,"identify whether block is CI, merge conflict, or review gate",2026-05-12T09:36:29Z
pr,183035,Fix arange int64 fractional-step zero handling on CPU and CUDA.,https://github.com/pytorch/pytorch/pull/183035,PR blocked,55,failed checks or blocked/unstable merge state; PR has not updated for at least 45 days,"identify whether block is CI, merge conflict, or review gate",2026-05-11T21:51:59Z
pr,181435,Add operator benchmark monitoring skill for daily regression detection,https://github.com/pytorch/pytorch/pull/181435,PR blocked,55,failed checks or blocked/unstable merge state; PR has not updated for at least 45 days,"identify whether block is CI, merge conflict, or review gate",2026-05-11T19:05:46Z
pr,177867,Add ASIMD_BF16 Vectorized class specialisation,https://github.com/pytorch/pytorch/pull/177867,PR blocked,55,failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-05-11T16:41:56Z
pr,181359,S390x: fixes for docker image build,https://github.com/pytorch/pytorch/pull/181359,PR blocked,55,failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-05-11T15:18:31Z
pr,182900,test: relax tolerance of test_Embedding_discontiguous_cuda,https://github.com/pytorch/pytorch/pull/182900,PR blocked,55,failed checks or blocked/unstable merge state; PR has not updated for at least 45 days,"identify whether block is CI, merge conflict, or review gate",2026-05-11T14:47:56Z
pr,183254,Add RISC-V native fp16 conversion paths,https://github.com/pytorch/pytorch/pull/183254,PR blocked,55,failed checks or blocked/unstable merge state; PR has not updated for at least 45 days,"identify whether block is CI, merge conflict, or review gate",2026-05-11T14:42:55Z
pr,181875,[ROCm][release/2.11] Fix test_transformers.py test suite for ROCm,https://github.com/pytorch/pytorch/pull/181875,PR blocked,55,failed checks or blocked/unstable merge state; PR has not updated for at least 45 days,"identify whether block is CI, merge conflict, or review gate",2026-05-11T11:46:55Z
pr,183105,Fix formatting in extensions.json,https://github.com/pytorch/pytorch/pull/183105,PR blocked,55,failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-05-10T20:02:14Z
pr,178245,Replace c10::call_once with static initialization,https://github.com/pytorch/pytorch/pull/178245,PR blocked,55,failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2026-05-10T02:45:08Z
pr,174025,Add test for CosineAnnealingLR when T_max = 0,https://github.com/pytorch/pytorch/pull/174025,PR blocked,55,failed checks or blocked/unstable merge state; PR has not updated for at least 45 days,"identify whether block is CI, merge conflict, or review gate",2026-05-05T15:53:30Z
pr,161254,Add `zero_grad` param for `optimizer.step`,https://github.com/pytorch/pytorch/pull/161254,PR blocked,55,failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-04-29T01:54:20Z
pr,170249,[do not review] Expose API to create ownership token for autograd.Function grad_fn,https://github.com/pytorch/pytorch/pull/170249,PR blocked,55,failed checks or blocked/unstable merge state; PR has not updated for at least 45 days,"identify whether block is CI, merge conflict, or review gate",2026-03-27T18:49:13Z
pr,170250,[do not review] Use new GradientEdge creation in PP,https://github.com/pytorch/pytorch/pull/170250,PR blocked,55,failed checks or blocked/unstable merge state; PR has not updated for at least 45 days,"identify whether block is CI, merge conflict, or review gate",2026-03-24T19:05:48Z
pr,130556,Allow pipeline stage to take non-tensor args and kwargs,https://github.com/pytorch/pytorch/pull/130556,PR blocked,55,failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/torchtitan,"identify whether block is CI, merge conflict, or review gate",2026-03-21T22:38:41Z
pr,167894,[inductor] split reduction even if all reads are broadcasted,https://github.com/pytorch/pytorch/pull/167894,PR blocked,55,"failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-03-06T22:02:57Z
pr,166732,Introducing statefulness for single process dataloader,https://github.com/pytorch/pytorch/pull/166732,PR blocked,55,failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-02-26T08:44:11Z
pr,157193,[c10d] Call destroy_process_group upon interpreter exit,https://github.com/pytorch/pytorch/pull/157193,PR blocked,55,failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-02-23T23:30:39Z
pr,140613,"Do not merge, do not close",https://github.com/pytorch/pytorch/pull/140613,PR blocked,55,failed checks or blocked/unstable merge state; PR has not updated for at least 45 days,"identify whether block is CI, merge conflict, or review gate",2026-01-11T02:18:04Z
pr,165800,[flex_attention]: adding a heuristic for USE_TMA,https://github.com/pytorch/pytorch/pull/165800,PR blocked,55,"failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/b200, ciflow/inductor, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2025-12-30T18:42:02Z
pr,164128,Consistently use c10_ovrsource in arvr mode everywhere,https://github.com/pytorch/pytorch/pull/164128,PR blocked,55,failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2025-12-29T04:45:55Z
pr,162033,[RFC] Switch all of c10d to use .h instead of .hpp,https://github.com/pytorch/pytorch/pull/162033,PR blocked,55,"failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/h100-symm-mem, ciflow/inductor","identify whether block is CI, merge conflict, or review gate",2025-12-21T03:17:44Z
pr,160888,Allow branching on whether you are in AC region with is_checkpoint_enabled,https://github.com/pytorch/pytorch/pull/160888,PR blocked,55,failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2025-12-05T01:43:53Z
pr,160878,[WIP] Introduce CachingDeviceAllocatorInterface as a base impl,https://github.com/pytorch/pytorch/pull/160878,PR blocked,55,failed checks or blocked/unstable merge state; PR has not updated for at least 45 days,"identify whether block is CI, merge conflict, or review gate",2025-11-27T01:35:47Z
pr,160872,[WIP] [3/N] Introduce a generic CachingDeviceAllocatorImpl for cross backend use,https://github.com/pytorch/pytorch/pull/160872,PR blocked,55,failed checks or blocked/unstable merge state; PR has not updated for at least 45 days,"identify whether block is CI, merge conflict, or review gate",2025-11-27T01:35:25Z
pr,160638,[WIP] [2/N] Introduce a generic CachingDeviceAllocatorImpl for cross backend use,https://github.com/pytorch/pytorch/pull/160638,PR blocked,55,failed checks or blocked/unstable merge state; PR has not updated for at least 45 days,"identify whether block is CI, merge conflict, or review gate",2025-11-27T01:35:02Z
pr,160182,[WIP] [1/N] Introduce a generic CachingDeviceAllocatorImpl for cross backend use,https://github.com/pytorch/pytorch/pull/160182,PR blocked,55,failed checks or blocked/unstable merge state; PR has not updated for at least 45 days,"identify whether block is CI, merge conflict, or review gate",2025-11-27T01:34:25Z
pr,148900,[RFC][BE] assume error checking is on by default (#141914),https://github.com/pytorch/pytorch/pull/148900,PR blocked,55,failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2025-11-22T02:02:11Z
pr,148668,Fix CUPTI lookup to include target directory,https://github.com/pytorch/pytorch/pull/148668,PR blocked,55,failed checks or blocked/unstable merge state; PR has not updated for at least 45 days,"identify whether block is CI, merge conflict, or review gate",2025-10-23T14:20:34Z
pr,112441,[RFC] Tensordict integration,https://github.com/pytorch/pytorch/pull/112441,PR blocked,55,failed checks or blocked/unstable merge state; PR has not updated for at least 45 days,"identify whether block is CI, merge conflict, or review gate",2025-05-28T21:34:32Z
pr,167224,[FlopCounter] Add skip_unsupported parameter for graceful handling of unsupported HOPs,https://github.com/pytorch/pytorch/pull/167224,waiting on contributor,53,review decision requests changes; PR closes or references open issue numbers,wait for contributor update,2026-07-09T01:53:29Z
pr,188006,[ROCm] Fix torch.pdist correctness on ROCm; re-enable test_pdist_norm_large,https://github.com/pytorch/pytorch/pull/188006,waiting on contributor,53,review decision requests changes; PR closes or references open issue numbers,wait for contributor update,2026-07-08T23:43:22Z
pr,181895,[CUDA] Fix linspace integer dtype precision mismatch with CPU,https://github.com/pytorch/pytorch/pull/181895,waiting on contributor,53,"review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/b200, ciflow/dtensor, ciflow/h100, ciflow/h100-symm-mem",wait for contributor update,2026-07-08T21:06:36Z
pr,189132,Register sspaddmm.out for XPU backend,https://github.com/pytorch/pytorch/pull/189132,waiting on contributor,53,review decision requests changes; PR closes or references open issue numbers,wait for contributor update,2026-07-08T20:39:59Z
pr,184121,Improve persistent tiled reduction defaults,https://github.com/pytorch/pytorch/pull/184121,waiting on contributor,53,"review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/inductor-perf-compare, ciflow/torchtitan",wait for contributor update,2026-07-08T19:39:04Z
pr,185150,Preserve OrderedDict outputs in FX tracing,https://github.com/pytorch/pytorch/pull/185150,waiting on contributor,53,review decision requests changes; PR closes or references open issue numbers,wait for contributor update,2026-07-08T19:12:06Z
pr,184319,[inductor] Align split reductions for tiled pointwise fusion,https://github.com/pytorch/pytorch/pull/184319,waiting on contributor,53,"review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-07-08T19:02:41Z
pr,185433,Fix FakeTensorMode handling for Inductor dynamic scalars,https://github.com/pytorch/pytorch/pull/185433,waiting on contributor,53,review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,wait for contributor update,2026-07-08T18:14:18Z
pr,184858,Support runtime ConstDict lookups for dynamic keys,https://github.com/pytorch/pytorch/pull/184858,waiting on contributor,53,"review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-07-08T17:31:22Z
pr,189215,[MPS] Add `largeTensorTest` to `test_group_norm_backward_large_input`,https://github.com/pytorch/pytorch/pull/189215,waiting on contributor,53,review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/mps,wait for contributor update,2026-07-08T17:28:51Z
pr,185984,Fix Dynamo out= resize for internal tensors,https://github.com/pytorch/pytorch/pull/185984,waiting on contributor,53,"review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-07-08T17:00:33Z
pr,184963,Fix FakeTensor shallow copy state,https://github.com/pytorch/pytorch/pull/184963,waiting on contributor,53,"review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-07-08T16:46:07Z
pr,184223,Run Inductor fallbacks below autograd,https://github.com/pytorch/pytorch/pull/184223,waiting on contributor,53,"review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-07-08T16:17:34Z
pr,189038,[MPS] Fix uint8 casting in MPS `copy_` kernels,https://github.com/pytorch/pytorch/pull/189038,waiting on contributor,53,review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/mps,wait for contributor update,2026-07-08T16:12:55Z
pr,184910,Reject direct aten.set in Dynamo,https://github.com/pytorch/pytorch/pull/184910,waiting on contributor,53,"review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-07-08T16:05:11Z
pr,185069,Fix AOTI Windows cross-compile package loading,https://github.com/pytorch/pytorch/pull/185069,waiting on contributor,53,"review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-07-08T16:01:31Z
pr,186897,Warn when custom op int schema specializes SymInts,https://github.com/pytorch/pytorch/pull/186897,waiting on contributor,53,review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,wait for contributor update,2026-07-08T15:53:41Z
pr,184380,Avoid fp32 materialization in checkpoint recompute,https://github.com/pytorch/pytorch/pull/184380,waiting on contributor,53,"review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-07-08T15:49:47Z
pr,185976,Fix Dynamo autograd.Function backward grad mode tracing,https://github.com/pytorch/pytorch/pull/185976,waiting on contributor,53,"review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-07-08T15:18:05Z
pr,186510,Fix Dynamo low-precision addc opmath,https://github.com/pytorch/pytorch/pull/186510,waiting on contributor,53,"review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-07-08T15:12:30Z
pr,184345,Use freezing compiler for traced inference graphs,https://github.com/pytorch/pytorch/pull/184345,waiting on contributor,53,review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,wait for contributor update,2026-07-08T13:19:57Z
pr,186397,Keep constant-false torch._check in compiled graph,https://github.com/pytorch/pytorch/pull/186397,waiting on contributor,53,"review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-07-08T11:51:17Z
pr,186420,Run torchrec tests in Dynamo CI,https://github.com/pytorch/pytorch/pull/186420,waiting on contributor,53,review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,wait for contributor update,2026-07-08T11:35:42Z
pr,172074,Fixes grad_dtype property being lost in fsdp,https://github.com/pytorch/pytorch/pull/172074,waiting on contributor,53,"review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-07-08T10:18:22Z
pr,184694,Fix AOTAutograd input overlap guards,https://github.com/pytorch/pytorch/pull/184694,waiting on contributor,53,review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,wait for contributor update,2026-07-08T09:35:56Z
pr,184573,Fix compile support for CTC loss tensor lengths,https://github.com/pytorch/pytorch/pull/184573,waiting on contributor,53,"review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-07-08T09:32:28Z
pr,184602,Preserve frozen ShapeEnv during fake tensor guards,https://github.com/pytorch/pytorch/pull/184602,waiting on contributor,53,review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,wait for contributor update,2026-07-08T09:05:31Z
pr,184632,Fix CUDA fake strides for mixed dtype pointwise ops,https://github.com/pytorch/pytorch/pull/184632,waiting on contributor,53,review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,wait for contributor update,2026-07-08T09:01:26Z
pr,184639,Support None returns from torch.cond,https://github.com/pytorch/pytorch/pull/184639,waiting on contributor,53,"review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-07-08T08:58:55Z
pr,185496,Fix export autograd.grad saved tensor tracing,https://github.com/pytorch/pytorch/pull/185496,waiting on contributor,53,review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,wait for contributor update,2026-07-08T08:44:24Z
pr,186444,Fix Adam capturable zero-lr NaNs,https://github.com/pytorch/pytorch/pull/186444,waiting on contributor,53,review decision requests changes; PR closes or references open issue numbers,wait for contributor update,2026-07-08T07:45:11Z
pr,185585,Fix mark_unbacked stride recompiles,https://github.com/pytorch/pytorch/pull/185585,waiting on contributor,53,"review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-07-08T07:13:24Z
pr,183871,[inductor] Match eval BatchNorm CUDA numerics,https://github.com/pytorch/pytorch/pull/183871,waiting on contributor,53,"review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-07-08T07:10:13Z
pr,184009,Handle SymPy boolean Inductor graph inputs,https://github.com/pytorch/pytorch/pull/184009,waiting on contributor,53,"review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-07-08T06:43:00Z
pr,188966,Add operator.concat/iconcat support to dynamo builtins,https://github.com/pytorch/pytorch/pull/188966,waiting on contributor,53,"review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-07-08T06:17:57Z
pr,185064,Add unroll option to scan,https://github.com/pytorch/pytorch/pull/185064,waiting on contributor,53,"review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-07-08T06:15:05Z
pr,186764,fix: Use typing_extensions.TypeAliasType for better reexport of `__module__`,https://github.com/pytorch/pytorch/pull/186764,waiting on contributor,53,review decision requests changes; PR closes or references open issue numbers,wait for contributor update,2026-07-08T05:48:41Z
pr,187908,[fix] torch.where silently overflows fp16 scalars (issue #187429),https://github.com/pytorch/pytorch/pull/187908,waiting on contributor,53,review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/trunk,wait for contributor update,2026-07-08T05:20:33Z
pr,173948,Fixes LazyConvTranspose modules when used with torch.compile(dynamic=True),https://github.com/pytorch/pytorch/pull/173948,waiting on contributor,53,"review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-07-08T04:51:20Z
pr,176072,Fix Triton launcher arg mismatch silent failure,https://github.com/pytorch/pytorch/pull/176072,waiting on contributor,53,"review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk",wait for contributor update,2026-07-08T04:50:40Z
pr,184938,Fix AOTAutograd forward output overwritten by backward mutation,https://github.com/pytorch/pytorch/pull/184938,waiting on contributor,53,review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,wait for contributor update,2026-07-08T04:26:00Z
pr,183881,Fix online softmax epilogue fusion for T5,https://github.com/pytorch/pytorch/pull/183881,waiting on contributor,53,"review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-07-08T04:09:02Z
pr,183944,Fix compiled autograd clear saved tensors semantics,https://github.com/pytorch/pytorch/pull/183944,waiting on contributor,53,"review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-07-08T04:01:54Z
pr,183954,Disallow mid-trace reference opaque creation,https://github.com/pytorch/pytorch/pull/183954,waiting on contributor,53,"review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk",wait for contributor update,2026-07-08T03:56:03Z
pr,186366,Support vmapped compiled autograd.Functions,https://github.com/pytorch/pytorch/pull/186366,waiting on contributor,53,review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,wait for contributor update,2026-07-08T03:44:03Z
pr,185764,Recommend nonstrict_trace for skipped C extensions,https://github.com/pytorch/pytorch/pull/185764,waiting on contributor,53,review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,wait for contributor update,2026-07-08T02:39:09Z
pr,183969,Fix CSE across unrelated mutation regions,https://github.com/pytorch/pytorch/pull/183969,waiting on contributor,53,review decision requests changes; PR closes or references open issue numbers,wait for contributor update,2026-07-08T02:36:28Z
pr,184958,Prevent escaped vjp TensorWrappers from reaching AOTAutograd,https://github.com/pytorch/pytorch/pull/184958,waiting on contributor,53,"review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-07-08T02:20:24Z
pr,180689,[ROCm] Handle profiler trace format differences in combo kernel tests,https://github.com/pytorch/pytorch/pull/180689,waiting on contributor,53,"review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/rocm-mi300",wait for contributor update,2026-07-08T02:18:49Z
pr,184961,Fix packed RNN export fake metadata,https://github.com/pytorch/pytorch/pull/184961,waiting on contributor,53,review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,wait for contributor update,2026-07-08T02:15:57Z
pr,184970,Preserve pin_memory in generated view copy kernels,https://github.com/pytorch/pytorch/pull/184970,waiting on contributor,53,review decision requests changes; PR closes or references open issue numbers,wait for contributor update,2026-07-08T02:05:15Z
pr,184051,Honor custom op layout constraints in fallbacks,https://github.com/pytorch/pytorch/pull/184051,waiting on contributor,53,"review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk",wait for contributor update,2026-07-08T01:40:54Z
pr,185017,Fix OpOverloadPacket tracing during initialization,https://github.com/pytorch/pytorch/pull/185017,waiting on contributor,53,review decision requests changes; PR closes or references open issue numbers,wait for contributor update,2026-07-08T00:45:01Z
pr,184481,Guard synthetic-base input alias layouts,https://github.com/pytorch/pytorch/pull/184481,waiting on contributor,53,"review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-07-08T00:38:07Z
pr,185035,Make symbolic formatting guard data-dependent scalars,https://github.com/pytorch/pytorch/pull/185035,waiting on contributor,53,"review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-07-08T00:23:05Z
pr,185043,Functionalize mutations in deferred local_map,https://github.com/pytorch/pytorch/pull/185043,waiting on contributor,53,review decision requests changes; PR closes or references open issue numbers,wait for contributor update,2026-07-08T00:08:21Z
pr,184944,Make OpaqueBase usable as a pybind base,https://github.com/pytorch/pytorch/pull/184944,waiting on contributor,53,"review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/dtensor, ciflow/inductor",wait for contributor update,2026-07-07T23:19:14Z
pr,185124,Fix prims.broadcast_in_dim functionalization,https://github.com/pytorch/pytorch/pull/185124,waiting on contributor,53,review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,wait for contributor update,2026-07-07T23:15:13Z
pr,185171,Document activation memory budget config,https://github.com/pytorch/pytorch/pull/185171,waiting on contributor,53,review decision requests changes; PR closes or references open issue numbers,wait for contributor update,2026-07-07T22:07:35Z
pr,185274,Fix ci expected accuracy updates without ClickHouse credentials,https://github.com/pytorch/pytorch/pull/185274,waiting on contributor,53,review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,wait for contributor update,2026-07-07T21:49:29Z
pr,187161,[FSDP] Cache post-forward DeviceMesh to avoid duplicate NCCL communicators per layer,https://github.com/pytorch/pytorch/pull/187161,waiting on contributor,53,"review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-07-07T20:54:49Z
pr,185075,Improve FakeTensorMode deepcopy error for real tensors,https://github.com/pytorch/pytorch/pull/185075,waiting on contributor,53,review decision requests changes; PR closes or references open issue numbers,wait for contributor update,2026-07-07T20:32:17Z
pr,182972,added graph break on exception based on if user code would catch it,https://github.com/pytorch/pytorch/pull/182972,waiting on contributor,53,"review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/dynamo, ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-07-07T19:53:45Z
pr,188605,Fix infinite recursion with pybind11 enum conversions,https://github.com/pytorch/pytorch/pull/188605,waiting on contributor,53,"review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-07-07T19:50:37Z
pr,185299,Run CSE in AOT inference graphs,https://github.com/pytorch/pytorch/pull/185299,waiting on contributor,53,review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,wait for contributor update,2026-07-07T19:20:47Z
pr,184052,Preserve CIA ops in AOT joint export,https://github.com/pytorch/pytorch/pull/184052,waiting on contributor,53,review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,wait for contributor update,2026-07-07T19:06:21Z
pr,185347,Support torch.compile under user FakeTensorMode,https://github.com/pytorch/pytorch/pull/185347,waiting on contributor,53,"review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-07-07T18:51:04Z
pr,185353,Normalize export state names for subclass decomposition,https://github.com/pytorch/pytorch/pull/185353,waiting on contributor,53,review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,wait for contributor update,2026-07-07T18:42:42Z
pr,185892,Improve export device mismatch errors,https://github.com/pytorch/pytorch/pull/185892,waiting on contributor,53,"review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-07-07T17:52:17Z
pr,184061,Fix fused moving avg observer functional meta,https://github.com/pytorch/pytorch/pull/184061,waiting on contributor,53,"review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-07-07T17:40:44Z
pr,184985,Fix nested compile handling in AOT joint export,https://github.com/pytorch/pytorch/pull/184985,waiting on contributor,53,"review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-07-07T17:10:59Z
pr,188994,[dynamic shapes] Preserve algebraic relationships in foreign unbacked SymInt transfer (#188723),https://github.com/pytorch/pytorch/pull/188994,waiting on contributor,53,review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,wait for contributor update,2026-07-07T17:08:55Z
pr,184312,Fix efficient zero tensor handling in torch.compile,https://github.com/pytorch/pytorch/pull/184312,waiting on contributor,53,"review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk",wait for contributor update,2026-07-07T17:08:27Z
pr,188948,Fix triangular_solve for sparse CPU tensors on non-MKL platforms,https://github.com/pytorch/pytorch/pull/188948,waiting on contributor,53,review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,wait for contributor update,2026-07-07T17:06:02Z
pr,184016,Fix Inductor DCE for fallback alias mutations,https://github.com/pytorch/pytorch/pull/184016,waiting on contributor,53,"review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-07-07T16:32:58Z
pr,185499,Fix unbacked memo reuse during decomposition retrace,https://github.com/pytorch/pytorch/pull/185499,waiting on contributor,53,review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,wait for contributor update,2026-07-07T16:25:41Z
pr,184091,Handle cudagraph liveness changes during replay,https://github.com/pytorch/pytorch/pull/184091,waiting on contributor,53,"review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-07-07T15:58:50Z
pr,184092,Skip tiny cudagraph partitions in max-autotune,https://github.com/pytorch/pytorch/pull/184092,waiting on contributor,53,"review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk",wait for contributor update,2026-07-07T15:54:29Z
pr,185690,Optimize AOT input view mutations,https://github.com/pytorch/pytorch/pull/185690,waiting on contributor,53,"review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/inductor-pallas",wait for contributor update,2026-07-07T15:06:00Z
pr,189043,Preload full bundled cuDNN set with RTLD_GLOBAL to prevent sublibrary version mismatch,https://github.com/pytorch/pytorch/pull/189043,waiting on contributor,53,review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/binaries,wait for contributor update,2026-07-07T13:27:21Z
pr,185759,Improve float32 flex attention backward tiling,https://github.com/pytorch/pytorch/pull/185759,waiting on contributor,53,review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,wait for contributor update,2026-07-07T13:11:26Z
pr,184130,Avoid torch imports in Inductor compile sidecar,https://github.com/pytorch/pytorch/pull/184130,waiting on contributor,53,"review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-07-07T12:24:05Z
pr,183904,Fix CPU addcmul tensor-value decomposition numerics,https://github.com/pytorch/pytorch/pull/183904,waiting on contributor,53,"review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/dtensor, ciflow/inductor, ciflow/torchtitan, ciflow/trunk",wait for contributor update,2026-07-07T11:39:48Z
pr,186522,Trace Tensor.requires_grad_ in pre-dispatch make_fx,https://github.com/pytorch/pytorch/pull/186522,waiting on contributor,53,review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,wait for contributor update,2026-07-07T11:04:20Z
pr,187633,Enhance documentation around the developer build,https://github.com/pytorch/pytorch/pull/187633,waiting on contributor,53,review decision requests changes; PR closes or references open issue numbers,wait for contributor update,2026-07-07T10:54:24Z
pr,184153,[Inductor] Clone cudagraph-owned grads before replay,https://github.com/pytorch/pytorch/pull/184153,waiting on contributor,53,"review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-07-07T10:47:22Z
pr,184161,Fix Inductor lazy negative input loads,https://github.com/pytorch/pytorch/pull/184161,waiting on contributor,53,"review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/inductor-pallas, ciflow/torchtitan",wait for contributor update,2026-07-07T10:43:09Z
pr,185899,Fix Dynamo data mutation on functorch tensors,https://github.com/pytorch/pytorch/pull/185899,waiting on contributor,53,"review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-07-07T08:50:40Z
pr,184336,fix RPATH in libtorch extraction from wheel,https://github.com/pytorch/pytorch/pull/184336,waiting on contributor,53,review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/binaries,wait for contributor update,2026-07-07T08:21:16Z
pr,186160,Fix FakeTensor forward AD batch_norm meta rewrapping,https://github.com/pytorch/pytorch/pull/186160,waiting on contributor,53,review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,wait for contributor update,2026-07-07T06:32:34Z
pr,184621,Fix Dynamo torch.Size from tensor shape inputs,https://github.com/pytorch/pytorch/pull/184621,waiting on contributor,53,"review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-07-07T05:17:12Z
pr,184997,Fix TVM backend scalar tensor inputs,https://github.com/pytorch/pytorch/pull/184997,waiting on contributor,53,"review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-07-07T03:30:38Z
pr,178493,[c10d][NCCL] avoid reusing scalable init store keys across communicat…,https://github.com/pytorch/pytorch/pull/178493,waiting on contributor,53,review decision requests changes; PR closes or references open issue numbers,wait for contributor update,2026-07-07T02:20:16Z
pr,181106,Fix torch.compile crash with inference_mode and view ops in forward (#180823),https://github.com/pytorch/pytorch/pull/181106,waiting on contributor,53,"review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/trunk",wait for contributor update,2026-07-06T22:13:47Z
pr,188249,[inductor] Fix bool equality value range analysis,https://github.com/pytorch/pytorch/pull/188249,waiting on contributor,53,review decision requests changes; PR closes or references open issue numbers,wait for contributor update,2026-07-06T21:30:13Z
pr,184303,Fix pattern matcher replacement graph cleanup safety,https://github.com/pytorch/pytorch/pull/184303,waiting on contributor,53,"review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-07-06T18:21:54Z
pr,184310,Fix flex attention shared memory config fallback,https://github.com/pytorch/pytorch/pull/184310,waiting on contributor,53,"review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-07-06T17:10:03Z
pr,189019,[MPS] Add torch.mps.MetalGraph capture/replay API,https://github.com/pytorch/pytorch/pull/189019,waiting on contributor,53,review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/mps,wait for contributor update,2026-07-06T14:46:00Z
pr,188748,[MPS] Make F.linear honor PYTORCH_MPS_PREFER_METAL,https://github.com/pytorch/pytorch/pull/188748,waiting on contributor,53,review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/mps,wait for contributor update,2026-07-06T13:00:28Z
pr,181665,Fix UBSan in weight_norm CPU kernel for empty input (#181510),https://github.com/pytorch/pytorch/pull/181665,waiting on contributor,53,review decision requests changes; PR closes or references open issue numbers,wait for contributor update,2026-07-06T02:20:21Z
pr,182276,[inductor] Unwrap ShapeAsConstantBuffer in tensor_constructor size,https://github.com/pytorch/pytorch/pull/182276,waiting on contributor,53,"review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-07-06T01:09:40Z
pr,188952,Fix C2872 'std' ambiguous symbol when nvcc compiles compiled_autograd.h on Windows extension builds,https://github.com/pytorch/pytorch/pull/188952,waiting on contributor,53,review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,wait for contributor update,2026-07-04T21:09:29Z
pr,188473,[MPS] Add complex64 support to linalg.lu_factor (unblocks complex det/slogdet/lu),https://github.com/pytorch/pytorch/pull/188473,waiting on contributor,53,review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/mps,wait for contributor update,2026-07-04T19:12:03Z
pr,187477,Add context manager support to torch.dtype,https://github.com/pytorch/pytorch/pull/187477,waiting on contributor,53,review decision requests changes; PR closes or references open issue numbers,wait for contributor update,2026-07-04T15:13:41Z
pr,178591,fixes:division by zero check for inductor,https://github.com/pytorch/pytorch/pull/178591,waiting on contributor,53,"review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-07-04T02:27:03Z
pr,177457,[DTensor] Coalesce collectives for multi-arg redistribution,https://github.com/pytorch/pytorch/pull/177457,waiting on contributor,53,review decision requests changes; PR closes or references open issue numbers,wait for contributor update,2026-07-03T21:14:35Z
pr,174718,[inductor] fix allocation with deterministic guard,https://github.com/pytorch/pytorch/pull/174718,waiting on contributor,53,"review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk",wait for contributor update,2026-07-02T20:30:43Z
pr,188801,[MPS] Leak MetalShaderLibrary bundled singleton to avoid exit-time destructor crashes,https://github.com/pytorch/pytorch/pull/188801,waiting on contributor,53,review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/mps,wait for contributor update,2026-07-02T19:51:02Z
pr,180675,Add test for autograd.Function fused fwd/bwd pattern (issue #180642),https://github.com/pytorch/pytorch/pull/180675,waiting on contributor,53,review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,wait for contributor update,2026-07-02T18:43:42Z
pr,187572,[Metal][ops] add EmbeddingBag offsets validation (#170370),https://github.com/pytorch/pytorch/pull/187572,waiting on contributor,53,review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/mps,wait for contributor update,2026-07-02T15:35:51Z
pr,180353,[MPS] Replace isin MPSGraph implementation with native Metal kernels,https://github.com/pytorch/pytorch/pull/180353,waiting on contributor,53,review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/mps,wait for contributor update,2026-07-01T17:56:01Z
pr,185831,Use minimum-norm subgradient for clamp/clamp_min/clamp_max at boundary,https://github.com/pytorch/pytorch/pull/185831,waiting on contributor,53,review decision requests changes; PR closes or references open issue numbers,wait for contributor update,2026-07-01T01:54:03Z
pr,188297,"Validate target range [0, 1] in binary_cross_entropy_with_logits",https://github.com/pytorch/pytorch/pull/188297,waiting on contributor,53,review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,wait for contributor update,2026-07-01T01:53:17Z
pr,186542,Support joint export with unused parameter grads,https://github.com/pytorch/pytorch/pull/186542,waiting on contributor,53,review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,wait for contributor update,2026-07-01T01:47:15Z
pr,188092,Fix cartesian_prod returning inconsistent shape for single tensor input,https://github.com/pytorch/pytorch/pull/188092,waiting on contributor,53,review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/trunk,wait for contributor update,2026-06-29T09:22:11Z
pr,187955,Make expandable-segment Handle a move-only RAII owner and exception-safe,https://github.com/pytorch/pytorch/pull/187955,waiting on contributor,53,review decision requests changes; PR closes or references open issue numbers,wait for contributor update,2026-06-29T01:21:43Z
pr,187316,Support save_on_cpu offloading under first-order torch.func transforms,https://github.com/pytorch/pytorch/pull/187316,waiting on contributor,53,"review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-06-27T05:24:16Z
pr,183980,[cpu][cuda] Extend native_group_norm_backward to handle grad from all outputs,https://github.com/pytorch/pytorch/pull/183980,waiting on contributor,53,"review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/mps, ciflow/trunk",wait for contributor update,2026-06-25T20:55:12Z
pr,187663,Avoid oversized storage for reduced mse_loss / smooth_l1_loss,https://github.com/pytorch/pytorch/pull/187663,waiting on contributor,53,review decision requests changes; PR closes or references open issue numbers,wait for contributor update,2026-06-25T07:25:40Z
pr,185519,Fix ZeroTensor meta redispatch for nested hessian compile,https://github.com/pytorch/pytorch/pull/185519,waiting on contributor,53,review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,wait for contributor update,2026-06-24T21:19:41Z
pr,174862,[FSDP2] Fix to differing gradient computation graphs in different ranks,https://github.com/pytorch/pytorch/pull/174862,waiting on contributor,53,"review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-06-24T19:51:16Z
pr,185163,[aot_autograd] Handle None in saved tensors for LSTM/RNN backward; enable allow_rnn by default,https://github.com/pytorch/pytorch/pull/185163,waiting on contributor,53,"review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/b200, ciflow/docker, ciflow/dtensor, ciflow/h100",wait for contributor update,2026-06-24T19:14:57Z
pr,186766,fix: [BE] Deduplicate auto_functionalized and triton_kernel_wrapper_functi...,https://github.com/pytorch/pytorch/pull/186766,waiting on contributor,53,review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,wait for contributor update,2026-06-23T19:23:30Z
pr,183508,Chain-based dispatch for custom_op fast path,https://github.com/pytorch/pytorch/pull/183508,waiting on contributor,53,review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/trunk,wait for contributor update,2026-06-22T22:53:24Z
pr,172727,Fix torch.topk out variant recompilations by adding meta function for…,https://github.com/pytorch/pytorch/pull/172727,waiting on contributor,53,review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,wait for contributor update,2026-06-22T15:22:24Z
pr,187260,Fix static cudnn building for cudnn v9,https://github.com/pytorch/pytorch/pull/187260,waiting on contributor,53,review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/binaries,wait for contributor update,2026-06-22T00:32:51Z
pr,187247,Type _add_docstr so type checkers see spectral ops as callable,https://github.com/pytorch/pytorch/pull/187247,waiting on contributor,53,review decision requests changes; PR closes or references open issue numbers,wait for contributor update,2026-06-21T16:33:20Z
pr,175490,Fix CPU `NotImplementedError` for bf16/fp16 antialiased upsampling (`interpolate`),https://github.com/pytorch/pytorch/pull/175490,waiting on contributor,53,review decision requests changes; PR closes or references open issue numbers,wait for contributor update,2026-06-20T01:26:27Z
pr,185355,[pipelining] Fix backward hook accumulation on recv buffer activations,https://github.com/pytorch/pytorch/pull/185355,waiting on contributor,53,"review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/torchtitan, ciflow/trunk",wait for contributor update,2026-06-19T07:58:02Z
pr,186445,[AOTI] Avoid serializing full storage for shared-storage constants,https://github.com/pytorch/pytorch/pull/186445,waiting on contributor,53,"review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-06-18T17:12:23Z
pr,187323,Fix Tensor.to() ignoring memory_format for non-contiguous tensors,https://github.com/pytorch/pytorch/pull/187323,waiting on contributor,53,review decision requests changes; PR closes or references open issue numbers,wait for contributor update,2026-06-18T07:16:17Z
pr,187558,Fix torch.sign to propagate NaN instead of returning 0,https://github.com/pytorch/pytorch/pull/187558,waiting on contributor,53,"review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-06-18T05:05:22Z
pr,182794,[inductor][rocm] Fix AOTInductor autotune int64 Triton kernel test and enable FP8 platform-aware handling,https://github.com/pytorch/pytorch/pull/182794,waiting on contributor,53,"review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/trunk",wait for contributor update,2026-06-17T21:24:04Z
pr,181465,Fix torch.fx.Tracer.record_stack_traces regression (user frames filtered by internal PyTorch frames),https://github.com/pytorch/pytorch/pull/181465,waiting on contributor,53,review decision requests changes; PR closes or references open issue numbers,wait for contributor update,2026-06-17T21:23:24Z
pr,186579,Fix mse_loss/smooth_l1_loss allocating oversized storage for scalar output,https://github.com/pytorch/pytorch/pull/186579,waiting on contributor,53,review decision requests changes; PR closes or references open issue numbers,wait for contributor update,2026-06-17T21:10:39Z
pr,187154,Use TORCH_INSTALL_LIB_DIR instead of hardcoded path 'lib',https://github.com/pytorch/pytorch/pull/187154,waiting on contributor,53,review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,wait for contributor update,2026-06-17T21:06:49Z
pr,184680,Inductor Lowering of Trunc Div should use div_rn,https://github.com/pytorch/pytorch/pull/184680,waiting on contributor,53,"review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk",wait for contributor update,2026-06-17T12:01:27Z
pr,186543,Fix Windows debug build issues,https://github.com/pytorch/pytorch/pull/186543,waiting on contributor,53,"review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/xpu",wait for contributor update,2026-06-17T05:11:10Z
pr,185645,Add batched small-matrix CUDA kernel for linalg.eig,https://github.com/pytorch/pytorch/pull/185645,waiting on contributor,53,review decision requests changes; PR closes or references open issue numbers,wait for contributor update,2026-06-16T15:55:48Z
pr,187106,[inductor] decompose broadcast-bias baddbmm into bmm + pointwise add,https://github.com/pytorch/pytorch/pull/187106,waiting on contributor,53,"review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-06-15T22:15:26Z
pr,185977,Fix index_add out-of-bounds parity between eager and inductor,https://github.com/pytorch/pytorch/pull/185977,waiting on contributor,53,"review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/dtensor, ciflow/inductor, ciflow/mps, ciflow/torchtitan",wait for contributor update,2026-06-15T18:47:12Z
pr,175964,Add validation for embedding_bag offsets to prevent illegal memory access,https://github.com/pytorch/pytorch/pull/175964,waiting on contributor,53,review decision requests changes; PR closes or references open issue numbers,wait for contributor update,2026-06-15T10:35:27Z
pr,185157,[Inductor] Fix CPU bool argmin/argmax incorrect indices,https://github.com/pytorch/pytorch/pull/185157,waiting on contributor,53,"review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk",wait for contributor update,2026-06-15T08:42:07Z
pr,184249,Fix extern autotune tensor kwargs,https://github.com/pytorch/pytorch/pull/184249,waiting on contributor,53,"review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-06-11T23:28:09Z
pr,186776,Fix __qualname__ for OpOverload and OpOverloadPacket,https://github.com/pytorch/pytorch/pull/186776,waiting on contributor,53,review decision requests changes; PR closes or references open issue numbers,wait for contributor update,2026-06-11T01:17:39Z
pr,185765,[Inductor] Fix correctness discrepancy in torch.minimum and torch.maximum with signed zeros (-0.0 vs 0.0),https://github.com/pytorch/pytorch/pull/185765,waiting on contributor,53,"review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-06-10T19:47:04Z
pr,182980,Clear exception tracebacks to avoid reference cycles,https://github.com/pytorch/pytorch/pull/182980,waiting on contributor,53,"review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/docker, ciflow/dtensor, ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-06-10T14:48:54Z
pr,186205,[dynamo] Add as_strided_ handler to fix metadata propagation (#185888),https://github.com/pytorch/pytorch/pull/186205,waiting on contributor,53,"review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-06-08T13:32:56Z
pr,186079,Add test cases for 0-sized tensor padding,https://github.com/pytorch/pytorch/pull/186079,waiting on contributor,53,review decision requests changes; PR closes or references open issue numbers,wait for contributor update,2026-06-05T11:10:05Z
pr,183691,[Test] Add tolerance and precision override mechanisms for parameteried tests,https://github.com/pytorch/pytorch/pull/183691,waiting on contributor,53,review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/trunk,wait for contributor update,2026-06-04T08:32:40Z
pr,185927,Fix clamp boundary subgradients,https://github.com/pytorch/pytorch/pull/185927,waiting on contributor,53,review decision requests changes; PR closes or references open issue numbers,wait for contributor update,2026-06-03T22:01:22Z
pr,182618,[Profiler] Forward declare nlohmann::json in public header,https://github.com/pytorch/pytorch/pull/182618,waiting on contributor,53,review decision requests changes; PR closes or references open issue numbers,wait for contributor update,2026-06-02T01:31:40Z
pr,185361,[pooling] Reject negative divisor_override in avg_pool2d/3d,https://github.com/pytorch/pytorch/pull/185361,waiting on contributor,53,"review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/mps, ciflow/torchtitan",wait for contributor update,2026-06-01T01:37:44Z
pr,185263,Fix torch.compile normal tensor std validation,https://github.com/pytorch/pytorch/pull/185263,waiting on contributor,53,"review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk",wait for contributor update,2026-05-31T05:25:25Z
pr,185048,[Bugfix] fix `torch.library._register_fake` crashes when any unrelated `LazyLoader` module in `sys.modules` fails to import,https://github.com/pytorch/pytorch/pull/185048,waiting on contributor,53,review decision requests changes; PR closes or references open issue numbers,wait for contributor update,2026-05-30T16:17:35Z
pr,177823,[Inductor] Cap num_warps for non-1x1 Triton conv kernels to fix incorrect results,https://github.com/pytorch/pytorch/pull/177823,waiting on contributor,53,"review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-05-29T21:44:31Z
pr,183535,[MPS] fix older gpu gen mac matmul failures,https://github.com/pytorch/pytorch/pull/183535,waiting on contributor,53,review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/mps,wait for contributor update,2026-05-29T16:12:23Z
pr,185256,Fix PendingUnbackedSymbolNotFound for tensor-derived slice bounds,https://github.com/pytorch/pytorch/pull/185256,waiting on contributor,53,"review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-05-29T15:38:10Z
pr,181526,Replace obsolete TORCHDYNAMO_DYNAMIC_SHAPES env var in torch.compile FAQ,https://github.com/pytorch/pytorch/pull/181526,waiting on contributor,53,review decision requests changes; PR closes or references open issue numbers,wait for contributor update,2026-05-29T01:46:26Z
pr,183023,[Inductor] Filter triton configs by launch geometry in triton heuristics,https://github.com/pytorch/pytorch/pull/183023,waiting on contributor,53,"review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/xpu",wait for contributor update,2026-05-28T19:17:24Z
pr,184531,[Inductor] Tolerance increase for test_bmm_large_batch_dynamic_cuda,https://github.com/pytorch/pytorch/pull/184531,waiting on contributor,53,"review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/inductor-rocm-mi200, ciflow/inductor-rocm-mi300, ciflow/inductor-rocm-mi355",wait for contributor update,2026-05-27T12:58:53Z
pr,179830,Fix upsample_nearest3d stride mismatch under torch.compile (#179272),https://github.com/pytorch/pytorch/pull/179830,waiting on contributor,53,"review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/inductor-pallas, ciflow/torchtitan",wait for contributor update,2026-05-26T17:22:28Z
pr,184483,Make DTensor.local_map pytree-aware on in/out_placements,https://github.com/pytorch/pytorch/pull/184483,draft/noise,53,"PR is draft; failed checks or blocked/unstable merge state; PR closes or references open issue numbers; PR has not updated for at least 45 days; CI lane labels: ciflow/dtensor, ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-05-20T08:45:10Z
pr,183029,[make_fx] Fix wrong 2nd-order gradients by decomposing detach to alias,https://github.com/pytorch/pytorch/pull/183029,draft/noise,53,"PR is draft; failed checks or blocked/unstable merge state; PR closes or references open issue numbers; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-05-11T17:39:34Z
pr,175969,[inductor][xpu] Fix convolution_backward accuracy for mixed channels-last / contiguous inputs,https://github.com/pytorch/pytorch/pull/175969,draft/noise,53,"PR is draft; failed checks or blocked/unstable merge state; PR closes or references open issue numbers; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk, ciflow/xpu","identify whether block is CI, merge conflict, or review gate",2026-05-11T07:32:58Z
issue,184878,Division by Zero Floating Point Exception in monitor aggregation,https://github.com/pytorch/pytorch/issues/184878,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: edge cases, module: error checking",review early; this can change release or regression risk,2026-06-01T17:14:12Z
issue,172011,torch.pixel_shuffle CPU-GPU crash with massive upscale factor,https://github.com/pytorch/pytorch/issues/172011,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: edge cases, module: error checking",review early; this can change release or regression risk,2026-01-09T20:15:00Z
issue,128691,Segmentation fault (core dumped) in `torch._remove_batch_dim`,https://github.com/pytorch/pytorch/issues/128691,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: edge cases",review early; this can change release or regression risk,2026-01-07T18:55:32Z
issue,150187,at::BlasBackend::Ck does not handle all ROCm BLAS gpus,https://github.com/pytorch/pytorch/issues/150187,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: regression, module: rocm",review early; this can change release or regression risk,2026-01-07T18:05:07Z
issue,68320,Potential race conditions between multiple workers trying to download and cache the same file in torch.hub.load_state_dict_from_url and torch.hub.download_url_to_file <- duplicate dataset/model downloads across DDP workers,https://github.com/pytorch/pytorch/issues/68320,high-priority/blocker,52,high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: hub,review early; this can change release or regression risk,2026-01-02T10:36:36Z
issue,154312,torch.logdet produces incorrect results for singular matrices on CUDA vs CPU,https://github.com/pytorch/pytorch/issues/154312,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: cuda, module: linear algebra, module: nans and infs",review early; this can change release or regression risk,2025-12-30T20:35:27Z
issue,171284,GRU: Segmentation fault (core dumped),https://github.com/pytorch/pytorch/issues/171284,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: error checking, module: rnn",review early; this can change release or regression risk,2025-12-25T12:05:09Z
issue,164402,NCCL-2.28.3 build locally is unusable on H100,https://github.com/pytorch/pytorch/issues/164402,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: build, module: ci, module: nccl, module: regression",review early; this can change release or regression risk,2025-12-24T22:50:02Z
issue,48365,Custom exception for out of memory,https://github.com/pytorch/pytorch/issues/48365,high-priority/blocker,52,high-priority/blocker/regression label present; old update age or low-urgency/stale label,review early; this can change release or regression risk,2025-12-22T16:03:51Z
issue,67237,[docs] Google finds docs pages that give 404,https://github.com/pytorch/pytorch/issues/67237,high-priority/blocker,52,high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: docs,review early; this can change release or regression risk,2025-12-16T22:18:05Z
issue,164957,Something broke `from functorch import make_fx` on main,https://github.com/pytorch/pytorch/issues/164957,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: build, module: functorch",review early; this can change release or regression risk,2025-12-11T00:53:30Z
issue,112583,Nesting no_grad in autocast causes backwards graph to be (partially) lost outside of no_grad,https://github.com/pytorch/pytorch/issues/112583,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision), module: autograd",review early; this can change release or regression risk,2025-12-09T20:12:38Z
issue,16897,Implement `numpy.random.choice` equivalent,https://github.com/pytorch/pytorch/issues/16897,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, module: numpy",review early; this can change release or regression risk,2025-11-23T10:33:22Z
issue,166004,[CUDA] torch.linalg.eigh input size reduced since PyTorch 2.8.0,https://github.com/pytorch/pytorch/issues/166004,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: linear algebra, module: regression",review early; this can change release or regression risk,2025-11-18T17:43:56Z
issue,157097,Segmentation fault in torch.repeat_interleave,https://github.com/pytorch/pytorch/issues/157097,high-priority/blocker,52,high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash,review early; this can change release or regression risk,2025-11-17T17:07:38Z
issue,146551,libtorch_cuda_linalg.so: undefined symbol: mkl_lapack_dsbrdbn on a source built PyTorch 2.6.0 with USE_STATIC_MKL=1 on CUDA platform,https://github.com/pytorch/pytorch/issues/146551,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: build, module: mkl, module: regression",review early; this can change release or regression risk,2025-11-10T23:21:29Z
issue,141358,PyTorch crashes with `mkl_lapack_dgetrf` if `from_numpy and set_num_threads` are used,https://github.com/pytorch/pytorch/issues/141358,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: intel, module: linear algebra, module: mkl",review early; this can change release or regression risk,2025-11-09T14:40:12Z
issue,165123,https://github.com/pytorch/pytorch/pull/128406 caused ISA-specific symbols leak,https://github.com/pytorch/pytorch/issues/165123,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: crash, module: regression",review early; this can change release or regression risk,2025-11-07T22:32:18Z
issue,158154,`torch.native_channel_shuffle` throws FPE when `groups` > size of the second dimension of the input tensor,https://github.com/pytorch/pytorch/issues/158154,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: error checking",review early; this can change release or regression risk,2025-11-06T18:50:50Z
issue,163675,Torch CPU wheels no longer available for download from PyTorch website,https://github.com/pytorch/pytorch/issues/163675,high-priority/blocker,52,high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: binaries,review early; this can change release or regression risk,2025-11-05T02:06:14Z
issue,124423,`__getitem__` fails to vmap for one dimensional tensors,https://github.com/pytorch/pytorch/issues/124423,high-priority/blocker,52,high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: functorch,review early; this can change release or regression risk,2025-11-03T14:53:56Z
issue,163539,conv3d bfloat16 wrong result,https://github.com/pytorch/pytorch/issues/163539,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: cuda, module: cudnn, module: regression",review early; this can change release or regression risk,2025-10-29T07:43:45Z
issue,154887,MPS batch_norm mixed dtype failure,https://github.com/pytorch/pytorch/issues/154887,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: mps, module: third_party",review early; this can change release or regression risk,2025-10-28T07:40:42Z
issue,164562,Update FlashInfer version used by vLLM tests on PyTorch CI to v0.4.1,https://github.com/pytorch/pytorch/issues/164562,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: ci, module: vllm",review early; this can change release or regression risk,2025-10-23T22:14:17Z
issue,153203,FPE Crash in torch.fmod with large negative integer modulo value,https://github.com/pytorch/pytorch/issues/153203,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: edge cases, module: int overflow, module: python frontend",review early; this can change release or regression risk,2025-10-23T06:53:49Z
issue,157461,vLLM tests failing in torch 2.8rc but passing with torch 2.7,https://github.com/pytorch/pytorch/issues/157461,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: bc-breaking, module: regression, module: vllm",review early; this can change release or regression risk,2025-10-15T20:47:15Z
issue,160296,[inductor][cpu] Performance regression in 2025-08-02 nightly release,https://github.com/pytorch/pytorch/issues/160296,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: performance, module: regression, oncall: cpu inductor",review early; this can change release or regression risk,2025-10-13T07:41:34Z
issue,58005,torch.distributed.nn.all_reduce incorrectly scales the gradient,https://github.com/pytorch/pytorch/issues/58005,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), oncall: distributed",review early; this can change release or regression risk,2025-09-27T19:02:25Z
issue,116928,"Mistral works on (2.0.1+cu117) but ""CUDA out of memory"" on (2.1.2+cu121)",https://github.com/pytorch/pytorch/issues/116928,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: memory usage, module: regression",review early; this can change release or regression risk,2025-09-26T23:33:38Z
issue,107699,index.Tensor_out & index_put.out errors or segfaults with indices list containing only null tensors,https://github.com/pytorch/pytorch/issues/107699,high-priority/blocker,52,high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: advanced indexing,review early; this can change release or regression risk,2025-09-26T23:31:14Z
issue,119261,Performance Concern about CrossEntropyLoss,https://github.com/pytorch/pytorch/issues/119261,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: nn, module: performance",review early; this can change release or regression risk,2025-09-26T23:25:54Z
issue,81180,MPS: Placeholder tensor is empty!,https://github.com/pytorch/pytorch/issues/81180,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: mps",review early; this can change release or regression risk,2025-09-26T18:51:25Z
issue,155393,crash in torch.histc,https://github.com/pytorch/pytorch/issues/155393,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: bfloat16, module: crash, module: edge cases",review early; this can change release or regression risk,2025-09-21T11:43:37Z
issue,162884,setup.py develop --cmake no longer works,https://github.com/pytorch/pytorch/issues/162884,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: build, module: regression",review early; this can change release or regression risk,2025-09-21T02:42:52Z
issue,146828,"On Linux, passing torch.Generator to multiprocessing.Process crashes for forkserver and spawn start method",https://github.com/pytorch/pytorch/issues/146828,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: multiprocessing, module: random",review early; this can change release or regression risk,2025-09-18T09:13:40Z
issue,161760,Invalid __shared__ read of size 8 bytes in torch.histc,https://github.com/pytorch/pytorch/issues/161760,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: edge cases, module: sanitizers",review early; this can change release or regression risk,2025-09-04T13:21:21Z
issue,154424,`Segmentation fault` in `torch.matmul` and `torch.sparse.addmm`,https://github.com/pytorch/pytorch/issues/154424,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: edge cases, module: empty tensor, module: error checking",review early; this can change release or regression risk,2025-09-02T06:29:00Z
issue,161874,Floating point exception in torch.nn.quantized.Conv2d.register_full_backward_hook,https://github.com/pytorch/pytorch/issues/161874,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: error checking",review early; this can change release or regression risk,2025-09-01T06:52:40Z
issue,161358,periodic tests are failing to build torchvision,https://github.com/pytorch/pytorch/issues/161358,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: ci, module: cpp-extensions",review early; this can change release or regression risk,2025-08-26T20:17:49Z
issue,161419,mobilenetv3_large_100 aot_eager accuracy failure,https://github.com/pytorch/pytorch/issues/161419,high-priority/blocker,52,high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: oncall: pt2,review early; this can change release or regression risk,2025-08-26T18:16:14Z
issue,155981,Floating point exception (core dumped) in torch.nn.functional.fold,https://github.com/pytorch/pytorch/issues/155981,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: error checking",review early; this can change release or regression risk,2025-08-20T15:30:13Z
issue,109014,Cannot install torchmetrics - ERROR 403,https://github.com/pytorch/pytorch/issues/109014,high-priority/blocker,52,high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: binaries,review early; this can change release or regression risk,2025-08-13T16:26:32Z
issue,154882,MPS max_pool2d_with_indices failure: destination values and indices length mismatch along axis,https://github.com/pytorch/pytorch/issues/154882,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: mps, module: third_party",review early; this can change release or regression risk,2025-08-12T16:49:25Z
issue,159713,[MPS Crash] Assertion failed in _getLSTMGradKernelDAGObject using LSTM on macOS with MPS backend,https://github.com/pytorch/pytorch/issues/159713,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: mps, module: rnn",review early; this can change release or regression risk,2025-08-04T10:45:24Z
issue,121711,torch.func.grad hangs on nested differentiation,https://github.com/pytorch/pytorch/issues/121711,high-priority/blocker,52,high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: functorch,review early; this can change release or regression risk,2025-07-29T14:18:38Z
issue,152824,`mypy` stage of `lintrunner -a` has intermittent but continuing crashes,https://github.com/pytorch/pytorch/issues/152824,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: flaky-tests, module: lint",review early; this can change release or regression risk,2025-07-29T14:18:13Z
issue,158728,[CI] Inductor-perf-nighly-macos started to fail to install torchvision,https://github.com/pytorch/pytorch/issues/158728,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: ci, module: regression",review early; this can change release or regression risk,2025-07-21T22:45:29Z
issue,115347,[Functorch] vmap over index_select expands the output,https://github.com/pytorch/pytorch/issues/115347,high-priority/blocker,52,high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: functorch,review early; this can change release or regression risk,2025-07-21T16:59:18Z
issue,153050,Process never ends when sending tensors through multiprocessing queues in Python 3.12+ with filesystem strategy,https://github.com/pytorch/pytorch/issues/153050,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: deadlock, module: macos, module: multiprocessing, module: regression",review early; this can change release or regression risk,2025-07-11T00:55:09Z
issue,141836,StrideAPI caused regression in channels-last logic,https://github.com/pytorch/pytorch/issues/141836,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: mps, module: regression",review early; this can change release or regression risk,2025-07-10T16:12:25Z
issue,155201,Importing xgboost before torch + openmp causes seg fault,https://github.com/pytorch/pytorch/issues/155201,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: macos, module: openmp, module: third_party",review early; this can change release or regression risk,2025-07-09T15:41:41Z
issue,152535,optree package status in PyTorch,https://github.com/pytorch/pytorch/issues/152535,high-priority/blocker,52,high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: pytree,review early; this can change release or regression risk,2025-07-09T07:43:18Z
issue,34058,libtorch.so file size is very large,https://github.com/pytorch/pytorch/issues/34058,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: cpp, oncall: mobile",review early; this can change release or regression risk,2025-07-09T02:31:02Z
issue,157098,Floating point exception in torch.nn.functional.conv_transpose3d,https://github.com/pytorch/pytorch/issues/157098,high-priority/blocker,52,high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash,review early; this can change release or regression risk,2025-07-04T01:31:49Z
issue,52663,code linking to libtorch cannot use thrust/cub functions,https://github.com/pytorch/pytorch/issues/52663,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: cuda",review early; this can change release or regression risk,2025-06-29T15:38:49Z
issue,142032,"Python 3.12 ""from functorch.einops import rearrange"" fails with ""RuntimeError: First class dim doesn't work with python 3.12""",https://github.com/pytorch/pytorch/issues/142032,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: functorch, module: regression",review early; this can change release or regression risk,2025-06-27T14:35:32Z
issue,148699,CUDA 12.6 Inductor accuracy test failures,https://github.com/pytorch/pytorch/issues/148699,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: ci, module: cuda",review early; this can change release or regression risk,2025-06-20T17:25:31Z
issue,155607,CUDA 12.6->12.8 slow and periodic failures,https://github.com/pytorch/pytorch/issues/155607,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: regression",review early; this can change release or regression risk,2025-06-17T20:26:47Z
issue,143123,MPS incompatibility: Calls into the C++ engine to run the backward pass,https://github.com/pytorch/pytorch/issues/143123,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: mps, module: regression",review early; this can change release or regression risk,2025-06-08T00:40:35Z
issue,149325,☂️ MPS support for large tensors,https://github.com/pytorch/pytorch/issues/149325,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: 64-bit, module: crash, module: mps",review early; this can change release or regression risk,2025-06-06T15:28:24Z
issue,155293,[MPS] 5D 4Mln rng attempts fail with internal error,https://github.com/pytorch/pytorch/issues/155293,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: 64-bit, module: crash, module: mps, module: random",review early; this can change release or regression risk,2025-06-06T00:20:54Z
issue,76191,"[Feature request] Exclusive prefix sum, `torch.cumsum(input, dim=0, exclusive=True)`",https://github.com/pytorch/pytorch/issues/76191,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: nestedtensor, module: reductions",review early; this can change release or regression risk,2025-06-05T03:18:33Z
issue,152292,`torch==2.6` broke `nn.Module.dtype` typing,https://github.com/pytorch/pytorch/issues/152292,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: regression, module: typing",review early; this can change release or regression risk,2025-06-05T03:13:23Z
issue,146286,Non-blocking GPU to CPU copy of complex numbers with the different conj status produces wrong results,https://github.com/pytorch/pytorch/issues/146286,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: complex, module: correctness (silent)",review early; this can change release or regression risk,2025-05-31T02:29:21Z
issue,99625,Conda Pytorch set processor affinity to the first physical core after fork,https://github.com/pytorch/pytorch/issues/99625,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: dependency bug, module: intel, module: mkl",review early; this can change release or regression risk,2025-05-30T15:14:50Z
issue,154422,`Segmentation fault` in `torch.fx.experimental.partitioner_utils.map_arg`,https://github.com/pytorch/pytorch/issues/154422,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: error checking",review early; this can change release or regression risk,2025-05-27T18:25:18Z
issue,127622,"Segfault, possibly due to recursion limit",https://github.com/pytorch/pytorch/issues/127622,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: multiprocessing",review early; this can change release or regression risk,2025-05-25T01:58:53Z
issue,152747,torch.compile causes stride mismatch in SDPA with non-contiguous query in torch 2.7,https://github.com/pytorch/pytorch/issues/152747,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: regression, oncall: pt2",review early; this can change release or regression risk,2025-05-22T15:38:08Z
issue,153961,[Windows][CPU] Root cause the VS2022 illegal instruction,https://github.com/pytorch/pytorch/issues/153961,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: crash, module: windows",review early; this can change release or regression risk,2025-05-22T08:09:15Z
issue,153733,[SOURCE] links for patch releases lead to 2.7.1,https://github.com/pytorch/pytorch/issues/153733,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: build, module: doc infra, module: docs",review early; this can change release or regression risk,2025-05-19T17:05:46Z
issue,153231,`torch.native_channel_shuffle` crashes with Floating Point Exception when given large integer parameter,https://github.com/pytorch/pytorch/issues/153231,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: edge cases, module: nn",review early; this can change release or regression risk,2025-05-14T18:33:23Z
issue,152913,Segmentation fault (core dumped) in torch.nn.functional.max_unpool2d,https://github.com/pytorch/pytorch/issues/152913,high-priority/blocker,52,high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash,review early; this can change release or regression risk,2025-05-12T20:15:02Z
issue,102904,Unable to checkpoint model and optimizer state when using Hybrid Sharding Strategy,https://github.com/pytorch/pytorch/issues/102904,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: fsdp, oncall: distributed, oncall: distributed checkpointing",review early; this can change release or regression risk,2025-05-09T00:24:54Z
issue,152979,FPE when using `torch.lcm_` with int32 tensor and int16 scalar,https://github.com/pytorch/pytorch/issues/152979,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: crash, module: edge cases, module: error checking",review early; this can change release or regression risk,2025-05-08T08:28:54Z
issue,2575,ImportError: dlopen: cannot load any more object with static TLS,https://github.com/pytorch/pytorch/issues/2575,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: assert failure, module: build, module: crash",review early; this can change release or regression risk,2025-05-07T15:56:54Z
issue,152777,Segmentation fault (core dumped) in torch.nn.functional.alpha_dropout,https://github.com/pytorch/pytorch/issues/152777,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: empty tensor, module: error checking, oncall: quantization",review early; this can change release or regression risk,2025-05-04T19:12:50Z
issue,151680,"undefined reference to `at::native::blas_impl::fp16_gemv_notrans(int, int, float, c10::Half const*, int, c10::Half const*, int, float, c10::Half*, int)'",https://github.com/pytorch/pytorch/issues/151680,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: arm, module: build, module: regression",review early; this can change release or regression risk,2025-04-18T18:10:08Z
issue,139298,CUDNN sdp attention causes loss explosion,https://github.com/pytorch/pytorch/issues/139298,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: cudnn, module: sdpa",review early; this can change release or regression risk,2025-04-17T22:03:17Z
issue,19739,Importing open3d after PyTorch causes free(): invalid pointer,https://github.com/pytorch/pytorch/issues/19739,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: pybind",review early; this can change release or regression risk,2025-04-13T20:13:49Z
issue,149201,Segment fault when used with FAISS on arm,https://github.com/pytorch/pytorch/issues/149201,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: arm, module: binaries, module: crash, module: macos",review early; this can change release or regression risk,2025-04-04T16:20:29Z
issue,150388,double free or corruption (out) in torch.as_strided_scatter,https://github.com/pytorch/pytorch/issues/150388,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: error checking",review early; this can change release or regression risk,2025-04-02T05:32:38Z
issue,147362,Floating point exception (core dumped) in torch.nn.functional.conv3d,https://github.com/pytorch/pytorch/issues/147362,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: mkldnn, module: nn",review early; this can change release or regression risk,2025-04-01T01:56:23Z
issue,149658,unexpected kwarg 'no_python_abi_suffix' when installing stable version of pytorch with sam2,https://github.com/pytorch/pytorch/issues/149658,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: build, module: cpp-extensions",review early; this can change release or regression risk,2025-03-31T20:27:40Z
issue,149820,`Segmentation fault` when using `torch.sparse.mm` with `torch.sparse_csr_tensor`,https://github.com/pytorch/pytorch/issues/149820,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: sparse",review early; this can change release or regression risk,2025-03-24T19:32:47Z
issue,128819,Environment gating of CUDA_VISIBLE_DEVICES returns a CUDA initialization error.,https://github.com/pytorch/pytorch/issues/128819,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: regression",review early; this can change release or regression risk,2025-03-23T00:14:11Z
issue,141083,ARM build failed with recent XNNPACK update: third_party/XNNPACK/src/reference/unary-elementwise.cc:125:14: error: invalid ‘static_cast’ from type ‘xnn_bfloat16’ to type ‘_Float16’,https://github.com/pytorch/pytorch/issues/141083,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: arm, module: build, module: regression, module: xnnpack",review early; this can change release or regression risk,2025-02-25T18:17:42Z
issue,146798,Torch 2.6 Unexpected Graph Break with SubConfigProxy,https://github.com/pytorch/pytorch/issues/146798,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: compile ux, module: graph breaks, module: regression, oncall: pt2",review early; this can change release or regression risk,2025-02-11T19:27:50Z
issue,146857,Segmentation Fault nn/test_convolution.py test_ConvTranspose2d_output_size_downsample_upsample on Aarch64,https://github.com/pytorch/pytorch/issues/146857,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: arm, module: crash",review early; this can change release or regression risk,2025-02-11T09:17:11Z
issue,92776,segfault when running torch.lu_unpack,https://github.com/pytorch/pytorch/issues/92776,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: edge cases, module: linear algebra",review early; this can change release or regression risk,2025-02-10T15:05:21Z
issue,146787,Floating Point Exception in `torch.ops.aten.native_channel_shuffle` with `groups=0`,https://github.com/pytorch/pytorch/issues/146787,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: empty tensor, module: error checking",review early; this can change release or regression risk,2025-02-09T20:17:30Z
issue,146789,Segmentation Fault in `torch.ops.aten.linalg_eigvals` After Invalid `unfold_copy`,https://github.com/pytorch/pytorch/issues/146789,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: linear algebra",review early; this can change release or regression risk,2025-02-09T20:17:01Z
issue,146791,Floating Point Exception in `torch.ops.aten.unfold_backward` with Specific Input,https://github.com/pytorch/pytorch/issues/146791,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: error checking",review early; this can change release or regression risk,2025-02-09T20:16:28Z
issue,146793,Segmentation Fault in `torch.ops.aten.as_strided` with Large `storage_offset`,https://github.com/pytorch/pytorch/issues/146793,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: error checking",review early; this can change release or regression risk,2025-02-09T20:14:42Z
issue,146794,Segmentation Fault in `torch.as_strided_copy` with Large `storage_offset`,https://github.com/pytorch/pytorch/issues/146794,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: error checking",review early; this can change release or regression risk,2025-02-09T20:14:31Z
issue,113663,FP8 types should not participate in type promotion and should have no math ops defined on them,https://github.com/pytorch/pytorch/issues/113663,high-priority/blocker,52,high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: floatx (formerly float8),review early; this can change release or regression risk,2025-02-08T02:16:18Z
issue,86120,TransformerEncoder src_key_padding_mask does not work in eval(),https://github.com/pytorch/pytorch/issues/86120,high-priority/blocker,52,high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: nn,review early; this can change release or regression risk,2025-02-07T12:55:25Z
issue,145259,No Range Check for `storage_offset` in `as_strided` Function,https://github.com/pytorch/pytorch/issues/145259,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: error checking",review early; this can change release or regression risk,2025-02-04T21:19:53Z
issue,48559,[FR] torch.load should support loading directly to pinned/shared memory,https://github.com/pytorch/pytorch/issues/48559,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: multiprocessing, module: serialization",review early; this can change release or regression risk,2025-02-03T18:44:54Z
issue,97109,"""Adaptive pool MPS: input sizes must be divisible by output sizes"", I keep getting this error even when I try to adjust for size",https://github.com/pytorch/pytorch/issues/97109,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: mps",review early; this can change release or regression risk,2025-02-01T11:10:55Z
issue,144563,Multiple tests not run / run as no-ops by `run_test.py`,https://github.com/pytorch/pytorch/issues/144563,high-priority/blocker,52,high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: tests,review early; this can change release or regression risk,2025-01-30T23:08:34Z
issue,134323,"Fix `torch.stft` and `torch.istft` when using `center=False`, non-rectangular windows and `win_length==hop_length`",https://github.com/pytorch/pytorch/issues/134323,high-priority/blocker,52,high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: python frontend,review early; this can change release or regression risk,2025-01-30T10:51:56Z
issue,145604,`torch.ops.aten.copy` causes SIGSEGV when handling sparse CSR tensors with invalid metadata,https://github.com/pytorch/pytorch/issues/145604,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: sparse",review early; this can change release or regression risk,2025-01-24T19:33:15Z
issue,145065,SIGFPE error when passing very large kernel_size to `avg_pool1d`,https://github.com/pytorch/pytorch/issues/145065,high-priority/blocker,52,high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash,review early; this can change release or regression risk,2025-01-21T15:40:46Z
issue,145071,Illegal memory access and segmentation fault due to large `storage_offset` in `as_strided`,https://github.com/pytorch/pytorch/issues/145071,high-priority/blocker,52,high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash,review early; this can change release or regression risk,2025-01-21T15:40:35Z
issue,145267,`torch.ops.aten.embedding_dense_backward` Crashes with Out-of-Bounds Indices On CPU,https://github.com/pytorch/pytorch/issues/145267,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: embedding, module: error checking",review early; this can change release or regression risk,2025-01-21T15:03:02Z
issue,92033,Unable to use vmap atop torch.distribution functionality,https://github.com/pytorch/pytorch/issues/92033,high-priority/blocker,52,high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: functorch,review early; this can change release or regression risk,2025-01-17T01:05:43Z
issue,143480,Segmentation fault (core dumped) in `conv1d`,https://github.com/pytorch/pytorch/issues/143480,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: edge cases, module: error checking, module: nn",review early; this can change release or regression risk,2025-01-14T15:00:04Z
issue,143481,Segmentation fault (core dumped) in `conv3d`,https://github.com/pytorch/pytorch/issues/143481,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: crash, module: error checking, module: nn",review early; this can change release or regression risk,2025-01-14T09:36:55Z
issue,104342,"Segmentation error while using F.cross_entropy with mps(for code that works fine with device= ""cpu"")",https://github.com/pytorch/pytorch/issues/104342,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: mps",review early; this can change release or regression risk,2025-01-09T19:27:40Z
issue,103966,'MPS' Issue Running HuggingFace Transformer Pix2Struct Model,https://github.com/pytorch/pytorch/issues/103966,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: mps",review early; this can change release or regression risk,2025-01-09T19:26:24Z
issue,102911,PackedSequences on MPS accelerator yields `grad_y` missing or crashes the kernel.,https://github.com/pytorch/pytorch/issues/102911,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: mps",review early; this can change release or regression risk,2025-01-09T19:23:50Z
issue,119367,Updating a slice on CPU with a slice from MPS results in an error,https://github.com/pytorch/pytorch/issues/119367,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: mps",review early; this can change release or regression risk,2025-01-09T19:04:02Z
issue,96056,MPS device throws error for `F.adaptive_avg_pool2d`,https://github.com/pytorch/pytorch/issues/96056,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: mps",review early; this can change release or regression risk,2025-01-09T18:49:22Z
issue,133625,Binary Cross Entropy crashes kernel on apple silicon with singleton dimensions,https://github.com/pytorch/pytorch/issues/133625,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: mps",review early; this can change release or regression risk,2025-01-09T18:18:29Z
issue,84489,"For PyTorch Nightly, failure when changing MPS device to CPU after PYTORCH_ENABLE_MPS_FALLBACK occurs.",https://github.com/pytorch/pytorch/issues/84489,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: mps",review early; this can change release or regression risk,2025-01-09T18:15:02Z
issue,13402,batch_norm doesn't bump version counter of running stats,https://github.com/pytorch/pytorch/issues/13402,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: nn",review early; this can change release or regression risk,2024-12-19T22:02:31Z
issue,143485,Aborted (core dumped) in `mkldnn_rnn_layer`,https://github.com/pytorch/pytorch/issues/143485,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: error checking, module: mkldnn",review early; this can change release or regression risk,2024-12-19T08:52:34Z
issue,143490,Segmentation fault (core dumped) in `replication_pad2d`,https://github.com/pytorch/pytorch/issues/143490,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: error checking",review early; this can change release or regression risk,2024-12-18T22:09:38Z
issue,143483,Segmentation fault (core dumped) in `embedding_bag.padding_idx`,https://github.com/pytorch/pytorch/issues/143483,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: embedding, module: empty tensor, module: error checking",review early; this can change release or regression risk,2024-12-18T16:02:40Z
issue,143482,Segmentation fault (core dumped) in `embedding_backward`,https://github.com/pytorch/pytorch/issues/143482,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: embedding, module: empty tensor, module: error checking",review early; this can change release or regression risk,2024-12-18T16:02:13Z
issue,142209,Segmentation fault (core dumped) in `convolution`,https://github.com/pytorch/pytorch/issues/142209,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: crash, module: edge cases, module: empty tensor",review early; this can change release or regression risk,2024-12-16T18:38:59Z
issue,35600,Increased memory usage in repetitive torch.jit.trace calls,https://github.com/pytorch/pytorch/issues/35600,high-priority/blocker,52,high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,review early; this can change release or regression risk,2024-12-16T08:30:28Z
issue,142838,Segmentation fault (core dumped) in `fractional_max_pool3d_backward`,https://github.com/pytorch/pytorch/issues/142838,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: edge cases, module: pooling",review early; this can change release or regression risk,2024-12-11T23:05:24Z
issue,142837,Segmentation fault (core dumped) in `embedding_dense_backward`,https://github.com/pytorch/pytorch/issues/142837,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: edge cases, module: embedding",review early; this can change release or regression risk,2024-12-11T23:05:07Z
issue,142221,[MPS] nansum crashes with scalar input,https://github.com/pytorch/pytorch/issues/142221,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: mps",review early; this can change release or regression risk,2024-12-07T01:43:02Z
issue,141847,Floating point exception (core dumped) in `slow_conv3d_forward`,https://github.com/pytorch/pytorch/issues/141847,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: edge cases",review early; this can change release or regression risk,2024-12-03T04:05:56Z
issue,141831,Nearest-neighbour upsampling with `interpolate` silently fails on CUDA when output size exceeds 2^31,https://github.com/pytorch/pytorch/issues/141831,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: 64-bit, module: cuda, module: nn",review early; this can change release or regression risk,2024-12-02T23:57:22Z
issue,134804,FlopsProfiler double counts sdpa flops,https://github.com/pytorch/pytorch/issues/134804,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: flop counter, module: regression, module: sdpa",review early; this can change release or regression risk,2024-12-02T18:19:12Z
issue,137347,cuDNN dBias error starting on 10/2 nightly,https://github.com/pytorch/pytorch/issues/137347,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: cudnn, module: regression, module: sdpa",review early; this can change release or regression risk,2024-12-02T18:19:04Z
issue,122617,[Regression PyTorch 2.1 & 2.2] `torch._transformer_encoder_layer_fwd` outputs on CUDA/NestedTensorCUDA backends significantly diverge from truth for torch>=2.1 (SDP math)),https://github.com/pytorch/pytorch/issues/122617,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: nestedtensor, module: nn",review early; this can change release or regression risk,2024-11-27T21:50:10Z
issue,141218,Floating point exception (core dumped) in `scaled_dot_product_attention`,https://github.com/pytorch/pytorch/issues/141218,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: edge cases, module: empty tensor, module: error checking",review early; this can change release or regression risk,2024-11-27T21:33:24Z
issue,141229,FPE in `torch._scaled_dot_product_attention_math`,https://github.com/pytorch/pytorch/issues/141229,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: edge cases, module: empty tensor, module: sdpa",review early; this can change release or regression risk,2024-11-27T20:11:51Z
issue,141219,Floating point exception (core dumped) in `_scaled_dot_product_flash_attention_for_cpu`,https://github.com/pytorch/pytorch/issues/141219,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: edge cases, module: empty tensor, module: sdpa",review early; this can change release or regression risk,2024-11-27T20:11:50Z
issue,128690,`torch._transform_bias_rescale_qkv`:FPE,https://github.com/pytorch/pytorch/issues/128690,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: edge cases, module: nn",review early; this can change release or regression risk,2024-11-27T20:02:11Z
issue,106700,C++ API `torch::nn::MultiheadAttention` Crashes by division by zero,https://github.com/pytorch/pytorch/issues/106700,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: crash, module: nn",review early; this can change release or regression risk,2024-11-27T19:52:28Z
issue,50804,Segmentation fault encountered when using nn.MultiheadAttention with v1.7.1,https://github.com/pytorch/pytorch/issues/50804,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: regression",review early; this can change release or regression risk,2024-11-27T19:48:31Z
issue,141481,Segmentation fault (core dumped) in `segment_reduce`,https://github.com/pytorch/pytorch/issues/141481,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: edge cases, module: python frontend",review early; this can change release or regression risk,2024-11-26T08:04:46Z
issue,140945,Segmentation fault (core dumped) in `replication_pad1d_backward`,https://github.com/pytorch/pytorch/issues/140945,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: edge cases, module: empty tensor",review early; this can change release or regression risk,2024-11-18T18:45:29Z
issue,140272,There is probably a destructor ordering accident going on with registerPythonModule,https://github.com/pytorch/pytorch/issues/140272,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: crash, module: python frontend",review early; this can change release or regression risk,2024-11-11T23:19:03Z
issue,139987,FPE in torch.nn.functional.pixel_shuffle,https://github.com/pytorch/pytorch/issues/139987,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: edge cases, module: nn",review early; this can change release or regression risk,2024-11-08T13:37:06Z
issue,139242,Tensor.index_reduce produces incorrect result,https://github.com/pytorch/pytorch/issues/139242,high-priority/blocker,52,high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: advanced indexing,review early; this can change release or regression risk,2024-11-04T18:21:10Z
issue,135885,☂️ 150+ MacOS tests were marked flaky recently,https://github.com/pytorch/pytorch/issues/135885,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: infra, module: macos",review early; this can change release or regression risk,2024-11-01T09:57:47Z
issue,138425,torch.div throws a floating point exception when rounding_mode is `trunc` or `floor`,https://github.com/pytorch/pytorch/issues/138425,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: edge cases, module: intel, module: python frontend",review early; this can change release or regression risk,2024-10-22T12:47:28Z
issue,17199,Deadlock with multiprocessing (using fork) and OpenMP / PyTorch should warn after OMP and fork that multithreading may be broken,https://github.com/pytorch/pytorch/issues/17199,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: docs, module: multiprocessing",review early; this can change release or regression risk,2024-10-11T07:29:28Z
issue,137098,[CI] Multiple tests fail with `Host and device pointer dont match with cudaHostRegister` and are being disabled,https://github.com/pytorch/pytorch/issues/137098,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: ci, module: cuda",review early; this can change release or regression risk,2024-10-05T06:18:10Z
issue,135637,MPS code contains references to undocumented APIs,https://github.com/pytorch/pytorch/issues/135637,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: build, module: mps",review early; this can change release or regression risk,2024-10-02T17:35:58Z
issue,136819,Aborted (core dumped) in `torch.smm`/`torch.hspmm`/`torch.hsmm`/`torch.sspaddmm`,https://github.com/pytorch/pytorch/issues/136819,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: sparse",review early; this can change release or regression risk,2024-09-30T17:27:34Z
issue,136704,Segmentation fault (core dumped) in `torch._fft_r2c`/`torch._fft_c2`/`torch._fft_c2r`,https://github.com/pytorch/pytorch/issues/136704,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: edge cases",review early; this can change release or regression risk,2024-09-27T02:16:03Z
issue,136709,Segmentation fault (core dumped) in `torch._weight_norm`/`torch._weight_int8pack_mm`,https://github.com/pytorch/pytorch/issues/136709,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: edge cases",review early; this can change release or regression risk,2024-09-27T02:15:46Z
issue,136714,Aborted (core dumped) in `torch.linalg.ldl_solve` with double free or corruption (out),https://github.com/pytorch/pytorch/issues/136714,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: linear algebra",review early; this can change release or regression risk,2024-09-27T01:59:42Z
issue,136719,Segmentation fault (core dumped) in `torch.nn.functional.max_pool1d`,https://github.com/pytorch/pytorch/issues/136719,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: edge cases, module: python frontend",review early; this can change release or regression risk,2024-09-27T01:57:45Z
issue,136374,Runbook for GCP A100 runners should be updated,https://github.com/pytorch/pytorch/issues/136374,high-priority/blocker,52,high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: infra,review early; this can change release or regression risk,2024-09-23T17:45:11Z
issue,136299,The results of baddbmm on CPU seem to have issues under certain conditions,https://github.com/pytorch/pytorch/issues/136299,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: arm, module: correctness (silent), module: regression",review early; this can change release or regression risk,2024-09-23T17:15:38Z
issue,136217,Segmentaion fault on Apple M1 caused by libomp on v2.4.1,https://github.com/pytorch/pytorch/issues/136217,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: macos, module: openmp",review early; this can change release or regression risk,2024-09-19T16:53:56Z
issue,122394,[MPS] Severe Performance Degradation with Deterministic Algorithm on macOS 14.4,https://github.com/pytorch/pytorch/issues/122394,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: mps, module: performance, module: regression",review early; this can change release or regression risk,2024-09-10T21:52:16Z
issue,82627,PyTorch crashes when running with OpenACC,https://github.com/pytorch/pytorch/issues/82627,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: openmp, module: third_party",review early; this can change release or regression risk,2024-09-08T13:46:39Z
issue,133334,torch._scaled_mm row-wise hits CUDA invalid memory access when M % 256 != 0,https://github.com/pytorch/pytorch/issues/133334,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: floatx (formerly float8)",review early; this can change release or regression risk,2024-08-22T19:44:12Z
issue,80431,CMake Error: File /opt/pytorch/build_variables.bzl does not exist.,https://github.com/pytorch/pytorch/issues/80431,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: docker, module: regression",review early; this can change release or regression risk,2024-08-15T11:43:10Z
issue,132086,Result of MPS memory operation can be corrupted if memory pressure is high,https://github.com/pytorch/pytorch/issues/132086,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: memory usage, module: mps",review early; this can change release or regression risk,2024-08-05T18:27:55Z
issue,70108,Derivative for _ctc_loss_backward,https://github.com/pytorch/pytorch/issues/70108,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: double backwards",review early; this can change release or regression risk,2024-07-24T16:56:00Z
issue,83818,torch.linalg.eigh crashe for matrices of size 2895×2895 or larger on eigen and M1,https://github.com/pytorch/pytorch/issues/83818,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: linear algebra, module: m1",review early; this can change release or regression risk,2024-07-12T20:37:35Z
issue,78829,Certain import order triggers segmentation fault,https://github.com/pytorch/pytorch/issues/78829,high-priority/blocker,52,high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash,review early; this can change release or regression risk,2024-07-10T10:20:41Z
issue,128503,crash@sleef_tryVXE2 () while trying to run torch.compile() BERT model,https://github.com/pytorch/pytorch/issues/128503,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: power, module: sleef",review early; this can change release or regression risk,2024-07-04T07:25:20Z
issue,125247,RuntimeError: could not create a primitive descriptor for a matmul primitive,https://github.com/pytorch/pytorch/issues/125247,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: crash",review early; this can change release or regression risk,2024-07-02T07:27:40Z
issue,39007,[docs] Urls changed => forum links would become invalid,https://github.com/pytorch/pytorch/issues/39007,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: doc infra, module: docs",review early; this can change release or regression risk,2024-06-24T15:23:51Z
issue,128694,Segmentation fault (core dumped) in `torch._fused_moving_avg_obs_fq_helper`,https://github.com/pytorch/pytorch/issues/128694,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: edge cases",review early; this can change release or regression risk,2024-06-15T03:11:38Z
issue,128695,Segmentation fault (core dumped) in `torch._weight_norm_interface`,https://github.com/pytorch/pytorch/issues/128695,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: edge cases, module: empty tensor",review early; this can change release or regression risk,2024-06-14T14:10:57Z
issue,128692,Segmentation fault (core dumped) in `torch._weight_int4pack_mm`,https://github.com/pytorch/pytorch/issues/128692,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: edge cases, module: error checking",review early; this can change release or regression risk,2024-06-14T13:38:42Z
issue,126385,Segmentation Faults loading Models pytorch v2.3.0 Apple M2,https://github.com/pytorch/pytorch/issues/126385,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: macos, module: openmp",review early; this can change release or regression risk,2024-06-10T20:43:46Z
issue,113037,"`TensorIteratorBase::is_scalar` return `false` for empty numpy tensors, but true for empty Torch ones",https://github.com/pytorch/pytorch/issues/113037,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: crash, module: regression",review early; this can change release or regression risk,2024-06-05T23:01:53Z
issue,127102,segfault when registering op with mismatched schema,https://github.com/pytorch/pytorch/issues/127102,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: custom-operators, module: opcheck",review early; this can change release or regression risk,2024-05-29T00:02:17Z
issue,66782,"Segmentation Fault when importing Torch, 2021 version",https://github.com/pytorch/pytorch/issues/66782,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: crash",review early; this can change release or regression risk,2024-05-16T17:39:05Z
issue,64794,linalg.lstsq out variant fails internal assert because it uses non-inplace view op for some inputs,https://github.com/pytorch/pytorch/issues/64794,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: ci, module: linear algebra",review early; this can change release or regression risk,2024-05-14T00:41:00Z
issue,120995,AttributeError: module 'mpmath' has no attribute 'rational',https://github.com/pytorch/pytorch/issues/120995,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: binaries, oncall: releng",review early; this can change release or regression risk,2024-05-09T18:43:09Z
issue,82068,upsample_bilinear2d() received an invalid combination of arguments,https://github.com/pytorch/pytorch/issues/82068,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: nn, module: regression",review early; this can change release or regression risk,2024-05-06T17:12:16Z
issue,120622,_native_batch_norm_legit doesn't bump version counter of buffers,https://github.com/pytorch/pytorch/issues/120622,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: codegen, module: correctness (silent), module: norms and normalization",review early; this can change release or regression risk,2024-05-06T15:10:03Z
issue,124644,torch.Tensor.remainder raises a floating point exception when divisor is -1,https://github.com/pytorch/pytorch/issues/124644,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: edge cases",review early; this can change release or regression risk,2024-04-29T17:18:50Z
issue,105943,PT2 is not thread safe,https://github.com/pytorch/pytorch/issues/105943,high-priority/blocker,52,high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: oncall: pt2,review early; this can change release or regression risk,2024-04-09T18:37:31Z
issue,123219,[CI] vec_test_all tests are not run in CI,https://github.com/pytorch/pytorch/issues/123219,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: ci, module: cpu, module: intel, module: regression",review early; this can change release or regression risk,2024-04-03T14:30:02Z
issue,2576,CUDA multinomial is limited to 2^24 categories,https://github.com/pytorch/pytorch/issues/2576,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: 64-bit, module: cuda, module: distributions",review early; this can change release or regression risk,2024-04-01T14:46:25Z
issue,121150,☂️ oneDNN update functional/regression,https://github.com/pytorch/pytorch/issues/121150,high-priority/blocker,52,high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: mkldnn,review early; this can change release or regression risk,2024-03-15T11:35:49Z
issue,101529,Inconsistent performance degradation of 3x3 convolution (torch 2.0.1+cu118),https://github.com/pytorch/pytorch/issues/101529,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: cudnn, module: performance, module: regression",review early; this can change release or regression risk,2024-03-02T02:06:59Z
issue,119547,CUDA Memory error while having a lot of free memory,https://github.com/pytorch/pytorch/issues/119547,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: memory usage, module: regression",review early; this can change release or regression risk,2024-02-28T12:02:21Z
issue,120597,floating point exception on `torch.fmod` when x is large negative integer and y is `-1`,https://github.com/pytorch/pytorch/issues/120597,high-priority/blocker,52,high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash,review early; this can change release or regression risk,2024-02-26T17:54:09Z
issue,47056,Training slows down and memory usage increases when upgrading from PyTorch 1.6 to 1.7,https://github.com/pytorch/pytorch/issues/47056,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision), module: performance, module: regression",review early; this can change release or regression risk,2024-02-22T21:50:34Z
issue,53625,torch.arange() issue,https://github.com/pytorch/pytorch/issues/53625,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: multiprocessing, module: regression, oncall: distributed",review early; this can change release or regression risk,2024-02-10T14:53:21Z
issue,119000,Performance regression in torch.nn.Linear,https://github.com/pytorch/pytorch/issues/119000,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: nn, module: performance, module: regression",review early; this can change release or regression risk,2024-02-05T16:48:23Z
issue,117431,Program killed by SIGFPE in the backward pass of torch.sparse.sampled_addmm when input has specific shape,https://github.com/pytorch/pytorch/issues/117431,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: cuda, module: sparse",review early; this can change release or regression risk,2024-01-16T15:11:56Z
issue,78924,"If large enough tensor is being cloned, parallel dataloading hangs on M1 Mac",https://github.com/pytorch/pytorch/issues/78924,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: arm, module: dataloader, module: deadlock, module: macos",review early; this can change release or regression risk,2024-01-14T18:50:07Z
issue,117254,```at::mse_loss```: Internal assertion failed,https://github.com/pytorch/pytorch/issues/117254,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: loss",review early; this can change release or regression risk,2024-01-11T19:57:09Z
issue,115416,torch.nn.functional.embedding segmentation fault on large negative ids,https://github.com/pytorch/pytorch/issues/115416,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: edge cases, module: int overflow",review early; this can change release or regression risk,2023-12-08T23:19:37Z
issue,112043,"AssertionError: graph-captured input #1, of type <class 'torch.Tensor'>, is not among original inputs of types…",https://github.com/pytorch/pytorch/issues/112043,high-priority/blocker,52,high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: oncall: export,review early; this can change release or regression risk,2023-11-28T17:08:45Z
issue,114529,Bug in element-wise multiplication of `torch.sparse_csr_tensor`s on GPU - 0's in result considered significant - PyTorch 2.1.1,https://github.com/pytorch/pytorch/issues/114529,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: regression, module: sparse",review early; this can change release or regression risk,2023-11-28T14:40:00Z
issue,50037,Torch quantized_lstm_cell op out-of-bounds access,https://github.com/pytorch/pytorch/issues/50037,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, oncall: quantization",review early; this can change release or regression risk,2023-11-17T22:54:35Z
issue,106432,inconsistent dtype of scale and zero_point in observers,https://github.com/pytorch/pytorch/issues/106432,high-priority/blocker,52,high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,review early; this can change release or regression risk,2023-11-17T22:42:45Z
issue,88882,Can't import torch --> OSError related to libcublasLt.so.11,https://github.com/pytorch/pytorch/issues/88882,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: cuda, module: regression",review early; this can change release or regression risk,2023-11-02T13:58:24Z
issue,99719,New feature requested: vmap for torch.histc,https://github.com/pytorch/pytorch/issues/99719,high-priority/blocker,52,high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: functorch,review early; this can change release or regression risk,2023-10-12T18:20:37Z
issue,110205,RuntimeError: Expected packed scalar Tensor to be of dimension 1. Got 0 instead.,https://github.com/pytorch/pytorch/issues/110205,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: optimizer, module: regression",review early; this can change release or regression risk,2023-10-04T21:39:47Z
issue,107668,torch.dot gives wrong result on Macos,https://github.com/pytorch/pytorch/issues/107668,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: macos",review early; this can change release or regression risk,2023-10-01T20:58:21Z
issue,94544,Abort Caused by Virtual Function,https://github.com/pytorch/pytorch/issues/94544,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: build, module: regression",review early; this can change release or regression risk,2023-09-08T02:03:21Z
issue,108532,"Breaking incompatibility with Cuda 12.2, pytorch stable, torchvision",https://github.com/pytorch/pytorch/issues/108532,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: crash, module: cuda",review early; this can change release or regression risk,2023-09-07T17:37:27Z
issue,107363,Regression in text encoding,https://github.com/pytorch/pytorch/issues/107363,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: multithreading, module: performance, module: regression",review early; this can change release or regression risk,2023-08-23T03:26:14Z
issue,107429,Abort when running torch.set_num_interop_threads,https://github.com/pytorch/pytorch/issues/107429,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: single threaded",review early; this can change release or regression risk,2023-08-20T00:54:18Z
issue,39388,"RuntimeError: NCCL error in: /pytorch/torch/lib/c10d/ProcessGroupNCCL.cpp:514, unhandled system error, NCCL version 2.4.8",https://github.com/pytorch/pytorch/issues/39388,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: nccl",review early; this can change release or regression risk,2023-08-18T18:28:22Z
issue,106362,Calling ops.aten.embedding_bag() function got silent crash,https://github.com/pytorch/pytorch/issues/106362,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: nn",review early; this can change release or regression risk,2023-08-14T17:02:56Z
issue,50013,torch.Tensor.repeat is divergent from np.repeat,https://github.com/pytorch/pytorch/issues/50013,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: deprecation, module: numpy",review early; this can change release or regression risk,2023-07-27T16:39:10Z
issue,103756,libtorch > 1.9.1 produces segfault on Qt5 gui application exit,https://github.com/pytorch/pytorch/issues/103756,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: crash",review early; this can change release or regression risk,2023-07-25T08:26:45Z
issue,42793,distributed training with c10d does not work with layerdrop in pytorch > 1.4,https://github.com/pytorch/pytorch/issues/42793,high-priority/blocker,52,high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,review early; this can change release or regression risk,2023-07-06T11:00:37Z
issue,103117,Observed regress in DataLoader spawn from PyTorch1.13 to PyTorch2.0,https://github.com/pytorch/pytorch/issues/103117,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: data, module: performance, module: regression",review early; this can change release or regression risk,2023-06-13T18:00:24Z
issue,99509,'Illegal instruction (core dumped)' for gpt-j bf16 generation task using greedy search,https://github.com/pytorch/pytorch/issues/99509,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: crash, module: intel",review early; this can change release or regression risk,2023-06-07T16:06:24Z
issue,99982,torch.cuda.is_available() crashes python in systems with disabled gpu,https://github.com/pytorch/pytorch/issues/99982,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: cuda, module: edge cases",review early; this can change release or regression risk,2023-06-03T15:16:36Z
issue,98537,Segfault when using torch.ops.* to access de-registered op,https://github.com/pytorch/pytorch/issues/98537,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: dispatch, module: library",review early; this can change release or regression risk,2023-04-28T14:50:07Z
issue,100085,[regression] torch.norm with out dtype bfloat16 cause runtime error,https://github.com/pytorch/pytorch/issues/100085,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: norms and normalization, module: regression",review early; this can change release or regression risk,2023-04-27T17:17:31Z
issue,98136,A Segment Fault can be triggered in torch._grid_sampler_2d_cpu_fallback,https://github.com/pytorch/pytorch/issues/98136,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: nn",review early; this can change release or regression risk,2023-04-14T15:10:18Z
issue,98503,Power VSX vectorization support disabled,https://github.com/pytorch/pytorch/issues/98503,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: build, module: power, module: regression",review early; this can change release or regression risk,2023-04-07T17:07:37Z
issue,55375,`DataParallel` (`broadcast_coalesced`) with complex tensors yield real views,https://github.com/pytorch/pytorch/issues/55375,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: complex, module: data parallel, oncall: distributed",review early; this can change release or regression risk,2023-04-04T13:59:09Z
issue,95432,Regression bug in `torch.nn.ReLU6` and `torch.nn.Hardtanh` that `inplace=True` doesn't work in PyTorch 1.10.0~1.13.1,https://github.com/pytorch/pytorch/issues/95432,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: memory usage, module: nn",review early; this can change release or regression risk,2023-04-03T18:32:30Z
issue,97086,A crash due to Floating Point Exception can be triggered in torch.index_select,https://github.com/pytorch/pytorch/issues/97086,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: edge cases, module: error checking",review early; this can change release or regression risk,2023-03-27T17:12:04Z
issue,96276,A Segment Fault can be triggered in torch.geqrf with an edge case,https://github.com/pytorch/pytorch/issues/96276,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: edge cases",review early; this can change release or regression risk,2023-03-09T12:04:23Z
issue,96275,A Segment Fault can be triggered in torch.pinverse,https://github.com/pytorch/pytorch/issues/96275,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: edge cases",review early; this can change release or regression risk,2023-03-09T12:03:55Z
issue,96277,A Segment Fault can be triggered in torch.adaptive_max_pool1d with an edge case,https://github.com/pytorch/pytorch/issues/96277,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: edge cases",review early; this can change release or regression risk,2023-03-09T12:03:45Z
issue,91754,CUDA error: initialization error,https://github.com/pytorch/pytorch/issues/91754,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: nvfuser, module: regression",review early; this can change release or regression risk,2023-03-03T04:07:14Z
issue,54418,Some Numba tests are failing,https://github.com/pytorch/pytorch/issues/54418,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: ci, module: numba, module: tests",review early; this can change release or regression risk,2023-02-26T10:16:29Z
issue,94808,Memory Corruption in torch.lstm caused by edge cases,https://github.com/pytorch/pytorch/issues/94808,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: edge cases, module: rnn",review early; this can change release or regression risk,2023-02-15T16:22:38Z
issue,94698,A segment fault can be triggered in torch.avg_pool1d,https://github.com/pytorch/pytorch/issues/94698,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: edge cases",review early; this can change release or regression risk,2023-02-15T16:22:09Z
issue,94668,A segment fault can be triggered in torch.lstm with edge cases,https://github.com/pytorch/pytorch/issues/94668,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: edge cases",review early; this can change release or regression risk,2023-02-15T16:21:43Z
issue,94669,A segment fault can be triggered in torch.svd,https://github.com/pytorch/pytorch/issues/94669,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: edge cases",review early; this can change release or regression risk,2023-02-15T16:21:26Z
issue,94696,A segment fault can be triggered in torch.max_pool1d_with_indices,https://github.com/pytorch/pytorch/issues/94696,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: edge cases",review early; this can change release or regression risk,2023-02-15T16:19:49Z
issue,94594,A segment fault can be triggered in torch.histogramdd,https://github.com/pytorch/pytorch/issues/94594,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: edge cases",review early; this can change release or regression risk,2023-02-10T17:12:51Z
issue,93846,large number of temporary files generated when using dataloader with num_workers>0,https://github.com/pytorch/pytorch/issues/93846,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: dataloader, module: openmp",review early; this can change release or regression risk,2023-02-10T04:29:56Z
issue,68349,"SIGILL, Illegal Instruction from libtorch_cpu.so when callling backward() function in a VM.",https://github.com/pytorch/pytorch/issues/68349,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: crash, module: dependency bug, module: mkldnn",review early; this can change release or regression risk,2023-02-08T07:39:55Z
issue,94125,A Floating Point Exception can be trigerred in torch._C._nn.slow_conv3d,https://github.com/pytorch/pytorch/issues/94125,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: edge cases",review early; this can change release or regression risk,2023-02-07T18:51:10Z
issue,91590,Crashes of linalg.ldl_solve on different edge cases not coming from linalg.ldl_factor,https://github.com/pytorch/pytorch/issues/91590,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: edge cases, module: linear algebra",review early; this can change release or regression risk,2023-01-23T16:08:58Z
issue,82894,linalg and lu tests fail when run in parallel on linux cuda,https://github.com/pytorch/pytorch/issues/82894,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: ci, module: cuda, module: linear algebra",review early; this can change release or regression risk,2023-01-16T10:34:45Z
issue,58414,cuDNN v8 API tracking issue,https://github.com/pytorch/pytorch/issues/58414,high-priority/blocker,52,high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: cudnn,review early; this can change release or regression risk,2023-01-11T00:17:59Z
issue,90535,Illegal hardware instruction using torch.nn.Conv2d on aarch64 (Raspberry Pi 4),https://github.com/pytorch/pytorch/issues/90535,high-priority/blocker,52,high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash,review early; this can change release or regression risk,2023-01-09T10:43:02Z
issue,41661,libgcc_s.so.1 must be installed for pthread_cancel to work,https://github.com/pytorch/pytorch/issues/41661,high-priority/blocker,52,high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: binaries,review early; this can change release or regression risk,2023-01-05T12:22:29Z
issue,83243,RPC: wait method of Future object return 0 sometimes in rpc framework,https://github.com/pytorch/pytorch/issues/83243,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: rpc, oncall: distributed",review early; this can change release or regression risk,2022-12-27T03:34:00Z
issue,90676,RuntimeError: Error in dlopen: libnvJitLink.so.12: cannot open shared object file: No such file or directory,https://github.com/pytorch/pytorch/issues/90676,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: distributions",review early; this can change release or regression risk,2022-12-12T15:07:55Z
issue,88842,test/test_ops.py is segfaulting on master build with DEBUG assets,https://github.com/pytorch/pytorch/issues/88842,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: ci, module: testing",review early; this can change release or regression risk,2022-12-06T20:11:14Z
issue,89718,`torch.Tensor.flatten` Trigger Segmentation Fault when trying to provide and output named dim,https://github.com/pytorch/pytorch/issues/89718,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: named tensor",review early; this can change release or regression risk,2022-11-30T10:25:17Z
issue,82072,cant build with USE_VULKAN=1,https://github.com/pytorch/pytorch/issues/82072,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: build, module: vulkan, oncall: mobile",review early; this can change release or regression risk,2022-11-23T22:26:42Z
issue,38122,Pybind11 cpp extensions broken with pytorch v1.5.0,https://github.com/pytorch/pytorch/issues/38122,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: build, module: cpp, module: docs, module: regression",review early; this can change release or regression risk,2022-11-22T15:05:59Z
issue,83393,Test public bindings in CI gives weird output on error,https://github.com/pytorch/pytorch/issues/83393,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: ci, module: python frontend, module: tests",review early; this can change release or regression risk,2022-11-15T04:22:12Z
issue,89006,Add smoke-tests for CPP extensions compilations,https://github.com/pytorch/pytorch/issues/89006,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: cpp-extensions, module: regression",review early; this can change release or regression risk,2022-11-14T19:19:05Z
issue,73179,Floating point exception in _nnpack_spatial_convolution,https://github.com/pytorch/pytorch/issues/73179,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: cpu, module: error checking, module: nnpack",review early; this can change release or regression risk,2022-11-07T22:55:03Z
issue,73174,`LayerNorm` triggers INTERNAL ASSERT,https://github.com/pytorch/pytorch/issues/73174,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: error checking, module: norms and normalization",review early; this can change release or regression risk,2022-11-07T08:22:24Z
issue,88036,A segment fault can be triggered in fbgemm_pack_gemm_matrix_fp16,https://github.com/pytorch/pytorch/issues/88036,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: half, module: third_party",review early; this can change release or regression risk,2022-10-31T19:12:43Z
issue,87283,Missing string parsing for some parameter types in python arg parsing logic,https://github.com/pytorch/pytorch/issues/87283,high-priority/blocker,52,high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: pybind,review early; this can change release or regression risk,2022-10-24T17:10:43Z
issue,86704,Performance tests mnist_hogwild-cpu_memory CPU memory increase by 30%,https://github.com/pytorch/pytorch/issues/86704,high-priority/blocker,52,high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: regression,review early; this can change release or regression risk,2022-10-17T17:06:45Z
issue,85538,Iterative Global Pruning Cause GPU Memory Leak,https://github.com/pytorch/pytorch/issues/85538,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: memory usage, module: nn",review early; this can change release or regression risk,2022-10-08T04:28:05Z
issue,85909,Steam Deck Core Dump,https://github.com/pytorch/pytorch/issues/85909,high-priority/blocker,52,high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: rocm,review early; this can change release or regression risk,2022-10-03T17:27:31Z
issue,80104,Provide error message when thread pool is exhausted in RPC,https://github.com/pytorch/pytorch/issues/80104,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: rpc, oncall: distributed",review early; this can change release or regression risk,2022-09-26T16:07:00Z
issue,85329,Crash in `torch.package.PackageExporter`,https://github.com/pytorch/pytorch/issues/85329,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: edge cases, oncall: package/deploy",review early; this can change release or regression risk,2022-09-21T19:49:07Z
issue,85215,Floating point exception in gather gradient computation.,https://github.com/pytorch/pytorch/issues/85215,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: edge cases, module: error checking",review early; this can change release or regression risk,2022-09-21T19:43:42Z
issue,85217,Segmentation fault in native_batch_norm,https://github.com/pytorch/pytorch/issues/85217,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: error checking, module: norms and normalization",review early; this can change release or regression risk,2022-09-21T19:43:33Z
issue,85214,Segmentation fault in mkldnn_reorder_conv2d_weight and mkldnn_reorder_conv3d_weight,https://github.com/pytorch/pytorch/issues/85214,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: edge cases, module: mkldnn",review early; this can change release or regression risk,2022-09-19T17:25:25Z
issue,84990,Segmentation fault in `torch.futures.collect_all`,https://github.com/pytorch/pytorch/issues/84990,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: error checking",review early; this can change release or regression risk,2022-09-16T17:24:38Z
issue,84181,Session of Google Colab crashes when `torch.utils::SummaryWriter` is called after importing `torchaudio`,https://github.com/pytorch/pytorch/issues/84181,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: tensorboard",review early; this can change release or regression risk,2022-09-12T17:11:56Z
issue,80488,Negative values still produced by torch.nn.functional.kl_div,https://github.com/pytorch/pytorch/issues/80488,high-priority/blocker,52,high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: nn,review early; this can change release or regression risk,2022-08-31T08:31:10Z
issue,84051,model.load_state_dict won't work in Child process if a sufficiently large tensor was padded in the Parent (even if empty padded),https://github.com/pytorch/pytorch/issues/84051,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: memory usage, module: multiprocessing",review early; this can change release or regression risk,2022-08-30T10:06:09Z
issue,83070,make_fx + aot_autograd segfaults,https://github.com/pytorch/pytorch/issues/83070,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: functorch, module: fx",review early; this can change release or regression risk,2022-08-22T14:53:58Z
issue,80946,torch.nn.functional.linear fails for multi-dimensional bias from torch 1.12,https://github.com/pytorch/pytorch/issues/80946,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: nn, module: regression",review early; this can change release or regression risk,2022-08-18T15:03:34Z
issue,83577,libtorch malloc cause coredump,https://github.com/pytorch/pytorch/issues/83577,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: crash, module: static linking",review early; this can change release or regression risk,2022-08-17T14:17:50Z
issue,82813,functionalize and make_fx are not composable resulting in segfault and cuda error,https://github.com/pytorch/pytorch/issues/82813,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: functorch, module: fx",review early; this can change release or regression risk,2022-08-10T16:12:21Z
issue,81195,torch._weight_norm with specified dim returns wrong output,https://github.com/pytorch/pytorch/issues/81195,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: error checking, module: nn, module: norms and normalization, module: regression",review early; this can change release or regression risk,2022-07-27T16:46:02Z
issue,77284,non-rentrant checkpointing uses same memory as non-checkpointed code,https://github.com/pytorch/pytorch/issues/77284,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: activation checkpointing, oncall: distributed",review early; this can change release or regression risk,2022-07-25T03:31:50Z
issue,81127,Docker updates cause subsequent builds to fail,https://github.com/pytorch/pytorch/issues/81127,high-priority/blocker,52,high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: ci,review early; this can change release or regression risk,2022-07-12T16:47:25Z
issue,80337,Illegal Memory Access from nonzero method when Tensor is Too Large,https://github.com/pytorch/pytorch/issues/80337,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: dependency bug, module: edge cases",review early; this can change release or regression risk,2022-06-27T21:45:57Z
issue,80012,static builds are broken by MKL_DNN,https://github.com/pytorch/pytorch/issues/80012,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: build, module: regression, module: third_party",review early; this can change release or regression risk,2022-06-22T17:45:34Z
issue,69801,"Dropout2d doesn't drop channels for (C, H, W)",https://github.com/pytorch/pytorch/issues/69801,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: nn",review early; this can change release or regression risk,2022-06-15T20:43:38Z
issue,13918,Pytorch very slow to convert list of numpy arrays into tensors,https://github.com/pytorch/pytorch/issues/13918,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: numpy, module: performance",review early; this can change release or regression risk,2022-06-15T09:00:31Z
issue,79014,RuntimeError: cuDNN error: CUDNN_STATUS_INTERNAL_ERROR,https://github.com/pytorch/pytorch/issues/79014,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: cuda, module: cudnn, module: ddp",review early; this can change release or regression risk,2022-06-13T18:41:56Z
issue,59745,[c10d] Work objects should have a general operator<<,https://github.com/pytorch/pytorch/issues/59745,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: c10d, oncall: distributed",review early; this can change release or regression risk,2022-06-03T21:10:44Z
issue,73186,Segmentation fault in fractional_max_pool3d,https://github.com/pytorch/pytorch/issues/73186,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: error checking, module: pooling",review early; this can change release or regression risk,2022-05-07T18:08:23Z
issue,73185,Segmentation fault in fractional_max_pool2d,https://github.com/pytorch/pytorch/issues/73185,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: error checking, module: pooling",review early; this can change release or regression risk,2022-05-07T18:06:58Z
issue,47953,Linear algebra GPU backend tracking issue [magma/cusolver/cublas],https://github.com/pytorch/pytorch/issues/47953,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: linear algebra, module: performance",review early; this can change release or regression risk,2022-05-05T13:28:14Z
issue,61223,optimize_for_inference does not handle MKLDNN for BatchNorm1d,https://github.com/pytorch/pytorch/issues/61223,high-priority/blocker,52,high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,review early; this can change release or regression risk,2022-05-01T09:42:28Z
issue,65815,Overflow error in torch.linalg uncaught,https://github.com/pytorch/pytorch/issues/65815,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: linear algebra, module: multithreading, module: numerical-stability",review early; this can change release or regression risk,2022-04-14T19:49:19Z
issue,73488,After XNNPack update `TestXNNPACKSerDes.test_linear` started to fail,https://github.com/pytorch/pytorch/issues/73488,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: regression, module: rocm, module: xnnpack",review early; this can change release or regression risk,2022-03-31T17:16:55Z
issue,74547,importing open3d before pytorch causes matmul to produce a segfault,https://github.com/pytorch/pytorch/issues/74547,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: cublas, module: cuda, module: regression",review early; this can change release or regression risk,2022-03-23T20:33:42Z
issue,38228,pybind11::gil_scoped_release: crash on exit with daemon threads,https://github.com/pytorch/pytorch/issues/38228,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: dependency bug",review early; this can change release or regression risk,2022-03-22T17:47:20Z
issue,74422,Embedding Pytorch in C++ using pybind fails on interpreter shutdown,https://github.com/pytorch/pytorch/issues/74422,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: pybind, module: third_party",review early; this can change release or regression risk,2022-03-22T00:56:08Z
issue,73792,Bug: torch.distributions.mixture_same_distribution._pad_mixture_dimension,https://github.com/pytorch/pytorch/issues/73792,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: distributions",review early; this can change release or regression risk,2022-03-07T19:01:49Z
issue,73349,Large performance difference of loss.backward() between torch-1.9.0 and torch-1.8.0,https://github.com/pytorch/pytorch/issues/73349,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: regression",review early; this can change release or regression risk,2022-02-25T08:57:52Z
issue,35151,Segfault when using misaligned data pointer (from joblib),https://github.com/pytorch/pytorch/issues/35151,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: numpy",review early; this can change release or regression risk,2022-02-24T22:37:54Z
issue,70350,Core dumped with large matmul on aarch64,https://github.com/pytorch/pytorch/issues/70350,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: arm, module: crash",review early; this can change release or regression risk,2022-02-24T13:06:18Z
issue,15849,DataLoader with option to re-use worker processes,https://github.com/pytorch/pytorch/issues/15849,high-priority/blocker,52,high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: dataloader,review early; this can change release or regression risk,2022-02-22T13:42:40Z
issue,71227,Move torch::deploy tests to their own workflow job,https://github.com/pytorch/pytorch/issues/71227,high-priority/blocker,52,high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: deploy,review early; this can change release or regression risk,2022-02-08T19:46:28Z
issue,72182,test_fn_fwgrad_bwgrad_special_ndtr_cuda_float64 fails,https://github.com/pytorch/pytorch/issues/72182,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: cuda",review early; this can change release or regression risk,2022-02-02T20:00:50Z
issue,72189,test_fn_fwgrad_bwgrad_[trapezoid|trapz]_cuda_complex128 causes CUDA memory exception,https://github.com/pytorch/pytorch/issues/72189,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: ci, module: crash, module: cuda",review early; this can change release or regression risk,2022-02-02T19:59:44Z
issue,29973,Indexing into tensor order of magnitude slower than numpy,https://github.com/pytorch/pytorch/issues/29973,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: advanced indexing, module: autograd, module: performance",review early; this can change release or regression risk,2022-02-01T15:49:01Z
issue,71391,Calling .backward() inside of an LBFGS closure function throws an exception in Libtorch v1.6.0+,https://github.com/pytorch/pytorch/issues/71391,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: regression",review early; this can change release or regression risk,2022-01-25T19:36:01Z
issue,42452,`device-side assert triggered` - when probability > 1.0,https://github.com/pytorch/pytorch/issues/42452,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: cuda",review early; this can change release or regression risk,2022-01-14T02:41:10Z
issue,42487,Support recursive data type in TorchScript,https://github.com/pytorch/pytorch/issues/42487,high-priority/blocker,52,high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,review early; this can change release or regression risk,2021-12-30T06:42:53Z
issue,69512,Inconsistent behavior of cosine_similarity between fp16 and fp32 inputs,https://github.com/pytorch/pytorch/issues/69512,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: distance functions, module: half, module: numerical-stability",review early; this can change release or regression risk,2021-12-13T18:21:56Z
issue,67448,Deadlock in test_multiprocessing_spawn.py,https://github.com/pytorch/pytorch/issues/67448,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: deadlock, module: multiprocessing",review early; this can change release or regression risk,2021-11-08T18:25:55Z
issue,67797,Conv2d kernel performance regression on CPU since PyTorch 1.9,https://github.com/pytorch/pytorch/issues/67797,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: mkldnn, module: performance, module: regression",review early; this can change release or regression risk,2021-11-05T19:56:50Z
issue,67247,pytorch_linux_xenial_py3_6_gcc5_4_test may timeout during test_multiprocessing_spawn,https://github.com/pytorch/pytorch/issues/67247,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: multiprocessing, module: tests",review early; this can change release or regression risk,2021-11-02T17:46:10Z
issue,60081,[Bug] numpy is no longer a required dependency,https://github.com/pytorch/pytorch/issues/60081,high-priority/blocker,52,high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: numpy,review early; this can change release or regression risk,2021-11-01T10:47:14Z
issue,59237,Multiplication of `torch.tensor` with `np.array` does the operation with numpy.,https://github.com/pytorch/pytorch/issues/59237,high-priority/blocker,52,high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: numpy,review early; this can change release or regression risk,2021-10-28T13:31:30Z
issue,65366,Exception thrown in final autograd callback (queue_callback) not caught if not on CPU thread: terminate called after throwing an instance of 'python_error',https://github.com/pytorch/pytorch/issues/65366,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: bootcamp, module: crash",review early; this can change release or regression risk,2021-10-11T17:25:21Z
issue,46155,Complex backward returns NaN values,https://github.com/pytorch/pytorch/issues/46155,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: complex, module: cuda",review early; this can change release or regression risk,2021-10-06T15:42:35Z
issue,38910,Memory Leak with Docker GPU,https://github.com/pytorch/pytorch/issues/38910,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: cudnn, module: memory usage, module: regression",review early; this can change release or regression risk,2021-10-05T19:20:41Z
issue,52633,CUDA `linalg.norm` matrix order +/-2 and nuclear norm for extreme values may be incorrect,https://github.com/pytorch/pytorch/issues/52633,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: linear algebra, module: nans and infs",review early; this can change release or regression risk,2021-09-14T16:20:13Z
issue,57178,MaxPool2D Returns Wrong Shape With Ceil_Mode,https://github.com/pytorch/pytorch/issues/57178,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: docs, module: pooling",review early; this can change release or regression risk,2021-09-13T16:08:34Z
issue,63486,Distributed / MPI / CUDA: Incorrect messages received from isend/irecv in PyTorch 1.9,https://github.com/pytorch/pytorch/issues/63486,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), oncall: distributed",review early; this can change release or regression risk,2021-09-02T15:14:16Z
issue,50344,NumPy Compatibility tracking issue,https://github.com/pytorch/pytorch/issues/50344,high-priority/blocker,52,high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: numpy,review early; this can change release or regression risk,2021-08-31T07:10:26Z
issue,63749,Segmentation fault when using C++/pybind11 module without also importing torch,https://github.com/pytorch/pytorch/issues/63749,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: cpp-extensions, module: pybind, module: regression",review early; this can change release or regression risk,2021-08-23T15:50:27Z
issue,50012,torch.split is divergent from np.split,https://github.com/pytorch/pytorch/issues/50012,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: deprecation, module: numpy",review early; this can change release or regression risk,2021-08-13T18:36:42Z
issue,59196,torch.zeros memory leak : cpu,https://github.com/pytorch/pytorch/issues/59196,high-priority/blocker,52,high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: memory usage,review early; this can change release or regression risk,2021-08-04T11:56:56Z
issue,48558,[FR] tensor ctors should support directly creating in shared memory,https://github.com/pytorch/pytorch/issues/48558,high-priority/blocker,52,high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: tensor creation,review early; this can change release or regression risk,2021-07-07T13:25:52Z
issue,60626,Impossible to raise the limit on number of shared memory tensors,https://github.com/pytorch/pytorch/issues/60626,high-priority/blocker,52,high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: multiprocessing,review early; this can change release or regression risk,2021-07-05T23:38:19Z
issue,58370,Add better python operators annotations for IDE type checking,https://github.com/pytorch/pytorch/issues/58370,high-priority/blocker,52,high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: typing,review early; this can change release or regression risk,2021-06-30T08:17:25Z
issue,60628,pip install nightly torch torchaudio and torchvision together will install 0.3.0 version of torchvsion on Ubuntu,https://github.com/pytorch/pytorch/issues/60628,high-priority/blocker,52,high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: binaries,review early; this can change release or regression risk,2021-06-28T20:48:47Z
issue,44634,torch.sparse improvements - tracking issue,https://github.com/pytorch/pytorch/issues/44634,high-priority/blocker,52,high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: sparse,review early; this can change release or regression risk,2021-06-28T11:12:20Z
issue,24823,"Problematic handling of NaN and inf in grid_sample, causing segfaults, corrupted CUDA memory, and incorrect results",https://github.com/pytorch/pytorch/issues/24823,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: cuda, module: interpolation",review early; this can change release or regression risk,2021-06-21T15:13:23Z
issue,37449,RuntimeError: CUDA error: an illegal memory access was encountered with channels_last,https://github.com/pytorch/pytorch/issues/37449,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: cuda, module: cudnn, module: dependency bug",review early; this can change release or regression risk,2021-06-14T14:11:26Z
issue,59645,Tensor.type() does not work with meta tensors,https://github.com/pytorch/pytorch/issues/59645,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: meta tensors",review early; this can change release or regression risk,2021-06-08T19:53:52Z
issue,6998,[pytorch] Not handling python reload properly,https://github.com/pytorch/pytorch/issues/6998,high-priority/blocker,52,high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash,review early; this can change release or regression risk,2021-06-06T22:11:41Z
issue,48491,SegmentationFault when pytorch is installed from source.,https://github.com/pytorch/pytorch/issues/48491,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: build, module: crash",review early; this can change release or regression risk,2021-05-28T12:02:28Z
issue,48382,hangs indefinitely at os.waitpid(),https://github.com/pytorch/pytorch/issues/48382,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: deadlock, module: docs, module: error checking, module: multiprocessing",review early; this can change release or regression risk,2021-05-14T02:34:34Z
issue,40137,Improve RPC test debugability,https://github.com/pytorch/pytorch/issues/40137,high-priority/blocker,52,high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: rpc,review early; this can change release or regression risk,2021-04-26T17:17:43Z
issue,49520,Segfault in torch.bincount,https://github.com/pytorch/pytorch/issues/49520,high-priority/blocker,52,high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash,review early; this can change release or regression risk,2021-04-21T16:09:11Z
issue,51134,segmentation fault in torch.nn.ReplicationPad3d/2d when padding is large,https://github.com/pytorch/pytorch/issues/51134,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: nn, module: padding",review early; this can change release or regression risk,2021-04-14T22:04:11Z
issue,55669,Segmentation fault when loss.backward(),https://github.com/pytorch/pytorch/issues/55669,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: cuda",review early; this can change release or regression risk,2021-04-12T17:24:23Z
issue,45136,Multiple torch.load in one file,https://github.com/pytorch/pytorch/issues/45136,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: regression, module: serialization",review early; this can change release or regression risk,2021-04-03T09:22:01Z
issue,50879,Unexpected slow dropout in stacked RNN/LSTM/GRU,https://github.com/pytorch/pytorch/issues/50879,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: cudnn, module: rnn",review early; this can change release or regression risk,2021-03-26T01:13:58Z
issue,20497,Statically make `__setstate__` set all attributes/parameters,https://github.com/pytorch/pytorch/issues/20497,high-priority/blocker,52,high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: serialization,review early; this can change release or regression risk,2021-03-15T17:48:44Z
issue,53796,foreach ops in autograd,https://github.com/pytorch/pytorch/issues/53796,high-priority/blocker,52,high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: autograd,review early; this can change release or regression risk,2021-03-15T17:21:29Z
issue,49877,The same tensor requires more memory on RTX3090,https://github.com/pytorch/pytorch/issues/49877,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: memory usage",review early; this can change release or regression risk,2021-03-09T21:03:38Z
issue,18300,NCCL backend fails when calling broadcast from different threads,https://github.com/pytorch/pytorch/issues/18300,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: nccl, oncall: distributed",review early; this can change release or regression risk,2021-03-03T08:13:12Z
issue,46491,cannot call rpc.init_rpc twice within a single process,https://github.com/pytorch/pytorch/issues/46491,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: rpc, oncall: distributed",review early; this can change release or regression risk,2021-02-23T02:07:43Z
issue,47844,Incorrect output loss value under specific CUDA version,https://github.com/pytorch/pytorch/issues/47844,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: dependency bug, module: linear algebra, module: numerical-stability",review early; this can change release or regression risk,2021-02-11T23:12:17Z
issue,51863,TestVectorizedMemoryAccess.CopyKernel starting to fail after driver to 460.39,https://github.com/pytorch/pytorch/issues/51863,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: ci, module: cuda",review early; this can change release or regression risk,2021-02-10T20:11:19Z
issue,51332,Cant compile from source 1.8.0a0 because 'magma_v2.h' file not found and undefined references to `magma_*',https://github.com/pytorch/pytorch/issues/51332,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: build, module: cuda, module: linear algebra, module: magma",review early; this can change release or regression risk,2021-02-08T18:15:40Z
issue,51732,torch.nn.functional.ctc_loss crash(segfault),https://github.com/pytorch/pytorch/issues/51732,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: loss, module: nn",review early; this can change release or regression risk,2021-02-05T19:57:13Z
issue,51083,apex internal assert failed,https://github.com/pytorch/pytorch/issues/51083,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: assert failure, module: dependency bug, module: regression",review early; this can change release or regression risk,2021-02-01T18:05:18Z
issue,49758,c10::scalar_to_tensor(...) uses should be audited for performance and type promotion impact,https://github.com/pytorch/pytorch/issues/49758,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: performance",review early; this can change release or regression risk,2021-01-22T18:01:54Z
issue,50038,Torch _remove_batch_dim OP out-of-bounds access,https://github.com/pytorch/pytorch/issues/50038,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: vmap",review early; this can change release or regression risk,2021-01-06T02:25:15Z
issue,42815,Add a guide for what to do when you think there's a memory leak,https://github.com/pytorch/pytorch/issues/42815,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: docs, module: memory usage",review early; this can change release or regression risk,2020-12-28T18:07:05Z
issue,49459,This error occurs occasionally during the run,https://github.com/pytorch/pytorch/issues/49459,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: cudnn",review early; this can change release or regression risk,2020-12-28T17:46:44Z
issue,49392,Uninitialized variable was not detected in ASAN CI config,https://github.com/pytorch/pytorch/issues/49392,high-priority/blocker,52,high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: ci,review early; this can change release or regression risk,2020-12-21T18:17:04Z
issue,49473,Pytorch 1.5+: Kernel dies / Segmentation Fault with torch.cuda.mermory_allocated() and torch.cuda.mermory_reserved(),https://github.com/pytorch/pytorch/issues/49473,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: regression",review early; this can change release or regression risk,2020-12-18T02:59:31Z
issue,47116,trying to Installing pytorch on python3.9 via pip results in an non-descript error,https://github.com/pytorch/pytorch/issues/47116,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: binaries, oncall: releng",review early; this can change release or regression risk,2020-12-11T01:07:44Z
issue,48400,conv_transpose3d returns different result when the input and kernel are mkldnn tensors,https://github.com/pytorch/pytorch/issues/48400,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: dependency bug, module: mkldnn",review early; this can change release or regression risk,2020-12-10T01:27:13Z
issue,38231,Python autograd engine threads never terminate in Python 3.5-3.8,https://github.com/pytorch/pytorch/issues/38231,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: ci, module: crash",review early; this can change release or regression risk,2020-11-16T21:04:05Z
issue,47161,segfault during shutdown with torch1.7,https://github.com/pytorch/pytorch/issues/47161,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: dependency bug, module: pybind, module: third_party",review early; this can change release or regression risk,2020-11-09T18:17:35Z
issue,39061,Some @slowtests are never run in CI,https://github.com/pytorch/pytorch/issues/39061,high-priority/blocker,52,high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: tests,review early; this can change release or regression risk,2020-11-06T19:12:56Z
issue,38623,Providing CUDA tensor to model on CPU causes a crash,https://github.com/pytorch/pytorch/issues/38623,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: cuda, module: rnn",review early; this can change release or regression risk,2020-11-06T19:07:46Z
issue,31558,More dynamic PyTorch APIs,https://github.com/pytorch/pytorch/issues/31558,high-priority/blocker,52,high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: ux,review early; this can change release or regression risk,2020-10-10T08:45:13Z
issue,43005,`torch.distributions.Categorical` crashes with illegal instruction,https://github.com/pytorch/pytorch/issues/43005,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: distributions",review early; this can change release or regression risk,2020-10-08T07:43:38Z
issue,42784,libtorch 1.5 crashes when used on macs when using torch::max without AVX support,https://github.com/pytorch/pytorch/issues/42784,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: crash, module: macos",review early; this can change release or regression risk,2020-10-07T03:49:08Z
issue,40457,DataParallel with Torch 1.5,https://github.com/pytorch/pytorch/issues/40457,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: data parallel, module: regression",review early; this can change release or regression risk,2020-09-30T08:20:37Z
issue,43484,pybind11_object_dealloc error,https://github.com/pytorch/pytorch/issues/43484,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: pybind",review early; this can change release or regression risk,2020-09-04T09:23:28Z
issue,35418,PyTorch / libtorch executables fail when built against libcuda stub library,https://github.com/pytorch/pytorch/issues/35418,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: build",review early; this can change release or regression risk,2020-07-29T13:06:37Z
issue,41226,"The name ""Contiguous"" in `torch::MemoryFormat::Contiguous` can mislead users into assuming contiguous memory",https://github.com/pytorch/pytorch/issues/41226,high-priority/blocker,52,high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: memory format,review early; this can change release or regression risk,2020-07-13T17:09:35Z
issue,5353,Bugs: Score Function approach in REINFORCE for PONG,https://github.com/pytorch/pytorch/issues/5353,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: cuda, module: loss, module: memory usage",review early; this can change release or regression risk,2020-04-20T21:26:45Z
issue,36040,[JIT] Huge delay (1274s vs 0.031s) when running scripted model,https://github.com/pytorch/pytorch/issues/36040,high-priority/blocker,52,high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,review early; this can change release or regression risk,2020-04-17T20:40:23Z
issue,36152,Seg Fault: import vaex with torch,https://github.com/pytorch/pytorch/issues/36152,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: crash",review early; this can change release or regression risk,2020-04-13T17:23:57Z
issue,35597,libtorch_global_deps.so not found.,https://github.com/pytorch/pytorch/issues/35597,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: build, module: regression",review early; this can change release or regression risk,2020-03-30T22:24:47Z
issue,32918,Interpolate in the “bicubic” mode with the same shape outputs zeros from second sample onwards,https://github.com/pytorch/pytorch/issues/32918,high-priority/blocker,52,high-priority/blocker/regression label present; old update age or low-urgency/stale label,review early; this can change release or regression risk,2020-03-28T20:52:04Z
issue,28980,Unsafe use of `at::parallel_for` in current codebase,https://github.com/pytorch/pytorch/issues/28980,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: internals, module: multithreading, module: performance",review early; this can change release or regression risk,2019-11-04T18:11:58Z
issue,25518,Assign torch.cuda.FloatTensor to List tensor,https://github.com/pytorch/pytorch/issues/25518,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: cuda",review early; this can change release or regression risk,2019-11-02T07:28:42Z
issue,16953,Seg fault with test_rnn_retain_variables on ppc64le,https://github.com/pytorch/pytorch/issues/16953,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: power",review early; this can change release or regression risk,2019-06-17T20:11:52Z
issue,20748,Segmentation fault when use torch::from_blob,https://github.com/pytorch/pytorch/issues/20748,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: abi, module: cpp, module: crash",review early; this can change release or regression risk,2019-05-22T18:57:11Z
pr,189329,[DEBUG] Profile the XPU SYCL device-link bottleneck,https://github.com/pytorch/pytorch/pull/189329,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/xpu,"identify whether block is CI, merge conflict, or review gate",2026-07-09T02:15:47Z
pr,188963,[XPU] Enable XPU testing for 50 test classes across 40 test files,https://github.com/pytorch/pytorch/pull/188963,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/b200, ciflow/h100, ciflow/inductor, ciflow/rocm-mi300","identify whether block is CI, merge conflict, or review gate",2026-07-09T02:13:43Z
pr,189229,[DO NOT MERGE][ROCm] Use torch.version.hip to detect ROCm when /opt/rocm is absent,https://github.com/pytorch/pytorch/pull/189229,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/docker, ciflow/dtensor, ciflow/inductor, ciflow/rocm-nightly","identify whether block is CI, merge conflict, or review gate",2026-07-09T01:53:50Z
pr,189275,Consolidate CD wheel build pipelines under .ci/wheel/ and unify the numpy pin,https://github.com/pytorch/pytorch/pull/189275,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/binaries_wheel,"identify whether block is CI, merge conflict, or review gate",2026-07-09T01:53:40Z
pr,189320,[inductor] Enable nested Inductor configs for invoke_subgraph regions,https://github.com/pytorch/pytorch/pull/189320,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-09T01:53:07Z
pr,189324,[T278829272][Claude Code] Gate TLX flex_attention choices on tlx_mode,https://github.com/pytorch/pytorch/pull/189324,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-09T01:53:05Z
pr,189321,[inductor] Enable per-region cudagraphs for invoke_subgraph via auto graph partition,https://github.com/pytorch/pytorch/pull/189321,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-09T01:53:04Z
pr,178393,[DO NOT MERGE] Dynamo 3.15 support,https://github.com/pytorch/pytorch/pull/178393,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/dynamo, ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-09T01:05:02Z
pr,189299,[MemoryViz] Decimate over-long stacked-area polygons so the summarize…,https://github.com/pytorch/pytorch/pull/189299,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-07-09T01:04:20Z
pr,179094,[DO NOT MERGE] Add docker image for python 3.15,https://github.com/pytorch/pytorch/pull/179094,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/docker,"identify whether block is CI, merge conflict, or review gate",2026-07-09T00:21:49Z
pr,189301,[ROCm] Bump AOTriton to 0.13b,https://github.com/pytorch/pytorch/pull/189301,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/rocm-mi300,"identify whether block is CI, merge conflict, or review gate",2026-07-08T23:49:17Z
pr,189092,[dynamo] reduce call_obj_hasattr overrides,https://github.com/pytorch/pytorch/pull/189092,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T23:23:53Z
pr,187707,[dynamo] remove GetAttrVariable,https://github.com/pytorch/pytorch/pull/187707,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T23:23:45Z
pr,189091,[dynamo] guard getattr() builtin constant-fold fallback,https://github.com/pytorch/pytorch/pull/189091,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T23:23:44Z
pr,187531,[dynamo] enable generic attribute mutation on non-UDOV VTs,https://github.com/pytorch/pytorch/pull/187531,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T23:23:35Z
pr,187532,[dynamo] fix __getattr__ and __getattribute__ dunder semantics,https://github.com/pytorch/pytorch/pull/187532,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T23:23:32Z
pr,188469,flex_gemm: support compressed local reduce aux outputs,https://github.com/pytorch/pytorch/pull/188469,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T23:12:49Z
pr,188435,[dynamo] Model uninitialized closure cells in __build_class__,https://github.com/pytorch/pytorch/pull/188435,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T18:44:50Z
pr,189216,[Dynamo] Implement DELETE_DEREF,https://github.com/pytorch/pytorch/pull/189216,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T17:02:56Z
pr,189165,[aot_compile] Add torch.compiler.precompile,https://github.com/pytorch/pytorch/pull/189165,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T16:58:43Z
pr,189237,Add c-shims for aten.unsqueeze and aten.expand,https://github.com/pytorch/pytorch/pull/189237,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2026-07-08T14:58:22Z
pr,188016,Update torch-xpu-ops commit pin,https://github.com/pytorch/pytorch/pull/188016,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/binaries_wheel, ciflow/xpu","identify whether block is CI, merge conflict, or review gate",2026-07-08T13:52:31Z
pr,189198,[DO NOT MERGE][ROCm] Validate 7.14 skip set on ROCm preview lane with TheRock ROCm deps preloading,https://github.com/pytorch/pytorch/pull/189198,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/docker, ciflow/dtensor, ciflow/inductor, ciflow/rocm-nightly","identify whether block is CI, merge conflict, or review gate",2026-07-08T13:01:49Z
pr,189218,[DO NOT MERGE][ROCm] Validate 7.14 skip set on TheRock preview lane,https://github.com/pytorch/pytorch/pull/189218,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/docker, ciflow/dtensor, ciflow/inductor, ciflow/rocm-nightly","identify whether block is CI, merge conflict, or review gate",2026-07-08T13:01:47Z
pr,188720,[ROCm] Enable cholesky_ex via hipSOLVER xgeev (ROCm >= 7.14),https://github.com/pytorch/pytorch/pull/188720,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/rocm-mi200, ciflow/rocm-mi300, ciflow/rocm-nightly","identify whether block is CI, merge conflict, or review gate",2026-07-08T13:01:38Z
pr,189250,[Test][Refactor] Decouple test_dataloader and test_comparison_utils according to the accelerator relevance.,https://github.com/pytorch/pytorch/pull/189250,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-07-08T13:01:10Z
pr,188236,[inductor]: Include convolution_backward in decide_layout_opt conv node detection,https://github.com/pytorch/pytorch/pull/188236,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk, ciflow/xpu","identify whether block is CI, merge conflict, or review gate",2026-07-08T10:30:31Z
pr,189240,[TEST] Refactor test_c10d_logger.py with hw_classification,https://github.com/pytorch/pytorch/pull/189240,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-07-08T10:13:21Z
pr,188539,Add support for win_arm64 CUDA builds,https://github.com/pytorch/pytorch/pull/188539,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-07-08T10:06:30Z
pr,189227,[xpu][feature]Extend SYCL device-specific math functions to NumericUtils,https://github.com/pytorch/pytorch/pull/189227,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-07-08T08:40:33Z
pr,189203,[DO NOT MERGE] This is a reference branch to test 7.14,https://github.com/pytorch/pytorch/pull/189203,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/docker, ciflow/inductor-perf-test-nightly-rocm-mi300, ciflow/inductor-perf-test-nightly-rocm-mi350","identify whether block is CI, merge conflict, or review gate",2026-07-08T06:51:35Z
pr,189084,[DO NOT MERGE] Testing Rocm 7.14,https://github.com/pytorch/pytorch/pull/189084,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/docker, ciflow/inductor-perf-test-nightly-rocm-mi300, ciflow/inductor-perf-test-nightly-rocm-mi350, ciflow/inductor-rocm-mi200","identify whether block is CI, merge conflict, or review gate",2026-07-08T05:57:52Z
pr,189205,[DO NOT MERGE] Harmless comment just to enable CI,https://github.com/pytorch/pytorch/pull/189205,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor-perf-test-nightly-rocm-mi300, ciflow/inductor-perf-test-nightly-rocm-mi350","identify whether block is CI, merge conflict, or review gate",2026-07-08T04:51:10Z
pr,180186,Try to enable LTO in OSS,https://github.com/pytorch/pytorch/pull/180186,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/binaries,"identify whether block is CI, merge conflict, or review gate",2026-07-08T03:41:17Z
pr,189209,[inductor] Fix with_effects star deps keyed by buffer name instead of operation name,https://github.com/pytorch/pytorch/pull/189209,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T02:23:04Z
pr,182334,[inductor][autotuning] pipeline AsyncCompile + CachingAutotuner (#182334),https://github.com/pytorch/pytorch/pull/182334,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T01:01:49Z
pr,163503,[dynamo] prevent recompilation limit exceeded on external_utils.wrap_inline,https://github.com/pytorch/pytorch/pull/163503,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T22:27:47Z
pr,188173,"[dynamo, nested graph breaks] propagate NGB suppression through intermediate inlined functions",https://github.com/pytorch/pytorch/pull/188173,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/dynamo, ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-07T22:00:17Z
pr,171826,"[dynamo] add break_graph_if_unsupported to COMPARE_OP, BINARY_*, and INPLACE_* instructions",https://github.com/pytorch/pytorch/pull/171826,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T20:52:40Z
pr,187650,[Pytorch Test] Add skip_if_lt_x_devices and migrate FSDP autograd tests to be device-agnostic,https://github.com/pytorch/pytorch/pull/187650,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-07-07T12:35:34Z
pr,181728,[xpu][4/4]Enable MXFP8/MXFP4/NVFP4 tests for XPU,https://github.com/pytorch/pytorch/pull/181728,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/b200, ciflow/h100, ciflow/inductor, ciflow/rocm-mi300","identify whether block is CI, merge conflict, or review gate",2026-07-07T10:42:46Z
pr,183613,[CD] Tighten ROCm manywheel build pipeline,https://github.com/pytorch/pytorch/pull/183613,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/binaries,"identify whether block is CI, merge conflict, or review gate",2026-07-07T08:22:23Z
pr,189128,Gil acquire bool conversion,https://github.com/pytorch/pytorch/pull/189128,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-07-07T08:21:34Z
pr,189070,Accept device-qualified gloo backend in monitored_barrier,https://github.com/pytorch/pytorch/pull/189070,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/torchtitan,"identify whether block is CI, merge conflict, or review gate",2026-07-07T04:25:35Z
pr,189073,Make subgroup name hash salt collective-consistent,https://github.com/pytorch/pytorch/pull/189073,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/torchtitan,"identify whether block is CI, merge conflict, or review gate",2026-07-07T04:25:34Z
pr,189074,Deduplicate comms when destroying a process group,https://github.com/pytorch/pytorch/pull/189074,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/torchtitan,"identify whether block is CI, merge conflict, or review gate",2026-07-07T04:25:33Z
pr,189072,Create TorchComms subgroups on the correct device,https://github.com/pytorch/pytorch/pull/189072,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/torchtitan,"identify whether block is CI, merge conflict, or review gate",2026-07-07T04:25:33Z
pr,189071,Remove new_group to split_group delegation on the TorchComms path,https://github.com/pytorch/pytorch/pull/189071,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/torchtitan,"identify whether block is CI, merge conflict, or review gate",2026-07-07T04:25:30Z
pr,189069,Route only TorchComms-owned backends through comms,https://github.com/pytorch/pytorch/pull/189069,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/torchtitan,"identify whether block is CI, merge conflict, or review gate",2026-07-07T01:03:17Z
pr,189068,Register nccl-lazy TorchComms backend,https://github.com/pytorch/pytorch/pull/189068,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/torchtitan,"identify whether block is CI, merge conflict, or review gate",2026-07-07T00:07:18Z
pr,182864,[donotrvw][broken][WIP] spmd_types + PP: _SpmdTypesMeta and nullcontext decorator fix,https://github.com/pytorch/pytorch/pull/182864,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/torchtitan,"identify whether block is CI, merge conflict, or review gate",2026-07-06T23:51:34Z
pr,185208,[PoC] uv.lock dependency-groups for runtime + build deps,https://github.com/pytorch/pytorch/pull/185208,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/binaries_wheel, ciflow/docker","identify whether block is CI, merge conflict, or review gate",2026-07-06T20:14:39Z
pr,182844,[CI] Pin miniforge to 26.1.1-3 to fix docker builds,https://github.com/pytorch/pytorch/pull/182844,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-07-06T19:08:18Z
pr,181215,Add cooperative GPU-sharing infrastructure via CUDA process-checkpoint,https://github.com/pytorch/pytorch/pull/181215,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/torchtitan,"identify whether block is CI, merge conflict, or review gate",2026-07-06T19:08:17Z
pr,188119,[inductor] Enable combo kernel compile time autotuning by default,https://github.com/pytorch/pytorch/pull/188119,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-06T12:16:07Z
pr,188326,feat(dynamo): backport recursive dict tags,https://github.com/pytorch/pytorch/pull/188326,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-06T11:07:14Z
pr,188203,[inductor] Fix combo-kernel peak-memory KeyError when reorder_for_peak_memory is disabled,https://github.com/pytorch/pytorch/pull/188203,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-06T09:15:11Z
pr,189003,[inductor] Coordinate-descent tune compile-time combo kernel blocks,https://github.com/pytorch/pytorch/pull/189003,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-06T09:15:09Z
pr,189002,[inductor] Don't combo-fuse data-independent producers feeding attention,https://github.com/pytorch/pytorch/pull/189002,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-06T09:15:09Z
pr,189001,[inductor] Fix flaky combo-kernel peak-memory test,https://github.com/pytorch/pytorch/pull/189001,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2026-07-06T09:15:08Z
pr,188886,[AOTI] Support PDL launch attributes via `cuLaunchKernelEx` in CUDA wrappers,https://github.com/pytorch/pytorch/pull/188886,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-06T07:51:18Z
pr,181324,"[dynamo] Remove TypingVariable, trace more of typing.py",https://github.com/pytorch/pytorch/pull/181324,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/dynamo, ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-06T06:01:50Z
pr,181611,[don't merge] test combo kernels,https://github.com/pytorch/pytorch/pull/181611,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-06T02:20:20Z
pr,174503,[Draft]Trying to see if cpp modification would work,https://github.com/pytorch/pytorch/pull/174503,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2026-07-05T20:49:00Z
pr,181205,"NOT MERGE, JUST FOR TEST",https://github.com/pytorch/pytorch/pull/181205,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/xpu,"identify whether block is CI, merge conflict, or review gate",2026-07-05T07:38:56Z
pr,182143,[OSDC] Migrate periodic.yml to OSDC via dial-up pattern,https://github.com/pytorch/pytorch/pull/182143,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/periodic,"identify whether block is CI, merge conflict, or review gate",2026-07-05T02:19:58Z
pr,182314,[OSDC] Migrate dynamo-unittest.yml to OSDC via dial-up pattern,https://github.com/pytorch/pytorch/pull/182314,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/dynamo,"identify whether block is CI, merge conflict, or review gate",2026-07-05T01:09:32Z
pr,180875,[CUDA][cuBLASLt] torch.addmm - improve numerics of bias fusion with reduced precision SplitK algorithms,https://github.com/pytorch/pytorch/pull/180875,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/b200, ciflow/h100, ciflow/rocm-mi300, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-04T11:55:40Z
pr,182415,[c10d] Apply record_comm names to NCCL launch annotations,https://github.com/pytorch/pytorch/pull/182415,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/torchtitan,"identify whether block is CI, merge conflict, or review gate",2026-07-04T09:18:40Z
pr,182389,Collective buffer detection for DDP and FSDP,https://github.com/pytorch/pytorch/pull/182389,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-07-04T02:20:07Z
pr,182388,Training hooks and public API for memory component tracking,https://github.com/pytorch/pytorch/pull/182388,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-07-04T02:20:06Z
pr,182387,Allocator-level memory attribution primitives,https://github.com/pytorch/pytorch/pull/182387,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-07-04T02:20:05Z
pr,181412,Add Claude-powered operator benchmark triage,https://github.com/pytorch/pytorch/pull/181412,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-07-03T17:59:34Z
pr,182226,[CI] Install git-core PPA without add-apt-repository,https://github.com/pytorch/pytorch/pull/182226,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-07-03T16:04:25Z
pr,188887,Enable cpp_wrapper for mixed order reduction,https://github.com/pytorch/pytorch/pull/188887,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-03T14:13:29Z
pr,188858,[DO NOT MERGE] Complex Wrapper Test,https://github.com/pytorch/pytorch/pull/188858,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-07-03T05:47:09Z
pr,188734,Add tracer kwarg to torch.compile for make_fx tracing,https://github.com/pytorch/pytorch/pull/188734,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2026-07-03T04:22:06Z
pr,186945,Enable nested Inductor configs for invoke_subgraph regions,https://github.com/pytorch/pytorch/pull/186945,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-02T22:50:21Z
pr,182810,[ROCm] Implement TORCH_HIP_VERSION,https://github.com/pytorch/pytorch/pull/182810,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/rocm-mi300, ciflow/rocm-mi355, ciflow/rocm-nightly, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-02T19:47:46Z
pr,188787,OneDNN 3.12.2 acceptance test,https://github.com/pytorch/pytorch/pull/188787,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/binaries_wheel,"identify whether block is CI, merge conflict, or review gate",2026-07-02T15:31:04Z
pr,182691,Parameterized cuda graph launch 3,https://github.com/pytorch/pytorch/pull/182691,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-01T22:03:04Z
pr,187909,[DONT MERGE] Test new xpu runners,https://github.com/pytorch/pytorch/pull/187909,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/trunk, ciflow/xpu","identify whether block is CI, merge conflict, or review gate",2026-07-01T21:29:39Z
pr,188136,[BE] Build all XPU wheels on a single runner,https://github.com/pytorch/pytorch/pull/188136,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/binaries_wheel,"identify whether block is CI, merge conflict, or review gate",2026-07-01T18:10:57Z
pr,188662,[ROCm] Updated mi300 runners to have 8 shards,https://github.com/pytorch/pytorch/pull/188662,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/inductor-rocm-mi300, ciflow/rocm-mi300, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-01T17:27:58Z
pr,188284,[Draft] Fix BUILD_SEPARATE_OPS cyclic dependency — register with xpu_hal.dll,https://github.com/pytorch/pytorch/pull/188284,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/trunk, ciflow/xpu","identify whether block is CI, merge conflict, or review gate",2026-07-01T17:27:47Z
pr,188629,docs: document S3-compatible object storage for distributed checkpoint,https://github.com/pytorch/pytorch/pull/188629,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-07-01T14:03:19Z
pr,188577,Migrate Build Triton Wheel to OSDC,https://github.com/pytorch/pytorch/pull/188577,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-07-01T06:46:07Z
pr,188259,Add XPU caching allocator checkpoint pool state bindings,https://github.com/pytorch/pytorch/pull/188259,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-01T01:30:26Z
pr,188592,vllm wheel 0630,https://github.com/pytorch/pytorch/pull/188592,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-06-30T23:51:15Z
pr,187206,[WIP][no-ci] auto_custom_op,https://github.com/pytorch/pytorch/pull/187206,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-06-30T21:46:47Z
pr,187878,Add c10d Watchdog: a libuv-backed timeout service,https://github.com/pytorch/pytorch/pull/187878,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/torchtitan,"identify whether block is CI, merge conflict, or review gate",2026-06-30T19:51:50Z
pr,188429,[ROCm][CI] Switch rocm-preview to TheRock wheels,https://github.com/pytorch/pytorch/pull/188429,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/docker, ciflow/rocm-nightly, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-30T17:53:53Z
pr,188463,[PU1] Add output_size parameter to torch.bincount for eager-meta support.,https://github.com/pytorch/pytorch/pull/188463,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/mps,"identify whether block is CI, merge conflict, or review gate",2026-06-30T00:05:47Z
pr,187870,compile-on-one-rank: device-agnostic inductor codegen and launcher,https://github.com/pytorch/pytorch/pull/187870,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/dtensor, ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-29T22:37:25Z
pr,188464,[spmd_types] infer R@FSDP axes from active mesh for empty params,https://github.com/pytorch/pytorch/pull/188464,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-29T21:44:37Z
pr,188448,Add DTensor sharding strategy for searchsorted,https://github.com/pytorch/pytorch/pull/188448,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/dtensor, ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-29T21:01:41Z
pr,188404,Inductor: Add metadata overlap dependency implementation,https://github.com/pytorch/pytorch/pull/188404,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-29T14:44:45Z
pr,181678,[spmd_types] CI CPU job w/o spmd_types,https://github.com/pytorch/pytorch/pull/181678,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-06-29T10:04:20Z
pr,181956,[do-not-review] cudagraph while node support,https://github.com/pytorch/pytorch/pull/181956,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-06-29T06:07:05Z
pr,187900,[XPU] Migrate CI Docker to Intel OMIX for oneAPI + GPU driver installation,https://github.com/pytorch/pytorch/pull/187900,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/docker, ciflow/xpu","identify whether block is CI, merge conflict, or review gate",2026-06-29T05:29:36Z
pr,181444,Scaled cublas grouped gemm sm90,https://github.com/pytorch/pytorch/pull/181444,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/b200, ciflow/dtensor, ciflow/h100, ciflow/inductor","identify whether block is CI, merge conflict, or review gate",2026-06-28T23:52:40Z
pr,181443,Cublas grouped gemm nvfp4,https://github.com/pytorch/pytorch/pull/181443,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/b200, ciflow/dtensor, ciflow/h100, ciflow/inductor","identify whether block is CI, merge conflict, or review gate",2026-06-28T23:52:39Z
pr,181627,[ ROCm] [ DRAFT ] 181288 + 177219,https://github.com/pytorch/pytorch/pull/181627,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/op-benchmark,"identify whether block is CI, merge conflict, or review gate",2026-06-27T18:54:03Z
pr,181628,[ ROCm] [ DRAFT ] 181288 + 177219 + 181615,https://github.com/pytorch/pytorch/pull/181628,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/op-benchmark,"identify whether block is CI, merge conflict, or review gate",2026-06-27T17:50:58Z
pr,177612,[inductor] Add pattern match for fused gate+up GEMM + SiLU on XPU,https://github.com/pytorch/pytorch/pull/177612,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/inductor-pallas, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-27T12:55:21Z
pr,181615,[ROCm] CI: skip operator matmul microbenchmark,https://github.com/pytorch/pytorch/pull/181615,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/op-benchmark,"identify whether block is CI, merge conflict, or review gate",2026-06-27T05:48:57Z
pr,181664,[FSDP2] Design notes: options for RS/AR back-edge in HSDP,https://github.com/pytorch/pytorch/pull/181664,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-06-27T02:21:55Z
pr,186271,[CUDA] Add TunableOp support for scaled GEMM,https://github.com/pytorch/pytorch/pull/186271,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/b200, ciflow/h100, ciflow/rocm-mi300","identify whether block is CI, merge conflict, or review gate",2026-06-27T01:21:00Z
pr,182271,[ci][WIP] Cut redundant before-test fanout + trim over-eager periodic crons,https://github.com/pytorch/pytorch/pull/182271,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/mps,"identify whether block is CI, merge conflict, or review gate",2026-06-27T01:01:49Z
pr,177038,Add cublaslt backend for scaled grouped GEMM,https://github.com/pytorch/pytorch/pull/177038,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/b200, ciflow/dtensor, ciflow/h100, ciflow/inductor","identify whether block is CI, merge conflict, or review gate",2026-06-26T23:50:32Z
pr,172682,[NOT FOR MERGE] Initial test of features disabled when doing cooperative reduction,https://github.com/pytorch/pytorch/pull/172682,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-26T20:50:46Z
pr,187981,[not for land] CODA backward epilogue fusion prototype,https://github.com/pytorch/pytorch/pull/187981,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-06-26T18:08:13Z
pr,184892,xpu: use stride-based conv descriptors and preserve grouped layout mapping,https://github.com/pytorch/pytorch/pull/184892,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/xpu,"identify whether block is CI, merge conflict, or review gate",2026-06-26T16:49:16Z
pr,187606,"[inductor] Per-arch ROCm channels_last layout-opt multipliers (gfx942, gfx950)",https://github.com/pytorch/pytorch/pull/187606,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/inductor-perf-test-nightly-rocm-mi300, ciflow/inductor-rocm-mi300, ciflow/rocm-mi300","identify whether block is CI, merge conflict, or review gate",2026-06-26T06:23:55Z
pr,188184,Update definition of ROCM_VERSION in cpp_extension.py,https://github.com/pytorch/pytorch/pull/188184,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-06-25T22:04:23Z
pr,188189,Fix split_group non-member handling regression for mixed backends,https://github.com/pytorch/pytorch/pull/188189,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/torchtitan,"identify whether block is CI, merge conflict, or review gate",2026-06-25T20:57:34Z
pr,188168,Fix hipBLASLt FP32 precision degradation on MI300X (ROCM-3139),https://github.com/pytorch/pytorch/pull/188168,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/b200, ciflow/h100, ciflow/rocm-mi300","identify whether block is CI, merge conflict, or review gate",2026-06-25T18:20:01Z
pr,178862,[FSDP] enable async H2D prefetching for world_size=1,https://github.com/pytorch/pytorch/pull/178862,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/dtensor, ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-25T17:35:10Z
pr,185258,[ROCm][DO NOT MERGE][DEBUGGING] MI355 hang reproduce and dump subprocess logs,https://github.com/pytorch/pytorch/pull/185258,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-25T04:53:37Z
pr,187882,[RFC] Specialize float attributes of nn.Modules,https://github.com/pytorch/pytorch/pull/187882,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-25T01:17:23Z
pr,187464,[ROCm] [Operator Benchmark] Add gfx950 ROCm nightly CI lane and docker image,https://github.com/pytorch/pytorch/pull/187464,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/docker, ciflow/op-benchmark","identify whether block is CI, merge conflict, or review gate",2026-06-24T15:35:10Z
pr,187818,[WIP] generalize unitttests,https://github.com/pytorch/pytorch/pull/187818,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2026-06-24T13:46:58Z
pr,187739,[ROCm][CI] Misc updates,https://github.com/pytorch/pytorch/pull/187739,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-06-24T08:36:03Z
pr,187971,[inductor] Add strict_numerics config to emit INNER_TREE reduction ordering for sum,https://github.com/pytorch/pytorch/pull/187971,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-24T06:07:52Z
pr,181199,[spmd types] fully_shard in spmd_types,https://github.com/pytorch/pytorch/pull/181199,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-24T05:52:05Z
pr,188003,[AMD][hipBLASLt] Enable streak-K work stealing for GEMM operations,https://github.com/pytorch/pytorch/pull/188003,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/b200, ciflow/h100, ciflow/rocm-mi300","identify whether block is CI, merge conflict, or review gate",2026-06-24T03:00:52Z
pr,181449,AOTI: Add use_fake_constants mode for zero-weight compilation,https://github.com/pytorch/pytorch/pull/181449,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-24T01:19:07Z
pr,175486,[WIP][inductor] Layout allocator approach for symm_mem graph inputs,https://github.com/pytorch/pytorch/pull/175486,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/h100-symm-mem, ciflow/inductor","identify whether block is CI, merge conflict, or review gate",2026-06-23T23:49:13Z
pr,181392,[AC] Add example_inputs param to remat pass and mark recomputed nodes as backward,https://github.com/pytorch/pytorch/pull/181392,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-23T19:14:45Z
pr,186033,[CD] Enable Python 3.15 & 3.15t Windows wheel builds and Triton XPU Windows build,https://github.com/pytorch/pytorch/pull/186033,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/binaries_wheel,"identify whether block is CI, merge conflict, or review gate",2026-06-23T13:24:37Z
pr,181218,FSDP2 chunked-loss ROCm race: stream-level wait_event at post_backward exit,https://github.com/pytorch/pytorch/pull/181218,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/rocm-mi300, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-23T05:53:04Z
pr,181269,[draft] refactor linux workflows,https://github.com/pytorch/pytorch/pull/181269,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-06-23T05:53:03Z
pr,187815,[Inductor][XPU] Unfuse addmm for all half-dtype inputs on XPU,https://github.com/pytorch/pytorch/pull/187815,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-23T02:58:19Z
pr,179149,[Inductor][RFC] Symbolic Analysis of User-Defined Triton Kernels,https://github.com/pytorch/pytorch/pull/179149,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-23T01:21:13Z
pr,187892,"[profiler][cupti] Perfetto-native (.pftrace) export, full chrome-args parity",https://github.com/pytorch/pytorch/pull/187892,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-06-23T00:28:36Z
pr,181278,[spmd_types][unused] local_map for spmd_types tensors,https://github.com/pytorch/pytorch/pull/181278,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/dtensor, ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-22T22:13:56Z
pr,187258,[aot_compile] Add torch.compiler.precompile,https://github.com/pytorch/pytorch/pull/187258,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-22T20:16:20Z
pr,187840,[inductor] Add allow_peak_memory_increasing_fusion config,https://github.com/pytorch/pytorch/pull/187840,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-22T19:03:50Z
pr,187841,[inductor] Avoid disjoint horizontal fusion when allow_peak_memory_increasing_fusion is off,https://github.com/pytorch/pytorch/pull/187841,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-22T19:03:46Z
pr,187839,[inductor] Resolve ProcessGroup attrs in overlap estimates,https://github.com/pytorch/pytorch/pull/187839,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-22T19:03:41Z
pr,187843,[inductor] Avoid late-output vertical fusion when allow_peak_memory_increasing_fusion is off,https://github.com/pytorch/pytorch/pull/187843,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-22T19:03:38Z
pr,187846,[inductor] Pick low-pressure buffer reuse when allow_peak_memory_increasing_fusion is off,https://github.com/pytorch/pytorch/pull/187846,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-22T19:03:37Z
pr,187845,[inductor] Prefer successor-unblocking LPMF when allow_peak_memory_increasing_fusion is off,https://github.com/pytorch/pytorch/pull/187845,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-22T19:03:30Z
pr,187844,[inductor] Realize reused pointwise add when allow_peak_memory_increasing_fusion is off,https://github.com/pytorch/pytorch/pull/187844,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-22T17:48:12Z
pr,180986,Enable nested_compile_region for AOTInductor + tensor-subclass params,https://github.com/pytorch/pytorch/pull/180986,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-22T15:18:58Z
pr,180754,[ROCm] Fix decorator for skipping FP8 tests,https://github.com/pytorch/pytorch/pull/180754,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/b200, ciflow/h100, ciflow/inductor, ciflow/rocm-mi200","identify whether block is CI, merge conflict, or review gate",2026-06-22T11:23:18Z
pr,181189,Tensor.record_use: precise cross-stream lifetime for the CUDA caching allocator,https://github.com/pytorch/pytorch/pull/181189,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-22T08:34:15Z
pr,180867,Use SavedTensor for SAC cached tensors to enable saved tensor hooks,https://github.com/pytorch/pytorch/pull/180867,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/torchtitan,"identify whether block is CI, merge conflict, or review gate",2026-06-21T23:56:02Z
pr,187559,[MPS] Migrate nll_loss to native Metal kernels,https://github.com/pytorch/pytorch/pull/187559,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/mps,"identify whether block is CI, merge conflict, or review gate",2026-06-21T22:38:31Z
pr,180866,Add save_tensor() API for selective activation checkpointing,https://github.com/pytorch/pytorch/pull/180866,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/torchtitan,"identify whether block is CI, merge conflict, or review gate",2026-06-21T21:58:40Z
pr,181047,[FSDP2] Introduce StreamHandoff; migrate AR keep-alive,https://github.com/pytorch/pytorch/pull/181047,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/h100-symm-mem, ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-21T11:31:47Z
pr,180933,[spmd_types] fully_shard,https://github.com/pytorch/pytorch/pull/180933,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-20T23:52:43Z
pr,187756,[testintro][DEMO - DO NOT LAND] exercise the test-diff CI comment,https://github.com/pytorch/pytorch/pull/187756,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-06-20T21:52:03Z
pr,180924,Remove allow_cache_entry_mutation from SAC,https://github.com/pytorch/pytorch/pull/180924,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/torchtitan,"identify whether block is CI, merge conflict, or review gate",2026-06-20T20:02:56Z
pr,180925,[dynamo] Extract dynamic graph break hints,https://github.com/pytorch/pytorch/pull/180925,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-06-20T02:25:37Z
pr,187479,Add Python c10d Backend trampoline,https://github.com/pytorch/pytorch/pull/187479,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/torchtitan,"identify whether block is CI, merge conflict, or review gate",2026-06-19T00:29:35Z
pr,187624,[DRAFT]codeowner updates finer grained groups,https://github.com/pytorch/pytorch/pull/187624,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-06-18T22:26:02Z
pr,187674,Preserve boxed lifetimes for regional compiled calls,https://github.com/pytorch/pytorch/pull/187674,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-18T19:37:05Z
pr,184886,[inductor] Diagonal-symmetric mm TritonTemplate for Gram matrices,https://github.com/pytorch/pytorch/pull/184886,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/b200, ciflow/h100, ciflow/inductor, ciflow/rocm-mi300","identify whether block is CI, merge conflict, or review gate",2026-06-18T17:03:20Z
pr,187435,[inductor] Support control_deps as fusion barriers,https://github.com/pytorch/pytorch/pull/187435,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-18T16:30:22Z
pr,187636,[AOTI] Flip the default value of autotune_at_compile_time to False,https://github.com/pytorch/pytorch/pull/187636,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-18T07:07:07Z
pr,187640,Support in-tree XNNPACK build on Windows ARM64,https://github.com/pytorch/pytorch/pull/187640,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-06-18T04:46:24Z
pr,187472,[dynamo] don't reconstruct live torch.Stream values as StreamContext across graph breaks,https://github.com/pytorch/pytorch/pull/187472,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-17T00:00:58Z
pr,180684,AOTI: Write tensor data directly to file without Python bytes copy,https://github.com/pytorch/pytorch/pull/180684,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-16T20:31:37Z
pr,180683,AOTI: Stream external weights to disk instead of in-memory blob,https://github.com/pytorch/pytorch/pull/180683,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-16T18:53:41Z
pr,180682,AOTI: Only clone mutated buffers in _unlift_graph,https://github.com/pytorch/pytorch/pull/180682,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-16T18:53:41Z
pr,185612,[WIP][CI][CRCR] Update mergebot rule to adapt CRCR L3,https://github.com/pytorch/pytorch/pull/185612,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-06-16T15:35:35Z
pr,177648,LinAlgUtils -- update the toolkit to address the MathLibs dispatch with redundant memory allocations/movements.,https://github.com/pytorch/pytorch/pull/177648,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2026-06-16T12:23:27Z
pr,186752,[Inductor][Triton] Add host-side TMA support for B200 GEMM templates,https://github.com/pytorch/pytorch/pull/186752,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/b200, ciflow/h100, ciflow/inductor, ciflow/rocm-mi300","identify whether block is CI, merge conflict, or review gate",2026-06-16T08:19:37Z
pr,186756,[Inductor][Host-side TMA GEMM] Enable host-side TMA for TMA epilogue store,https://github.com/pytorch/pytorch/pull/186756,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-16T08:19:28Z
pr,186988,Move header-only c10/util Layer 1 utils to torch/headeronly,https://github.com/pytorch/pytorch/pull/186988,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/docker, ciflow/inductor","identify whether block is CI, merge conflict, or review gate",2026-06-16T06:07:48Z
pr,186987,Move header-only c10/util large/vendored headers to torch/headeronly,https://github.com/pytorch/pytorch/pull/186987,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/docker, ciflow/inductor","identify whether block is CI, merge conflict, or review gate",2026-06-16T06:07:36Z
pr,186986,Move header-only c10/util string/filesystem utils to torch/headeronly,https://github.com/pytorch/pytorch/pull/186986,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2026-06-16T06:06:59Z
pr,186984,Move header-only c10/util container/storage utils to torch/headeronly,https://github.com/pytorch/pytorch/pull/186984,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2026-06-16T06:06:56Z
pr,180423,Fix checkpoint DeviceContext leak onto TorchFunction mode stack,https://github.com/pytorch/pytorch/pull/180423,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/torchtitan,"identify whether block is CI, merge conflict, or review gate",2026-06-16T03:48:36Z
pr,183579,[OpenReg][Quantization] Register dequantize_tensor_per_tensor_affine stub for PrivateUse1,https://github.com/pytorch/pytorch/pull/183579,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-06-16T02:38:34Z
pr,183270,[inductor][overlap] Post-scheduling foreach_mm for optimizer mm batching,https://github.com/pytorch/pytorch/pull/183270,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-15T22:49:47Z
pr,182631,[aten] Add aten._foreach_mm operator,https://github.com/pytorch/pytorch/pull/182631,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/b200, ciflow/h100, ciflow/inductor, ciflow/rocm-mi300","identify whether block is CI, merge conflict, or review gate",2026-06-15T22:47:50Z
pr,180597,[nocommit] symmem custreamwaitvalue32,https://github.com/pytorch/pytorch/pull/180597,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/h100-symm-mem,"identify whether block is CI, merge conflict, or review gate",2026-06-15T22:18:01Z
pr,180407,[nocommit] symmem nccl ib,https://github.com/pytorch/pytorch/pull/180407,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/h100-symm-mem,"identify whether block is CI, merge conflict, or review gate",2026-06-15T22:18:00Z
pr,180411,stateless dtensor,https://github.com/pytorch/pytorch/pull/180411,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/dtensor, ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-15T01:36:39Z
pr,180409,stateless ghstack,https://github.com/pytorch/pytorch/pull/180409,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-06-14T21:56:10Z
pr,180410,debug,https://github.com/pytorch/pytorch/pull/180410,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-06-14T21:56:09Z
pr,177438,Infer device_type in fork_rng from devices argument,https://github.com/pytorch/pytorch/pull/177438,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-06-14T18:55:52Z
pr,180412,stateless spmd,https://github.com/pytorch/pytorch/pull/180412,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-06-14T06:06:23Z
pr,187249,perf(dynamo): add guard partial memo,https://github.com/pytorch/pytorch/pull/187249,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2026-06-13T16:44:55Z
pr,177401,Allow non-CL images to go through NEON path in `F.interpolate()` for bilinear and bicubic modes when antialias=True.,https://github.com/pytorch/pytorch/pull/177401,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-06-13T15:10:10Z
pr,187056,[TESTING] Separated ROCm triton commit pin,https://github.com/pytorch/pytorch/pull/187056,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/docker, ciflow/inductor, ciflow/inductor-rocm, ciflow/rocm-mi300","identify whether block is CI, merge conflict, or review gate",2026-06-13T09:03:51Z
pr,180207,[TEST] Refactor test_structured_trace.py to be device-generic,https://github.com/pytorch/pytorch/pull/180207,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-06-13T07:47:52Z
pr,187116,Add stable tie handling to topk,https://github.com/pytorch/pytorch/pull/187116,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/mps, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-13T02:08:04Z
pr,177471,tag on ops,https://github.com/pytorch/pytorch/pull/177471,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/dtensor, ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-12T22:03:54Z
pr,187050,[TESTING] [Inductor] Add regression tests for large-batch BMM grid Y/Z split,https://github.com/pytorch/pytorch/pull/187050,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/inductor-rocm, ciflow/rocm-mi300, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-12T18:01:04Z
pr,186343,Enable FA build for SDPA on Windows+CUDA,https://github.com/pytorch/pytorch/pull/186343,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/binaries_wheel,"identify whether block is CI, merge conflict, or review gate",2026-06-12T11:45:28Z
pr,179675,[WIP]port dynamo cases to Intel GPU,https://github.com/pytorch/pytorch/pull/179675,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2026-06-12T08:05:06Z
pr,180192,[DO NOT REVIEW] fix torchtitan CI error,https://github.com/pytorch/pytorch/pull/180192,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/b200, ciflow/h100, ciflow/rocm-mi300","identify whether block is CI, merge conflict, or review gate",2026-06-12T03:48:16Z
pr,187004,perf(dynamo): add guard lookup fast paths,https://github.com/pytorch/pytorch/pull/187004,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-11T14:53:09Z
pr,186753,[fx] Promote raw-SymInt arg warning to RuntimeError,https://github.com/pytorch/pytorch/pull/186753,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-10T19:35:55Z
pr,178485,[DO NOT MERGE][Inductor] Migrate oneDNN Inner Product to oneDNN MatMul for MKLDNN linear ops,https://github.com/pytorch/pytorch/pull/178485,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-10T12:04:13Z
pr,186590,[Draft][xpu] Add NVFP4 support on Intel GPU,https://github.com/pytorch/pytorch/pull/186590,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/b200, ciflow/h100, ciflow/inductor, ciflow/rocm-mi300","identify whether block is CI, merge conflict, or review gate",2026-06-10T08:10:27Z
pr,186209,Unify triton-xpu CD nightly build into 'triton' package with version suffix,https://github.com/pytorch/pytorch/pull/186209,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-06-10T06:01:31Z
pr,186835,[RFC]: Auto-select and run Operator Performance Benchmarks from PR Diffs,https://github.com/pytorch/pytorch/pull/186835,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-06-10T02:01:33Z
pr,178277,Skip fake autocast tests for `linalg.pinv` and `pinverse` on XPU,https://github.com/pytorch/pytorch/pull/178277,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/xpu,"identify whether block is CI, merge conflict, or review gate",2026-06-09T23:09:15Z
pr,186798,[TESTING ONLY] [ROCm] testing heuristic config approach,https://github.com/pytorch/pytorch/pull/186798,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/inductor-perf-test-nightly-rocm-mi350, ciflow/inductor-rocm-mi355, ciflow/rocm-mi355","identify whether block is CI, merge conflict, or review gate",2026-06-09T23:08:29Z
pr,186541,[MPS] [WIP] Port all MPSGraph matmuls to Metal,https://github.com/pytorch/pytorch/pull/186541,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/mps,"identify whether block is CI, merge conflict, or review gate",2026-06-09T19:15:22Z
pr,186378,[activation_checkpointing] Support AutoNamingMode + policy name keys under torch.compile,https://github.com/pytorch/pytorch/pull/186378,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-05T19:51:42Z
pr,186382,[activation_checkpointing] AutoNamingMode + name/module-FQN keys in checkpoint policy= (eager),https://github.com/pytorch/pytorch/pull/186382,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-05T19:51:35Z
pr,186277,[activation_checkpointing] Add policy= kwarg and name() to torch.utils.checkpoint,https://github.com/pytorch/pytorch/pull/186277,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-05T19:09:33Z
pr,185555,Add torch._rust Rust extension and build plumbing (#185555),https://github.com/pytorch/pytorch/pull/185555,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/docker, ciflow/inductor","identify whether block is CI, merge conflict, or review gate",2026-06-05T06:35:13Z
pr,179769,[cuda graphs] Integrate kernel annotations into inductor cudagraph trees,https://github.com/pytorch/pytorch/pull/179769,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-04T17:45:42Z
pr,186002,"Temporary Commit at 6/2/2026, 2:55:48 PM",https://github.com/pytorch/pytorch/pull/186002,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-06-03T03:45:31Z
pr,185935,[DO NOT MERGE] ci: add manywheel build telemetry (May 6f7bb01c),https://github.com/pytorch/pytorch/pull/185935,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-06-02T18:31:35Z
pr,185822,[DO NOT MERGE] ci: add manywheel build telemetry (Feb baseline 0cd9ecaf),https://github.com/pytorch/pytorch/pull/185822,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/binaries_wheel, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-02T14:54:15Z
pr,185936,[DO NOT MERGE] ci: add manywheel build telemetry (Apr midpoint 977c5623),https://github.com/pytorch/pytorch/pull/185936,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-06-02T14:31:34Z
pr,185706,Fix typos across comments and docstrings in torch,https://github.com/pytorch/pytorch/pull/185706,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/dtensor, ciflow/h100-symm-mem, ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-02T14:01:32Z
pr,181578,[Draft] [04/27/26] Dashboard result update for ROCm,https://github.com/pytorch/pytorch/pull/181578,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/inductor-periodic, ciflow/inductor-rocm-mi300, ciflow/inductor-rocm-mi355","identify whether block is CI, merge conflict, or review gate",2026-06-02T11:45:26Z
pr,183626,[Inductor] Seed combo-kernel autotune from standalone subkernels,https://github.com/pytorch/pytorch/pull/183626,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-02T05:18:51Z
pr,185835,[c10d] Add delegate backend to FakeProcessGroup for numerics testing,https://github.com/pytorch/pytorch/pull/185835,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-06-01T21:41:43Z
pr,185286,Add version counter check for gradient tensors in the autograd engine,https://github.com/pytorch/pytorch/pull/185286,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-06-01T14:40:50Z
pr,185787,[WIP] Migrate XPU test runners to linux.xpu.test and disable legacy jobs,https://github.com/pytorch/pytorch/pull/185787,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/xpu,"identify whether block is CI, merge conflict, or review gate",2026-06-01T07:12:16Z
pr,185712,"Fix typos in comments, docstrings, and strings across the codebase",https://github.com/pytorch/pytorch/pull/185712,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/dtensor, ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-05-30T20:18:47Z
pr,185713,"Fix typos in comments, docstrings, and error messages",https://github.com/pytorch/pytorch/pull/185713,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/dtensor, ciflow/h100-symm-mem, ciflow/inductor, ciflow/mps","identify whether block is CI, merge conflict, or review gate",2026-05-30T19:27:43Z
pr,185705,"Fix typos in comments, docstrings, and error messages",https://github.com/pytorch/pytorch/pull/185705,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-05-30T16:46:33Z
pr,185565,upstream torch.comms,https://github.com/pytorch/pytorch/pull/185565,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/docker, ciflow/h100-distributed, ciflow/h100-symm-mem, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-05-30T09:28:11Z
pr,185339,[not-for-land][aten] Add cuBLAS/cublasLt grouped GEMM backends for _foreach_mm,https://github.com/pytorch/pytorch/pull/185339,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/b200, ciflow/h100, ciflow/rocm-mi300","identify whether block is CI, merge conflict, or review gate",2026-05-29T22:43:48Z
pr,185548,[test] Migrate in-test OpInfo skip to OpInfo skip DB,https://github.com/pytorch/pytorch/pull/185548,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-05-29T06:28:59Z
pr,177445,Add SymInt and SymFloat to Number,https://github.com/pytorch/pytorch/pull/177445,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-05-29T02:21:36Z
pr,185364,[not-for-land][native_dsl] Add _grouped_mm_from_ptrs pointer-packing path,https://github.com/pytorch/pytorch/pull/185364,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/b200, ciflow/h100, ciflow/rocm-mi300","identify whether block is CI, merge conflict, or review gate",2026-05-28T08:58:03Z
pr,185172,Add support to custom devices to operator_benchmark.,https://github.com/pytorch/pytorch/pull/185172,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/op-benchmark,"identify whether block is CI, merge conflict, or review gate",2026-05-28T08:57:01Z
pr,185363,[not-for-land][native_dsl] Python _foreach_mm override via stack + _grouped_mm,https://github.com/pytorch/pytorch/pull/185363,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-05-28T07:42:14Z
pr,185457,[Inductor][xpu] Enable grouped_mm Triton template on XPU with TMA disabled,https://github.com/pytorch/pytorch/pull/185457,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/b200, ciflow/h100, ciflow/inductor, ciflow/rocm-mi300","identify whether block is CI, merge conflict, or review gate",2026-05-28T07:41:19Z
pr,185376,[not-for-land][native_dsl] Add _foreach_mm_from_ptrs pointer-bridge and py_ptrs override,https://github.com/pytorch/pytorch/pull/185376,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/b200, ciflow/h100, ciflow/rocm-mi300","identify whether block is CI, merge conflict, or review gate",2026-05-27T23:06:36Z
pr,184302,Add Stream-K MM backend support to Inductor,https://github.com/pytorch/pytorch/pull/184302,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/b200, ciflow/h100, ciflow/inductor, ciflow/inductor-rocm-mi300","identify whether block is CI, merge conflict, or review gate",2026-05-27T21:13:51Z
pr,185340,[not-for-land][native_dsl] Add Python _foreach_mm override via _grouped_mm,https://github.com/pytorch/pytorch/pull/185340,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/b200, ciflow/h100, ciflow/rocm-mi300","identify whether block is CI, merge conflict, or review gate",2026-05-27T15:01:38Z
pr,185284,Add checkpoint_name() API for explicit tensor naming in SAC,https://github.com/pytorch/pytorch/pull/185284,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-05-27T01:25:54Z
pr,185287,Add memory_budget() context manager for per-region activation memory budgets,https://github.com/pytorch/pytorch/pull/185287,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-05-27T01:25:20Z
pr,185285,Allow debug mode with context_fn in checkpoint(),https://github.com/pytorch/pytorch/pull/185285,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/torchtitan,"identify whether block is CI, merge conflict, or review gate",2026-05-27T00:21:48Z
pr,184991,[inductor] Add memory-aware graph-level CSE pass,https://github.com/pytorch/pytorch/pull/184991,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-05-26T22:23:37Z
pr,185253,[c10d] Make FakeProcessGroup reduce collectives numerically consistent,https://github.com/pytorch/pytorch/pull/185253,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-05-26T19:37:58Z
pr,185219,[cherry-pick] no topological sort in manual overlaps,https://github.com/pytorch/pytorch/pull/185219,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-05-26T12:36:13Z
pr,185218,[not-for-land][squash] Regional inductor cache fix,https://github.com/pytorch/pytorch/pull/185218,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/dtensor, ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-05-26T12:36:12Z
pr,159718,[Don't Review] Test XPU CI,https://github.com/pytorch/pytorch/pull/159718,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/trunk, ciflow/xpu","identify whether block is CI, merge conflict, or review gate",2026-05-26T10:08:57Z
pr,184909,[Inductor][Draft] A Helion backend for Inductor,https://github.com/pytorch/pytorch/pull/184909,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-05-25T23:36:49Z
issue,171675,PyTorch quantization `QConfig` triggers INTERNAL ASSERT FAILED when printing config with invalid observer `dtype`,https://github.com/pytorch/pytorch/issues/171675,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,categorize and decide next maintainer owner/action,2026-01-06T04:13:26Z
issue,171516,How to verify that default_decompositions successfully reduce operators to the Core ATen IR set?,https://github.com/pytorch/pytorch/issues/171516,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: export, oncall: pt2",categorize and decide next maintainer owner/action,2026-01-05T16:23:29Z
issue,171377,PyTorch JIT `get_debug_state()` triggers INTERNAL ASSERT FAILED after `torch.jit.trace`,https://github.com/pytorch/pytorch/issues/171377,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2026-01-05T14:58:15Z
issue,154613,`torch.jit.script` doesn't accept `axis` as alias for `dim`,https://github.com/pytorch/pytorch/issues/154613,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2026-01-01T19:19:47Z
issue,170548,unable to export a model with bucketize,https://github.com/pytorch/pytorch/issues/170548,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: export, oncall: pt2",categorize and decide next maintainer owner/action,2025-12-23T01:32:59Z
issue,170783,New higher order ops to handle loops,https://github.com/pytorch/pytorch/issues/170783,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: export, oncall: pt2",categorize and decide next maintainer owner/action,2025-12-21T04:07:21Z
issue,170680,Error when composing jit.script and jacrev,https://github.com/pytorch/pytorch/issues/170680,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2025-12-20T23:38:52Z
issue,170673,`torch.export` Fails with RuntimeError when Accessing Parameter via `.data` in `forward`,https://github.com/pytorch/pytorch/issues/170673,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: export, oncall: pt2",categorize and decide next maintainer owner/action,2025-12-20T23:36:42Z
issue,170774,CUPTI metrics not recorded in profiler,https://github.com/pytorch/pytorch/issues/170774,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: profiler,categorize and decide next maintainer owner/action,2025-12-20T23:06:06Z
issue,170867,Operator benchmark: option to measure GPU execution time only (less CPU noise),https://github.com/pytorch/pytorch/issues/170867,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: profiler,categorize and decide next maintainer owner/action,2025-12-20T22:52:15Z
issue,170319,torch.profiler shows GPU ops (Triton / FlashAttention) to have `cpu_op` category for some reason,https://github.com/pytorch/pytorch/issues/170319,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: profiler,categorize and decide next maintainer owner/action,2025-12-17T04:29:26Z
issue,170169,questions about default_weight_fake_quant config with pytorch quantizations,https://github.com/pytorch/pytorch/issues/170169,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,categorize and decide next maintainer owner/action,2025-12-11T16:51:50Z
issue,170137,PyTorch memory profiler conflicts with torch.util.swap_tensors due to strict reference counter check,https://github.com/pytorch/pytorch/issues/170137,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: python frontend, oncall: profiler",categorize and decide next maintainer owner/action,2025-12-11T04:40:15Z
issue,161179,Consolidate magma-cuda build for Linux and Windows,https://github.com/pytorch/pytorch/issues/161179,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: ci, oncall: releng",categorize and decide next maintainer owner/action,2025-12-08T20:38:41Z
issue,169666,Inconsistent`flop_counter.get_total_flops()` results for a custom decoder model,https://github.com/pytorch/pytorch/issues/169666,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: profiler,categorize and decide next maintainer owner/action,2025-12-05T15:10:05Z
issue,165829,Floating Point Exception (Core Dumped) with Dynamic Quantization on nn.Linear with Output Dimension 0,https://github.com/pytorch/pytorch/issues/165829,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,categorize and decide next maintainer owner/action,2025-12-05T00:45:18Z
issue,169357,"prepareProfiler() skips collecting info of ""privateuseone""",https://github.com/pytorch/pytorch/issues/169357,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: privateuse1, oncall: profiler",categorize and decide next maintainer owner/action,2025-12-03T12:11:39Z
issue,162481,Incosistent tracking of device activities when calling profiler.step() in torch profiler,https://github.com/pytorch/pytorch/issues/162481,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: profiler,categorize and decide next maintainer owner/action,2025-12-01T18:41:45Z
issue,151536,[inductor][cpu] pytorch_CycleGAN_and_pix2pix AMP/AMP_FP16 multiple thread performance regression in 2025-04-07 nightly release,https://github.com/pytorch/pytorch/issues/151536,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: cpu inductor, oncall: pt2",categorize and decide next maintainer owner/action,2025-11-26T07:19:42Z
issue,169069,Triton Pin update 3.6 related - All periodic linux-jammy-cuda12.4-py3.10-gcc11 are in broken state. Some linux-jammy-cuda12.8-py3.10-gcc11-debug are broken as well,https://github.com/pytorch/pytorch/issues/169069,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: ci, oncall: releng",categorize and decide next maintainer owner/action,2025-11-25T18:48:55Z
issue,169009,Dynamo export: Qwen3 Omni model running into data-dependent expression errors,https://github.com/pytorch/pytorch/issues/169009,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: export, oncall: pt2",categorize and decide next maintainer owner/action,2025-11-25T00:39:27Z
issue,168077,scripted models wrongly cast integers to floats,https://github.com/pytorch/pytorch/issues/168077,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2025-11-18T21:39:56Z
issue,65607,torch.dtype is int in JIT instead of torch.dtype,https://github.com/pytorch/pytorch/issues/65607,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2025-11-18T11:42:23Z
issue,167730,Feedback about ExportDB,https://github.com/pytorch/pytorch/issues/167730,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: export, oncall: pt2",categorize and decide next maintainer owner/action,2025-11-15T01:35:00Z
issue,147818,Bug in `torch.ao.nn.quantized.Sigmoid` Parameter Restoration after `state_dict ` Loading,https://github.com/pytorch/pytorch/issues/147818,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,categorize and decide next maintainer owner/action,2025-11-14T18:46:17Z
issue,167199,[CPU] torch.compile consumes too much extra memory,https://github.com/pytorch/pytorch/issues/167199,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: cpu inductor, oncall: pt2",categorize and decide next maintainer owner/action,2025-11-14T05:34:44Z
issue,167412,How can I train in C++ using a Pytorch torchscript model,https://github.com/pytorch/pytorch/issues/167412,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2025-11-10T19:11:24Z
issue,126703,PyTorch's Execution Trace Observer Producing Invalid JSON,https://github.com/pytorch/pytorch/issues/126703,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: profiler,categorize and decide next maintainer owner/action,2025-11-09T03:36:41Z
issue,166939,Reduce aten::empty Calls for Models with Constant Input Shapes in AOTInductor,https://github.com/pytorch/pytorch/issues/166939,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: aotinductor, oncall: export, oncall: pt2",categorize and decide next maintainer owner/action,2025-11-08T00:24:54Z
issue,166311,Unable to export profiler trace,https://github.com/pytorch/pytorch/issues/166311,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: profiler,categorize and decide next maintainer owner/action,2025-11-07T09:19:37Z
issue,47163,Grad strides do not match bucket view strides.,https://github.com/pytorch/pytorch/issues/47163,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: ddp, module: memory format, oncall: distributed",categorize and decide next maintainer owner/action,2025-11-06T15:17:58Z
issue,164145,Improvements to profiler for bitwise equivalence use case,https://github.com/pytorch/pytorch/issues/164145,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: profiler,categorize and decide next maintainer owner/action,2025-10-26T03:18:33Z
issue,165395,[export] run_decompositions fails with index error,https://github.com/pytorch/pytorch/issues/165395,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: export, oncall: pt2",categorize and decide next maintainer owner/action,2025-10-21T21:37:04Z
issue,165929,Numerical mismatch b/w eager & aot_eager_decomp_partition backend in SegFormer layer (2.9.0),https://github.com/pytorch/pytorch/issues/165929,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: aotinductor, oncall: export, oncall: pt2",categorize and decide next maintainer owner/action,2025-10-21T14:27:18Z
issue,165900,Converting weights `.pt` content between `dict` and `RecursiveScriptModule`,https://github.com/pytorch/pytorch/issues/165900,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2025-10-20T14:27:26Z
issue,128114,Import Error: cannot import name 'XNNPACKQuantizer' from 'torch.ao.quantization.quantizer',https://github.com/pytorch/pytorch/issues/128114,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,categorize and decide next maintainer owner/action,2025-10-20T11:35:30Z
issue,163791,torch.export fails on Phi-3.5 model,https://github.com/pytorch/pytorch/issues/163791,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: export, oncall: pt2",categorize and decide next maintainer owner/action,2025-10-18T17:54:39Z
issue,165230,_int_mm is slow with row-major rhs matrix,https://github.com/pytorch/pytorch/issues/165230,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,categorize and decide next maintainer owner/action,2025-10-17T04:28:23Z
issue,165400,aten::bitwise_xor,https://github.com/pytorch/pytorch/issues/165400,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2025-10-14T13:33:35Z
issue,156814,torchscript: RuntimeError Unsupported dtype complex,https://github.com/pytorch/pytorch/issues/156814,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: complex, oncall: jit",categorize and decide next maintainer owner/action,2025-10-14T07:40:12Z
issue,164083,Broaden support of unflatten to split graphs,https://github.com/pytorch/pytorch/issues/164083,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: export,categorize and decide next maintainer owner/action,2025-10-07T15:50:53Z
issue,164534,DISABLED RecordDebugHandles.ScopedCallbacks (__main__.test_jit),https://github.com/pytorch/pytorch/issues/164534,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: flaky-tests, module: unknown, oncall: jit",categorize and decide next maintainer owner/action,2025-10-03T14:31:07Z
issue,164208,CUDA Illegal Memory Access in Multi-threaded LibTorch Applications,https://github.com/pytorch/pytorch/issues/164208,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2025-09-30T18:31:23Z
issue,69178,DISABLED test_remote_timeout_to_here_in_jit (__main__.FaultyJitFaultyAgentRpcTest),https://github.com/pytorch/pytorch/issues/69178,needs triage,47,"triage signal present; old update age or low-urgency/stale label; component/oncall labels: module: flaky-tests, module: rpc, oncall: distributed, oncall: jit",categorize and decide next maintainer owner/action,2025-09-29T23:19:15Z
issue,70865,Support hooks-based checkpointing API with DDP,https://github.com/pytorch/pytorch/issues/70865,needs triage,47,"triage signal present; old update age or low-urgency/stale label; component/oncall labels: module: ddp, oncall: distributed",categorize and decide next maintainer owner/action,2025-09-26T23:46:38Z
issue,163945,Failed to run a backward op in TorchScript interpreter,https://github.com/pytorch/pytorch/issues/163945,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2025-09-26T11:32:33Z
issue,149737,`INTERNAL ASSERT FAILED` in `interpolate` and `torch.import_ir_module`,https://github.com/pytorch/pytorch/issues/149737,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: error checking, oncall: quantization",categorize and decide next maintainer owner/action,2025-09-21T09:47:56Z
issue,162948,Metadata object loses arbitrary attributes added dynamically,https://github.com/pytorch/pytorch/issues/162948,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: distributed checkpointing,categorize and decide next maintainer owner/action,2025-09-18T01:13:04Z
issue,162365,[Windows][CPU] 3 UT of test/inductor/test_torchinductor.py failed with 0903 nightly whl on windows,https://github.com/pytorch/pytorch/issues/162365,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: windows, oncall: cpu inductor, oncall: pt2",categorize and decide next maintainer owner/action,2025-09-11T15:30:21Z
issue,162451,Some Windows AOTI UTs regression,https://github.com/pytorch/pytorch/issues/162451,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: aotinductor, module: dynamo, module: windows, oncall: export",categorize and decide next maintainer owner/action,2025-09-11T00:12:36Z
issue,162135,[inductor][cpu]dla102 AMP default wrapper hang issue,https://github.com/pytorch/pytorch/issues/162135,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: cpu inductor, oncall: pt2",categorize and decide next maintainer owner/action,2025-09-10T22:50:40Z
issue,161815,Windows AMI release issues and improvements,https://github.com/pytorch/pytorch/issues/161815,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: windows, oncall: releng",categorize and decide next maintainer owner/action,2025-09-09T20:26:36Z
issue,130407,[RFC][c10d] a new Pytorch API (split_group) to create a process group through ncclCommSplit,https://github.com/pytorch/pytorch/issues/130407,needs triage,47,"triage signal present; old update age or low-urgency/stale label; component/oncall labels: module: c10d, oncall: distributed",categorize and decide next maintainer owner/action,2025-09-05T23:03:35Z
issue,161935,torch.jit.trace_module.register_parameter :cannot assign 'torch.cuda.FloatTensor' object to parameter 'new_param' (torch.nn.Parameter or None required),https://github.com/pytorch/pytorch/issues/161935,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2025-09-02T13:04:18Z
issue,161854,torch.ScriptObjectProperty.name.deleter occurs Segmentation fault (core dumped),https://github.com/pytorch/pytorch/issues/161854,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2025-09-02T06:08:31Z
issue,154094,"[inductor][cpu] aoti shared memory run on multiple cores, performance will drop",https://github.com/pytorch/pytorch/issues/154094,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: aotinductor, oncall: cpu inductor, oncall: export, oncall: pt2",categorize and decide next maintainer owner/action,2025-09-01T03:00:33Z
issue,63476,Error loading Pytorch lite model in Android,https://github.com/pytorch/pytorch/issues/63476,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: mobile,categorize and decide next maintainer owner/action,2025-08-29T14:27:38Z
issue,157728,"fill_ overflows on uint64 in range [2**63, 2**64) when profiler is engaged",https://github.com/pytorch/pytorch/issues/157728,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: profiler,categorize and decide next maintainer owner/action,2025-08-28T16:11:29Z
issue,151969,Unrecognized in PyTorch traces w/ Cuda 12.6,https://github.com/pytorch/pytorch/issues/151969,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: profiler,categorize and decide next maintainer owner/action,2025-08-18T19:05:28Z
issue,49460,Libtorch: Segmentation fault when running torch::jit::load,https://github.com/pytorch/pytorch/issues/49460,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2025-08-17T10:50:56Z
issue,160631,torch.quantize_per_tensor not implemented for SparseCPU tensors,https://github.com/pytorch/pytorch/issues/160631,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: sparse, oncall: quantization",categorize and decide next maintainer owner/action,2025-08-15T23:23:52Z
issue,160391,`torch.compile` backward pass fails with `AssertionError` in Inductor C++ codegen when model returns tuple outputs,https://github.com/pytorch/pytorch/issues/160391,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: cpu inductor, oncall: pt2",categorize and decide next maintainer owner/action,2025-08-15T18:28:19Z
issue,160392,bug in libtorch optimize_for_inference,https://github.com/pytorch/pytorch/issues/160392,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2025-08-13T20:33:33Z
issue,160005,[BE][download.pytorch.org] Index fix user navigation,https://github.com/pytorch/pytorch/issues/160005,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: releng,categorize and decide next maintainer owner/action,2025-08-12T20:28:33Z
issue,144770,[ARM] multiple test failures in TestQuantizedConv on Aarch64,https://github.com/pytorch/pytorch/issues/144770,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: arm, module: tests, oncall: quantization",categorize and decide next maintainer owner/action,2025-08-12T09:43:42Z
issue,159316,Support for non-scalar valued Loss Tensor in `_export_forward_backward()`,https://github.com/pytorch/pytorch/issues/159316,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: export, oncall: pt2",categorize and decide next maintainer owner/action,2025-08-09T00:57:34Z
issue,159505,Cannot pass through None for example_inputs in prepare_fx,https://github.com/pytorch/pytorch/issues/159505,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,categorize and decide next maintainer owner/action,2025-08-07T17:25:26Z
issue,55282,Poor error message when trying to jit a function instead of a module (RuntimeError: Cannot insert a Tensor that requires grad as a constant.),https://github.com/pytorch/pytorch/issues/55282,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2025-08-07T14:51:52Z
issue,159169,[DTensor][FSDP2][DDP] benchmark dtensor cpu overhead for adam optimizer,https://github.com/pytorch/pytorch/issues/159169,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: profiler,categorize and decide next maintainer owner/action,2025-08-07T03:39:02Z
issue,159256,can kineto profile only one thread with corresponding gpu tasks among multithreads,https://github.com/pytorch/pytorch/issues/159256,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: profiler,categorize and decide next maintainer owner/action,2025-07-31T15:28:22Z
issue,159269,Feature Request: Allow torch::inductor::ModelPackageLoader to load from an in-memory buffer,https://github.com/pytorch/pytorch/issues/159269,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: aotinductor, oncall: export, oncall: pt2",categorize and decide next maintainer owner/action,2025-07-29T14:54:04Z
issue,159265,Unpin dependency version when possible,https://github.com/pytorch/pytorch/issues/159265,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: cuda, oncall: package/deploy",categorize and decide next maintainer owner/action,2025-07-28T22:01:03Z
issue,158331,DISABLED test_conv2d_unary_cpu_cpp_wrapper (__main__.TestCppWrapper),https://github.com/pytorch/pytorch/issues/158331,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: flaky-tests, oncall: cpu inductor",categorize and decide next maintainer owner/action,2025-07-15T06:50:49Z
issue,158227,Degraded JIT execution performance on 5090 Windows,https://github.com/pytorch/pytorch/issues/158227,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2025-07-14T18:08:15Z
issue,156915,Unable to compile with Vulkan support,https://github.com/pytorch/pytorch/issues/156915,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: build, module: vulkan, oncall: mobile",categorize and decide next maintainer owner/action,2025-07-14T07:21:35Z
issue,148902,Remove Direct Arm Compute Libray (ACL) Integration for Quantized Matmuls: `qlinear`/`qlinear_dynamic`,https://github.com/pytorch/pytorch/issues/148902,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: arm, oncall: quantization",categorize and decide next maintainer owner/action,2025-07-09T17:03:58Z
issue,155862,Export Huggingface models with StaticCache,https://github.com/pytorch/pytorch/issues/155862,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: export, oncall: pt2",categorize and decide next maintainer owner/action,2025-07-09T16:48:49Z
issue,156882,Export always give a value range with max length - 1,https://github.com/pytorch/pytorch/issues/156882,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: export, oncall: pt2",categorize and decide next maintainer owner/action,2025-07-09T00:55:21Z
issue,157625,Is there some official method to extract the featuremap of each node in pt2 graph like the function torchvision.models.feature_extraction.create_feature_extractor(),https://github.com/pytorch/pytorch/issues/157625,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: export, oncall: pt2",categorize and decide next maintainer owner/action,2025-07-08T17:33:56Z
issue,153429,[Export] Non-strict mode can't handle conditionals on tensor subclass types,https://github.com/pytorch/pytorch/issues/153429,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: export, oncall: pt2",categorize and decide next maintainer owner/action,2025-07-07T17:41:39Z
issue,157603,Incorrect inference of the groups parameter type for channel_stuffle (int misclassified as Tensor),https://github.com/pytorch/pytorch/issues/157603,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2025-07-07T08:07:53Z
issue,156544,`TorchScript` does not allow accessing methods of nested tensors,https://github.com/pytorch/pytorch/issues/156544,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: nestedtensor, oncall: jit",categorize and decide next maintainer owner/action,2025-06-30T10:56:11Z
issue,157103,Issue with super() resolution in TorchScript when inheriting from custom class and nn.Module,https://github.com/pytorch/pytorch/issues/157103,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2025-06-30T01:57:19Z
issue,156955,`torch.jit.load()` crashes with C++ `c10::Error` if file system is full (no graceful handling),https://github.com/pytorch/pytorch/issues/156955,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: edge cases, module: error checking, oncall: jit",categorize and decide next maintainer owner/action,2025-06-26T14:49:29Z
issue,156558,Using collections. namedtuple in the forward method of the model resulted in compilation failure,https://github.com/pytorch/pytorch/issues/156558,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: export, oncall: jit, oncall: pt2",categorize and decide next maintainer owner/action,2025-06-26T01:29:10Z
issue,156557,[JIT] TorchScript fails to compile model using random.choice() with confusing error message,https://github.com/pytorch/pytorch/issues/156557,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2025-06-22T19:22:45Z
issue,112844,torch.export does not support torchaudio.transforms.Spectrogram,https://github.com/pytorch/pytorch/issues/112844,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: export",categorize and decide next maintainer owner/action,2025-06-20T23:40:06Z
issue,155550,"[torch.export] Cannot export TorchVision raft_small, raft_large",https://github.com/pytorch/pytorch/issues/155550,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: export, oncall: pt2",categorize and decide next maintainer owner/action,2025-06-19T10:50:53Z
issue,155321,mishandling torch.package.PackageExporter raises Aborted when given a tensor instead of an importer,https://github.com/pytorch/pytorch/issues/155321,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: package/deploy,categorize and decide next maintainer owner/action,2025-06-09T14:59:56Z
issue,137252,RuntimeError: Expression of type - cannot be used in a type expression: __torch__.transformers_modules.code-5p-110m-embedding.modeling_codet5p_embedding.___torch_mangle_1368.CodeT5pEmbeddingModel ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ <--- HERE,https://github.com/pytorch/pytorch/issues/137252,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2025-06-05T08:17:14Z
issue,111480,torch.jit.script persistently changes default from utf-8 to ascii,https://github.com/pytorch/pytorch/issues/111480,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2025-06-04T03:03:00Z
issue,155003,Internal Assertion Failure with Invalid Arguments in max_unpool2d under TorchScript,https://github.com/pytorch/pytorch/issues/155003,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2025-06-03T18:01:40Z
issue,155004,Internal Assertion Failure with Invalid Arguments in max_pool2d under TorchScript,https://github.com/pytorch/pytorch/issues/155004,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2025-06-03T18:01:30Z
issue,155007,Internal Assertion Failure with Invalid Arguments in max_pool1d under TorchScript,https://github.com/pytorch/pytorch/issues/155007,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2025-06-03T18:01:21Z
issue,155009,Internal Assertion Failure with Invalid Arguments in max_unpool1d under TorchScript,https://github.com/pytorch/pytorch/issues/155009,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2025-06-03T18:01:12Z
issue,154853,"Add node.meta[""stack_trace""] to make_fx",https://github.com/pytorch/pytorch/issues/154853,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: export, oncall: pt2",categorize and decide next maintainer owner/action,2025-06-02T17:01:30Z
issue,130309,"【Pytorch mobile android】torch jit forward fail,Calling torch.geqrf on a CPU tensor requires compiling PyTorch with LAPACK. Please use PyTorch built with LAPACK support.",https://github.com/pytorch/pytorch/issues/130309,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2025-05-31T20:11:48Z
issue,154607,Internal Assertion Failure with torch.cuda.Stream in @script_method on CPU-only Systems,https://github.com/pytorch/pytorch/issues/154607,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2025-05-29T13:00:01Z
issue,154478,`__jit_ignored_attributes__` is not respected in `torch.jit.trace_module`,https://github.com/pytorch/pytorch/issues/154478,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2025-05-28T13:39:42Z
issue,154502,`typing.get_type_hints` fails on TorchScript model after loading with `torch.jit.load`,https://github.com/pytorch/pytorch/issues/154502,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2025-05-28T13:34:47Z
issue,154407,[JIT] torch.jit.script raises an exception with view(dtype),https://github.com/pytorch/pytorch/issues/154407,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2025-05-28T08:09:26Z
issue,154423,`Segmentation fault` in `torch.jit.ignore`,https://github.com/pytorch/pytorch/issues/154423,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2025-05-27T18:57:18Z
issue,154323,"when pass a scale generated from a registered_buff of a nn.Module to fake_quantize_per_channel_affine, onnx runntime will raise Exception： “For per axis quantization, scale must be 1D tensor with”",https://github.com/pytorch/pytorch/issues/154323,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,categorize and decide next maintainer owner/action,2025-05-27T18:24:47Z
issue,154324,torch.jit.script gives false results for autograd if complex data types are involved,https://github.com/pytorch/pytorch/issues/154324,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: complex, oncall: jit",categorize and decide next maintainer owner/action,2025-05-25T13:05:16Z
issue,154101,"INTERNAL ASSERT FAILED at ""/pytorch/torch/csrc/jit/ir/ir.cpp"":1307",https://github.com/pytorch/pytorch/issues/154101,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2025-05-22T12:46:27Z
issue,153901,torch.profiler support for tf32,https://github.com/pytorch/pytorch/issues/153901,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: profiler,categorize and decide next maintainer owner/action,2025-05-20T15:41:54Z
issue,153931,TorchScript Fails to Handle type() for Any Input Type (Tensor/Non-Tensor),https://github.com/pytorch/pytorch/issues/153931,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2025-05-20T14:16:16Z
issue,150526,Consider context when tuning kernels for max-autotune to more accurately reflect the performance of real workloads,https://github.com/pytorch/pytorch/issues/150526,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: cpu inductor, oncall: pt2",categorize and decide next maintainer owner/action,2025-05-20T05:50:02Z
issue,153722,JIT Script doesn't handle LLONG_MIN correctly,https://github.com/pytorch/pytorch/issues/153722,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2025-05-16T14:34:24Z
issue,152921,[inductor][cpu] pytorch_CycleGAN_and_pix2pix AMP multiple thread performance regression in 2025-04-27 nightly release,https://github.com/pytorch/pytorch/issues/152921,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: cpu inductor, oncall: pt2",categorize and decide next maintainer owner/action,2025-05-15T08:48:47Z
issue,153315,[JIT] Compilation-induced discrepancy in F.instance_norm when passing input as running stats,https://github.com/pytorch/pytorch/issues/153315,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), oncall: jit",categorize and decide next maintainer owner/action,2025-05-13T17:51:18Z
issue,153358,torch.dequantize result inconsistent on CPU and GPU,https://github.com/pytorch/pytorch/issues/153358,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,categorize and decide next maintainer owner/action,2025-05-13T07:23:37Z
issue,152813,Mismatch in dynamic quantization performance for torchao and torch.quantization,https://github.com/pytorch/pytorch/issues/152813,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,categorize and decide next maintainer owner/action,2025-05-12T05:48:32Z
issue,153224,[JIT] Inconsistent results between eager and script modes for instance_norm with custom running stats,https://github.com/pytorch/pytorch/issues/153224,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2025-05-10T11:29:31Z
issue,153179,export: No error raised on data-dependent assertion.,https://github.com/pytorch/pytorch/issues/153179,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: dynamic shapes, module: error checking, oncall: export, oncall: pt2",categorize and decide next maintainer owner/action,2025-05-09T20:24:48Z
issue,152465,torch.export with dynamic shapes on Static Cache HF LLama model fails,https://github.com/pytorch/pytorch/issues/152465,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: export, oncall: pt2",categorize and decide next maintainer owner/action,2025-05-05T23:57:45Z
issue,152617,Pytorch Profiler crashes while using it with Pytorch Lightning module,https://github.com/pytorch/pytorch/issues/152617,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: profiler,categorize and decide next maintainer owner/action,2025-05-02T13:28:56Z
issue,152612,Enable AOTI for Metal inductor,https://github.com/pytorch/pytorch/issues/152612,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: aotinductor, module: mps, oncall: export, oncall: pt2",categorize and decide next maintainer owner/action,2025-05-01T22:58:02Z
issue,149500,[RFC] : Dynamically Quantized 8-bit Matrix Multiplication support,https://github.com/pytorch/pytorch/issues/149500,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,categorize and decide next maintainer owner/action,2025-05-01T19:47:48Z
issue,152283,Forward compatibility in torch.export,https://github.com/pytorch/pytorch/issues/152283,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: export, oncall: pt2",categorize and decide next maintainer owner/action,2025-04-28T23:03:31Z
issue,150607,`torch.jit.script` does not respect `torch.set_default_dtype`,https://github.com/pytorch/pytorch/issues/150607,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2025-04-28T02:05:26Z
issue,151746,[AotInductor][Export][Triton] how to export custom triton kernels when use torch.export.export,https://github.com/pytorch/pytorch/issues/151746,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: aotinductor, module: user triton, oncall: export, oncall: pt2",categorize and decide next maintainer owner/action,2025-04-25T23:11:04Z
issue,94974,Slow inference of torchscript model in C++,https://github.com/pytorch/pytorch/issues/94974,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2025-04-24T21:19:47Z
issue,148557,AOTI is OOM-ing when eager doesn't,https://github.com/pytorch/pytorch/issues/148557,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: aotinductor, oncall: export, oncall: pt2",categorize and decide next maintainer owner/action,2025-04-22T00:26:55Z
issue,151468,"Bug At `d\\dependencies\\libtorch\\include\\ATen\\core\\jit_type_base.h"":289`",https://github.com/pytorch/pytorch/issues/151468,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2025-04-17T13:50:36Z
issue,151428,TorchScript Model Saved on x86 Returns NaNs When Loaded on s390x,https://github.com/pytorch/pytorch/issues/151428,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2025-04-16T14:55:07Z
issue,110332,`torch.jit.load()` might unresponsive in IBM s390x when loading some certain torchscript saved by x86 machine.,https://github.com/pytorch/pytorch/issues/110332,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: power, oncall: jit",categorize and decide next maintainer owner/action,2025-04-15T23:28:04Z
issue,151233,Code Injection via `torch.jit.annotations.parse_type_line()`,https://github.com/pytorch/pytorch/issues/151233,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2025-04-15T01:36:30Z
issue,151016,[Dynamo][Export] Untraceable issues when exporting the Stable Diffusion 3.5 model,https://github.com/pytorch/pytorch/issues/151016,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: export, oncall: pt2",categorize and decide next maintainer owner/action,2025-04-14T18:43:55Z
issue,151165,Bug on running TorchScript on H100,https://github.com/pytorch/pytorch/issues/151165,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2025-04-13T23:01:42Z
issue,150577,[TorchScript] Enum scripting failures in python 3.11+,https://github.com/pytorch/pytorch/issues/150577,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2025-04-03T23:40:13Z
issue,92855,Profiler documentation doesn't mention some exports are mutually exclusive,https://github.com/pytorch/pytorch/issues/92855,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: docs, oncall: profiler",categorize and decide next maintainer owner/action,2025-04-03T17:58:38Z
issue,146914,TestSelectAlgorithmCPU.test_int8_woq_mm fails on Aarch64,https://github.com/pytorch/pytorch/issues/146914,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: arm, module: tests, oncall: cpu inductor, oncall: jit",categorize and decide next maintainer owner/action,2025-04-01T23:24:25Z
issue,150342,Torch trace doesn't respect @torch.jit.ignore on torch.nn.Module forward method,https://github.com/pytorch/pytorch/issues/150342,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2025-04-01T00:03:16Z
issue,137691,[inductor][cpu]float32 and AMP performance regression in 2024-09-30 nightly release,https://github.com/pytorch/pytorch/issues/137691,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: cpu inductor, oncall: pt2",categorize and decide next maintainer owner/action,2025-03-26T18:28:48Z
issue,149186,torch.fx.symbolic_trace failed on deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B,https://github.com/pytorch/pytorch/issues/149186,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: fx, oncall: export, oncall: pt2",categorize and decide next maintainer owner/action,2025-03-26T15:34:31Z
issue,134679,[CPU] jx_nest_base float32 both inductor and eager performance regression in 2024-08-26 nightly release,https://github.com/pytorch/pytorch/issues/134679,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: cpu inductor, oncall: pt2",categorize and decide next maintainer owner/action,2025-03-26T00:21:47Z
issue,149800,`Segmentation fault` in `torch.jit.jit_module_from_flatbuffer`,https://github.com/pytorch/pytorch/issues/149800,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2025-03-24T03:58:03Z
issue,149736,`INTERNAL ASSERT FAILED` in `torch.jit.script`,https://github.com/pytorch/pytorch/issues/149736,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2025-03-21T15:54:16Z
issue,149619,Unexpected return type from `torch.split` under `@torch.jit.script` decorator,https://github.com/pytorch/pytorch/issues/149619,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2025-03-20T19:29:44Z
issue,149501,Inductor produce significantly different inference results with the originl original model,https://github.com/pytorch/pytorch/issues/149501,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: cpu inductor, oncall: pt2",categorize and decide next maintainer owner/action,2025-03-20T08:53:28Z
issue,148946,export deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B failed,https://github.com/pytorch/pytorch/issues/148946,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: export, oncall: pt2",categorize and decide next maintainer owner/action,2025-03-19T22:06:41Z
issue,149391,Cannot torch.jit.script nn.MultiheadAttention when bias is set to False,https://github.com/pytorch/pytorch/issues/149391,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2025-03-18T21:26:30Z
issue,82550,`torch.cat` can break `torch.jit.ScriptModule` when in inference mode,https://github.com/pytorch/pytorch/issues/82550,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2025-03-18T13:15:32Z
issue,149251,false INTERNAL ASSERT FAILED in `torch.jit.trace`,https://github.com/pytorch/pytorch/issues/149251,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2025-03-17T15:27:11Z
issue,107568,RuntimeError: Unsupported value kind: Tensor while torch.jit.script nn.Module,https://github.com/pytorch/pytorch/issues/107568,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2025-03-17T10:07:28Z
issue,149223,Inconsistent results in using torch.jit.script API from API documentation.,https://github.com/pytorch/pytorch/issues/149223,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2025-03-14T23:20:43Z
issue,136618,[inductor][cpu]pyhpc_isoneutral_mixing singel thread AMP cpp wrapper performance regression in 2024-09-22 nightly release,https://github.com/pytorch/pytorch/issues/136618,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: cpu inductor, oncall: pt2",categorize and decide next maintainer owner/action,2025-03-11T01:34:13Z
issue,148382,[Inductor-CPU] Debug util request: fine-grained mechanism to disable out-of-template epilogues,https://github.com/pytorch/pytorch/issues/148382,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: cpu inductor, oncall: pt2",categorize and decide next maintainer owner/action,2025-03-10T19:54:13Z
issue,148071,[inductor][cpu] maml performance regression in 2025-02-24 nightly release,https://github.com/pytorch/pytorch/issues/148071,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: cpu inductor, oncall: pt2",categorize and decide next maintainer owner/action,2025-03-10T19:54:13Z
issue,147919,[inductor][cpu]AOT inductor AMP static shape default wrapper occupied almost 3x disk than before,https://github.com/pytorch/pytorch/issues/147919,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: cpu inductor, oncall: pt2",categorize and decide next maintainer owner/action,2025-03-10T19:54:12Z
issue,81552,Support Swift Package Manager (SPM) for iOS,https://github.com/pytorch/pytorch/issues/81552,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: mobile,categorize and decide next maintainer owner/action,2025-03-06T18:58:48Z
issue,148543,"When Ues torch::jit::Module in UE env, there are some structured exception caused by calling module.forward()",https://github.com/pytorch/pytorch/issues/148543,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2025-03-06T01:49:47Z
issue,148447,"Union type raise error when running python with argument ""-O"" for torch script.",https://github.com/pytorch/pytorch/issues/148447,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2025-03-06T01:14:12Z
issue,148171,Inference llama after Export PTQ,https://github.com/pytorch/pytorch/issues/148171,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,categorize and decide next maintainer owner/action,2025-03-03T18:23:06Z
issue,36061,[JIT] support list of nn.Module in torchscript,https://github.com/pytorch/pytorch/issues/36061,needs triage,47,triage signal present; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2025-03-03T08:27:33Z
issue,146824,Inductor-CPU might load (and store) fewer elements than the vector-width,https://github.com/pytorch/pytorch/issues/146824,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: cpu inductor, oncall: pt2",categorize and decide next maintainer owner/action,2025-03-01T23:26:12Z
issue,93134,Failed to Open libnvrtc-builtins.so.11.7,https://github.com/pytorch/pytorch/issues/93134,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: cuda, oncall: jit",categorize and decide next maintainer owner/action,2025-02-27T16:31:10Z
issue,147844,Python warnings are printed multiple times,https://github.com/pytorch/pytorch/issues/147844,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2025-02-25T19:37:54Z
issue,147739,torchvision export model error:: torchvision.models.detection.retinanet_resnet50_fpn_v2,https://github.com/pytorch/pytorch/issues/147739,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2025-02-25T15:28:44Z
issue,147722,Segmentation fault in `torch.ops.profiler._call_end_callbacks_on_jit_fut`,https://github.com/pytorch/pytorch/issues/147722,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: profiler,categorize and decide next maintainer owner/action,2025-02-25T15:25:18Z
issue,147604,RuntimeError when running profiler in a loop,https://github.com/pytorch/pytorch/issues/147604,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: profiler,categorize and decide next maintainer owner/action,2025-02-25T15:24:55Z
issue,146974,[inductor][cpu]detectron2_fcos_r_50_fpn multiple thread accuracy failure in 2025-02-10 nightly release,https://github.com/pytorch/pytorch/issues/146974,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: cpu inductor, oncall: pt2",categorize and decide next maintainer owner/action,2025-02-24T16:06:15Z
issue,146483,[ARM] Unit test failure on Neoverse-N1 - CPUReproTests.test_lstm_packed,https://github.com/pytorch/pytorch/issues/146483,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: arm, module: tests, oncall: cpu inductor, oncall: pt2",categorize and decide next maintainer owner/action,2025-02-24T16:05:51Z
issue,147444,NaN values originating from the normalize_weight_jit function during the backward pass,https://github.com/pytorch/pytorch/issues/147444,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2025-02-24T15:56:27Z
issue,145394,[EXPORT AOTI] `aoti_compile_and_package` custom_ops dependecies,https://github.com/pytorch/pytorch/issues/145394,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: aotinductor, oncall: export, oncall: pt2",categorize and decide next maintainer owner/action,2025-02-20T20:29:47Z
issue,146127,When I try to run rt-detr model in C++ libtorch i face the given error,https://github.com/pytorch/pytorch/issues/146127,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: windows, oncall: jit",categorize and decide next maintainer owner/action,2025-02-19T09:17:41Z
issue,147187,[torch.export] How to export a model with kv cache,https://github.com/pytorch/pytorch/issues/147187,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: export, oncall: pt2",categorize and decide next maintainer owner/action,2025-02-18T19:20:39Z
issue,144331,[Export] UserWarning: Attempted to insert a get_attr Node with no underlying reference in the owning GraphModule,https://github.com/pytorch/pytorch/issues/144331,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: export, oncall: pt2",categorize and decide next maintainer owner/action,2025-02-15T12:53:16Z
issue,146315,[export] dynamic_shapes with `Dim` fails when `DYNAMIC` succeeds,https://github.com/pytorch/pytorch/issues/146315,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: export, oncall: pt2",categorize and decide next maintainer owner/action,2025-02-15T03:01:55Z
issue,75960,"jit/_trace.py"", line 71, in _unique_state_dict filtered_dict[k] = v.detach() AttributeError: 'torch.dtype' object has no attribute 'detach'",https://github.com/pytorch/pytorch/issues/75960,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2025-02-14T09:36:04Z
issue,147112,Add quantized version of BatchNorm1d module,https://github.com/pytorch/pytorch/issues/147112,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,categorize and decide next maintainer owner/action,2025-02-13T18:20:01Z
issue,147055,INTERNAL ASSERT FAILED or SEGFAULT when JITting a function that can return different types,https://github.com/pytorch/pytorch/issues/147055,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2025-02-13T07:01:40Z
issue,146915,Unit test failure on Aarch64 - TestSelectAlgorithmCPU.test_linear_with_in_out_buffer,https://github.com/pytorch/pytorch/issues/146915,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: arm, module: tests, oncall: cpu inductor, oncall: pt2",categorize and decide next maintainer owner/action,2025-02-11T18:44:19Z
issue,146698,Possible memory leak when running test_model_exports_to_core_aten.py test,https://github.com/pytorch/pytorch/issues/146698,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: memory usage, oncall: export, oncall: pt2",categorize and decide next maintainer owner/action,2025-02-11T10:40:40Z
issue,146773,"INTERNAL ASSERT FAILED at ""/pytorch/aten/src/ATen/native/quantized/cpu/qsigmoid.cpp"":65, please report a bug to PyTorch.",https://github.com/pytorch/pytorch/issues/146773,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit, oncall: quantization",categorize and decide next maintainer owner/action,2025-02-10T21:46:01Z
issue,143289,RFC: Dynamically Quantized 4 bit matmul API and usage,https://github.com/pytorch/pytorch/issues/143289,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,categorize and decide next maintainer owner/action,2025-02-10T15:40:10Z
issue,146760,"[torch.jit] INTERNAL ASSERT FAILED at ""../aten/src/ATen/core/ivalue_inl.h"":1967, please report a bug to PyTorch.",https://github.com/pytorch/pytorch/issues/146760,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2025-02-09T20:30:55Z
issue,146758,"INTERNAL ASSERT FAILED at ""/pytorch/torch/csrc/autograd/functions/utils.h"":74, please report a bug to PyTorch",https://github.com/pytorch/pytorch/issues/146758,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2025-02-09T20:30:46Z
issue,146759,"INTERNAL ASSERT FAILED at ""../torch/csrc/jit/ir/alias_analysis.cpp"":617, please report a bug to PyTorch.",https://github.com/pytorch/pytorch/issues/146759,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2025-02-09T20:30:36Z
issue,146776,"INTERNAL ASSERT FAILED at ""/pytorch/torch/csrc/jit/testing/file_check.cpp"":607, please report a bug to PyTorch",https://github.com/pytorch/pytorch/issues/146776,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: testing, oncall: jit",categorize and decide next maintainer owner/action,2025-02-09T20:30:24Z
issue,146778,"[torch.jit] INTERNAL ASSERT FAILED at ""/pytorch/torch/csrc/jit/mobile/register_ops_common_utils.cpp"":34, please report a bug to PyTorch.",https://github.com/pytorch/pytorch/issues/146778,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2025-02-09T20:29:58Z
issue,146775,[torch.jit] Crash would be raised when using torch.jit.script,https://github.com/pytorch/pytorch/issues/146775,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2025-02-09T20:12:40Z
issue,146757,"[torch.jit.script] INTERNAL ASSERT FAILED at ""./torch/csrc/jit/ir/ir.h"":505, please report a bug to PyTorch",https://github.com/pytorch/pytorch/issues/146757,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2025-02-08T14:59:57Z
issue,114112,[JIT] - torch.script - 'Optional[Tensor]' object has no attribute or method 'size',https://github.com/pytorch/pytorch/issues/114112,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2025-02-07T10:50:58Z
issue,146576,Torch Dynamo Export Failed on RetinaNet from Torchvison,https://github.com/pytorch/pytorch/issues/146576,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: export, oncall: pt2",categorize and decide next maintainer owner/action,2025-02-06T17:28:58Z
issue,146368,DeepSeek: block quantization,https://github.com/pytorch/pytorch/issues/146368,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,categorize and decide next maintainer owner/action,2025-02-05T08:34:55Z
issue,145456,torch._dynamo.exc.Unsupported: Graph break due to unsupported builtin torch._C._dynamo.eval_frame.set_eval_frame.,https://github.com/pytorch/pytorch/issues/145456,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: export, oncall: pt2",categorize and decide next maintainer owner/action,2025-02-05T01:20:34Z
issue,145711,[Inductor-CPU] Templated codegened kernel name may not correctly reflect epilogue fusions,https://github.com/pytorch/pytorch/issues/145711,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: cpu inductor, oncall: pt2",categorize and decide next maintainer owner/action,2025-02-04T22:33:11Z
issue,145709,[Inductor-CPU] Templated codegened kernel names are ambiguous in PyTorch profiler results,https://github.com/pytorch/pytorch/issues/145709,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: cpu inductor, oncall: pt2",categorize and decide next maintainer owner/action,2025-02-04T22:33:10Z
issue,145743,TorchScript run_method fails from 2.5.0 onward on Ubuntu,https://github.com/pytorch/pytorch/issues/145743,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: deadlock, oncall: jit",categorize and decide next maintainer owner/action,2025-02-03T10:31:47Z
issue,145995,torch.jit.trace does not move tensors to GPU.,https://github.com/pytorch/pytorch/issues/145995,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2025-01-30T01:12:10Z
issue,129889,custom gradient for int8,https://github.com/pytorch/pytorch/issues/129889,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,categorize and decide next maintainer owner/action,2025-01-22T07:55:12Z
issue,68559,LibTorch -> TorchScript -> PyTorch (Python) fails with `AttributeError: 'RecursiveScriptModule' object has no attribute 'forward'`,https://github.com/pytorch/pytorch/issues/68559,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2025-01-20T22:21:17Z
issue,144778,`torch.profiler.record_function` doesn't register kernels from `backward` function,https://github.com/pytorch/pytorch/issues/144778,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: profiler,categorize and decide next maintainer owner/action,2025-01-16T03:56:16Z
issue,144613,The label marked by torch.profiler.profile.record_function() appears twice in the output,https://github.com/pytorch/pytorch/issues/144613,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: profiler,categorize and decide next maintainer owner/action,2025-01-13T17:27:47Z
issue,127571,[export] Cannot mutate tensors with frozen storage,https://github.com/pytorch/pytorch/issues/127571,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: export,categorize and decide next maintainer owner/action,2025-01-08T22:04:05Z
issue,112372,[Tracer] RuntimeError: _Map_base::at when tracing using autograd.Function,https://github.com/pytorch/pytorch/issues/112372,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2025-01-08T02:02:58Z
issue,144148,"Unable to parse and visualize ""trace view"" with tensorboard",https://github.com/pytorch/pytorch/issues/144148,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: tensorboard, oncall: profiler",categorize and decide next maintainer owner/action,2025-01-06T15:40:24Z
issue,76774,Support Positional-only Arguments in JIT,https://github.com/pytorch/pytorch/issues/76774,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2025-01-03T20:06:13Z
issue,144039,[Mac/M1] torch.compile() -- expm1 returns an inaccurate result compared to the interpreted version,https://github.com/pytorch/pytorch/issues/144039,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: cpu inductor, oncall: pt2",categorize and decide next maintainer owner/action,2025-01-03T04:20:48Z
issue,141854,RuntimeError: quantized::conv2d_relu (ONEDNN): data type of input should be QUint8.,https://github.com/pytorch/pytorch/issues/141854,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: onnx, oncall: quantization",categorize and decide next maintainer owner/action,2024-12-30T21:53:10Z
issue,143886,[RFC] Identifying dynamic int8 symmetric vs asymmetric quantization of activation/input in Inductor-CPU,https://github.com/pytorch/pytorch/issues/143886,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: cpu inductor, oncall: pt2",categorize and decide next maintainer owner/action,2024-12-27T17:21:50Z
issue,143854,Can't script a tensorrt model,https://github.com/pytorch/pytorch/issues/143854,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-12-27T02:59:11Z
issue,135689,[inductor][cpu]GPTNeoForSequenceClassification AMP single/multiple thread static/dynamic shape default/cpp accuracy failure,https://github.com/pytorch/pytorch/issues/135689,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: cpu inductor, oncall: pt2",categorize and decide next maintainer owner/action,2024-12-27T02:35:03Z
issue,143831,"After pth is converted into ptl, the prediction result is very different from pth",https://github.com/pytorch/pytorch/issues/143831,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-12-26T15:42:23Z
issue,143857,`@torch.jit.script` causes `pytest-cov` to miss function body,https://github.com/pytorch/pytorch/issues/143857,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-12-26T14:29:55Z
issue,143565,Failed to load TorchScript of SSDLite in android,https://github.com/pytorch/pytorch/issues/143565,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: mobile,categorize and decide next maintainer owner/action,2024-12-20T05:15:26Z
issue,132009,2:4 semi structured sparsity matmul not supported on hopper,https://github.com/pytorch/pytorch/issues/132009,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: cuda, oncall: quantization",categorize and decide next maintainer owner/action,2024-12-19T09:48:39Z
issue,143379,Running decompositions over torch.export program adds one more output to the graph result,https://github.com/pytorch/pytorch/issues/143379,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: export, oncall: pt2",categorize and decide next maintainer owner/action,2024-12-17T21:38:08Z
issue,140689,[inductor][cpu]opacus_cifar10 and functorch_dp_cifar10 AMP performance regression in 2024-11-11 nightly release,https://github.com/pytorch/pytorch/issues/140689,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: cpu inductor, oncall: pt2",categorize and decide next maintainer owner/action,2024-12-17T01:19:15Z
issue,143258,Proper support for optionals in TorchScript,https://github.com/pytorch/pytorch/issues/143258,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-12-16T16:19:16Z
issue,81912,Unknown builtin op: aten::broadcast_shapes,https://github.com/pytorch/pytorch/issues/81912,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-12-16T08:04:07Z
issue,127521,Modeling ViT does not support quantized models,https://github.com/pytorch/pytorch/issues/127521,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,categorize and decide next maintainer owner/action,2024-12-08T18:23:02Z
issue,141893,Forward from pytorch/executorch: export error with Macbook pro M4,https://github.com/pytorch/pytorch/issues/141893,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: export,categorize and decide next maintainer owner/action,2024-12-07T01:13:45Z
issue,124190,Re-enable skipped 3.12 dynamo-wrapped profiler tests,https://github.com/pytorch/pytorch/issues/124190,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: python version, oncall: profiler",categorize and decide next maintainer owner/action,2024-12-06T21:30:51Z
issue,142170,[3.13] Fix failing torch.package dynamo-wrapped test,https://github.com/pytorch/pytorch/issues/142170,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: python frontend, oncall: package/deploy",categorize and decide next maintainer owner/action,2024-12-05T22:00:33Z
issue,141996,[export] Move pytree serialization into export serializer,https://github.com/pytorch/pytorch/issues/141996,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: export, oncall: pt2",categorize and decide next maintainer owner/action,2024-12-03T21:19:51Z
issue,141855,Can not convert Llama-3.2-11B-Vision-Instruct by torch.jit.script / torch.jit.trace,https://github.com/pytorch/pytorch/issues/141855,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-12-02T16:21:36Z
issue,115511,[Export] aten::_transformer_encoder_layer_fwd is not core aten and has no decomposition.,https://github.com/pytorch/pytorch/issues/115511,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: nn, oncall: export",categorize and decide next maintainer owner/action,2024-11-27T19:55:38Z
issue,140805,[AOTI] Test case `test_aot_inductor.py::test_multi_device_cpu` is Running GPU AOT Inductor in CPU container runner.,https://github.com/pytorch/pytorch/issues/140805,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: aotinductor, oncall: export, oncall: pt2",categorize and decide next maintainer owner/action,2024-11-25T07:12:45Z
issue,92041,jit testing fails on 3.11 debug build,https://github.com/pytorch/pytorch/issues/92041,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-11-23T20:23:55Z
issue,141116,How to fuse batchnorm to conv2d in the graph exported by torch.export,https://github.com/pytorch/pytorch/issues/141116,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: export, oncall: pt2",categorize and decide next maintainer owner/action,2024-11-20T19:06:46Z
issue,111142,'torch._C.Node' object has no attribute 'cs',https://github.com/pytorch/pytorch/issues/111142,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-11-20T15:02:09Z
issue,137202,"Met error ""we don't have an op for aten::full but it isn't a special case."" when running microsoft/Florence-2-base",https://github.com/pytorch/pytorch/issues/137202,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-11-20T02:24:56Z
issue,81855,Enable jit error when using FSDP,https://github.com/pytorch/pytorch/issues/81855,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-11-19T21:22:44Z
issue,121581,The results of PyTorch and TorchScript models are inconsistent.,https://github.com/pytorch/pytorch/issues/121581,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-11-19T14:27:21Z
issue,140468,[TorchScript] Failure if you script a wrapper module and then an interface-implementing submodule.,https://github.com/pytorch/pytorch/issues/140468,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-11-18T18:06:54Z
issue,132240,torch.nn.functional.interpolate completely broken with torch.jit.script and torch.fx,https://github.com/pytorch/pytorch/issues/132240,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-11-16T13:03:17Z
issue,140684,Segmentation fault (core dumped) in `torch.autograd.profiler.profile`,https://github.com/pytorch/pytorch/issues/140684,needs triage,47,triage signal present; old update age or low-urgency/stale label; component/oncall labels: oncall: profiler,categorize and decide next maintainer owner/action,2024-11-15T17:28:10Z
issue,140702,`Tensor.prod` gives a jit warning,https://github.com/pytorch/pytorch/issues/140702,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-11-14T11:35:18Z
issue,140539,torch.jit.trace doesn't trace all operations completely (no data-dependent control flow),https://github.com/pytorch/pytorch/issues/140539,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-11-14T03:07:29Z
issue,68893,NCCL Network is unreachable / Connection refused when initializing DDP,https://github.com/pytorch/pytorch/issues/68893,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: nccl, oncall: distributed",categorize and decide next maintainer owner/action,2024-11-13T09:18:50Z
issue,76026,Jit torchscript for prediction is missing 'forward' when using forward hooks,https://github.com/pytorch/pytorch/issues/76026,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-11-11T13:32:33Z
issue,135569,[inductor][cpu]AOT inductor performance regression and accuracy failure in 2024-09-04 nightly release,https://github.com/pytorch/pytorch/issues/135569,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: cpu inductor, oncall: pt2",categorize and decide next maintainer owner/action,2024-11-11T07:53:24Z
issue,66046,PicklingError when saving a ddp module with torch.save(),https://github.com/pytorch/pytorch/issues/66046,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,categorize and decide next maintainer owner/action,2024-11-09T13:29:58Z
issue,140101,Inconsistent gradients in torch script function,https://github.com/pytorch/pytorch/issues/140101,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-11-08T17:11:11Z
issue,120166,QAT using pytorch-quantization cause accuracy loss after exporting to onnx,https://github.com/pytorch/pytorch/issues/120166,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: onnx, oncall: quantization",categorize and decide next maintainer owner/action,2024-11-08T10:17:10Z
issue,139470,serialization of PT2E model impacts torch.fx.passes.utils.source_matcher_utils.get_source_partitions,https://github.com/pytorch/pytorch/issues/139470,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: export,categorize and decide next maintainer owner/action,2024-11-05T19:29:38Z
issue,45398,build_android.sh,https://github.com/pytorch/pytorch/issues/45398,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: mobile,categorize and decide next maintainer owner/action,2024-11-05T09:20:03Z
issue,96726,Memory not release after jit.trace/freeze,https://github.com/pytorch/pytorch/issues/96726,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-11-05T08:59:43Z
issue,139556,[JIT] ModuleList getitem error when saving and loading with torchscript,https://github.com/pytorch/pytorch/issues/139556,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-11-04T14:02:30Z
issue,139464,Maybe there is a precision issue with torch.quantize_per_channel?,https://github.com/pytorch/pytorch/issues/139464,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,categorize and decide next maintainer owner/action,2024-11-01T19:49:27Z
issue,139297,`torch.package` warning -- `TypedStorage` is deprecated,https://github.com/pytorch/pytorch/issues/139297,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: package/deploy,categorize and decide next maintainer owner/action,2024-10-31T21:44:35Z
issue,139063,Quantization API examples in the docs are outdated,https://github.com/pytorch/pytorch/issues/139063,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: docs, oncall: quantization",categorize and decide next maintainer owner/action,2024-10-28T21:20:10Z
issue,138973,"When using torch.jit.trace for inference, the element of the shape attribute of tensor and the return value of numel() are tensor",https://github.com/pytorch/pytorch/issues/138973,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-10-28T14:27:24Z
issue,70544,Tensor loses `bool` method during scripting,https://github.com/pytorch/pytorch/issues/70544,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-10-27T10:28:18Z
issue,138820,RuntimeError: HalfTensor is not supported,https://github.com/pytorch/pytorch/issues/138820,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,categorize and decide next maintainer owner/action,2024-10-25T15:07:34Z
issue,138341,`torch.package` bug,https://github.com/pytorch/pytorch/issues/138341,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: package/deploy,categorize and decide next maintainer owner/action,2024-10-18T20:12:17Z
issue,138221,auto_functionalized_v2 doesn't support export,https://github.com/pytorch/pytorch/issues/138221,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: export, oncall: pt2",categorize and decide next maintainer owner/action,2024-10-18T19:58:11Z
issue,138193,[profiler] are CUDA events mutually exclusive with _ExperimentalConfig.profiler_metrics?,https://github.com/pytorch/pytorch/issues/138193,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: profiler,categorize and decide next maintainer owner/action,2024-10-18T02:22:43Z
issue,70450,torch.jit.optimized_execution is not mentioned anywhere in the docs,https://github.com/pytorch/pytorch/issues/70450,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-10-15T08:39:52Z
issue,137561,Segmentation fault would be triggered when using torch.jit.trace and torch.jit._fork,https://github.com/pytorch/pytorch/issues/137561,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-10-09T13:30:42Z
issue,91000,node.stack_trace does not handle escaping correctly,https://github.com/pytorch/pytorch/issues/91000,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: fx,categorize and decide next maintainer owner/action,2024-10-04T15:29:05Z
issue,131452,"JIT load traced model with ""Expression of type"" error",https://github.com/pytorch/pytorch/issues/131452,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-10-03T05:05:29Z
issue,69250,TorchScript incomptaible with non-persistent buffers,https://github.com/pytorch/pytorch/issues/69250,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-10-01T14:27:11Z
issue,136829,false INTERNAL ASSERT FAILED in `torch.jit.set_fusion_strategy`,https://github.com/pytorch/pytorch/issues/136829,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-09-27T14:04:29Z
issue,136828,false INTERNAL ASSERT FAILED in `torch.empty`/`torch.ones`,https://github.com/pytorch/pytorch/issues/136828,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,categorize and decide next maintainer owner/action,2024-09-27T14:04:19Z
issue,136654,[TorchScript] typing_extensions.deprecated doesn't work,https://github.com/pytorch/pytorch/issues/136654,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-09-25T18:27:20Z
issue,134971,"[torch.jit] INTERNAL ASSERT FAILED at ""../torch/csrc/jit/passes/symbolic_shape_analysis.cpp"":741, please report a bug to PyTorch. input_shapes size: 1 n inputs size: 2",https://github.com/pytorch/pytorch/issues/134971,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-09-23T23:01:39Z
issue,135690,[inductor][cpu] hf_BigBird float32 multiple threads performance regression in 2024-09-08 nightly release,https://github.com/pytorch/pytorch/issues/135690,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: cpu inductor, oncall: pt2",categorize and decide next maintainer owner/action,2024-09-23T22:59:06Z
issue,69078,RuntimeError: Global alloc not supported yet in TorchScript,https://github.com/pytorch/pytorch/issues/69078,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-09-20T17:56:18Z
issue,134924,`TorchScriptTesting._PickleTester` caused a segmentation fault with empty input lists in a jit traced function,https://github.com/pytorch/pytorch/issues/134924,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-09-20T16:16:39Z
issue,134923,A crash would be raised when using `._StackString` with empty lists,https://github.com/pytorch/pytorch/issues/134923,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-09-20T16:16:31Z
issue,134922,[torch.jit.script] An INTERNAL ASSERT error will be raised when permuting the inputs of a jit scripted graph with mismatched indices.,https://github.com/pytorch/pytorch/issues/134922,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-09-20T16:16:13Z
issue,134903,[torch.jit] Crash would be raised when using TorchScript modules,https://github.com/pytorch/pytorch/issues/134903,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-09-20T16:15:47Z
issue,134898,Segmentation fault would be triggered when using `torch.jit.script` and `torch._C._jit_shape_compute_graph_for_node`.,https://github.com/pytorch/pytorch/issues/134898,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-09-20T16:15:34Z
issue,134937,"[torch.jit.script] INTERNAL ASSERT FAILED at ""../aten/src/ATen/core/jit_type_base.h"":556, please report a bug to PyTorch",https://github.com/pytorch/pytorch/issues/134937,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-09-20T16:10:16Z
issue,130229,"RuntimeError: 0 INTERNAL ASSERT FAILED at ""../torch/csrc/jit/ir/alias_analysis.cpp"":615, please report a bug to PyTorch. We don't have an op for aten::full but it isn't a special case. Argument types: int[], bool, int, NoneType, Device, bool,",https://github.com/pytorch/pytorch/issues/130229,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-09-18T22:30:56Z
issue,136256,TorchScript raises an OSError when using frozen dataclass,https://github.com/pytorch/pytorch/issues/136256,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-09-18T16:48:23Z
issue,68866,wait() does not block the default stream for NCCL's asynchronous P2P operations,https://github.com/pytorch/pytorch/issues/68866,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: nccl, oncall: distributed",categorize and decide next maintainer owner/action,2024-09-18T04:25:18Z
issue,58298,"[JIT] Can't handle creating new objects in compiled methods: ""Class does not have an __init__ function defined""",https://github.com/pytorch/pytorch/issues/58298,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-09-17T02:05:57Z
issue,136094,Quantized model is way slower than regular model.,https://github.com/pytorch/pytorch/issues/136094,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,categorize and decide next maintainer owner/action,2024-09-16T15:52:35Z
issue,136127,A division-by-zero operation in a scripted function can cause the core dumped,https://github.com/pytorch/pytorch/issues/136127,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-09-16T15:46:57Z
issue,122813,cpu performance for int4mm kernels,https://github.com/pytorch/pytorch/issues/122813,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: performance, oncall: cpu inductor, oncall: pt2",categorize and decide next maintainer owner/action,2024-09-15T16:34:49Z
issue,133444,Collect tensor shapes only when using record_shapes in profiler,https://github.com/pytorch/pytorch/issues/133444,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: profiler,categorize and decide next maintainer owner/action,2024-09-13T16:28:54Z
issue,135580,Inconsistent results between libtorch and pytorch for model with shared weights,https://github.com/pytorch/pytorch/issues/135580,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-09-11T19:26:24Z
issue,62474,Distributed: RuntimeError: one of the variables needed for gradient computation has been modified by an inplace operation: [torch.cuda.FloatTensor [2048]] is at version 4; expected version 3 instead,https://github.com/pytorch/pytorch/issues/62474,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,categorize and decide next maintainer owner/action,2024-09-11T08:49:51Z
issue,134220,[Operator Benchmark] [CPU]OP: weight_norm_sparsifier_step functionality failed with parts of long config shapes,https://github.com/pytorch/pytorch/issues/134220,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,categorize and decide next maintainer owner/action,2024-09-10T03:21:28Z
issue,93943,"`torch.jit.trace` memory usage increase although forward is constant, and gets much slower than forward with model depth increase",https://github.com/pytorch/pytorch/issues/93943,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-09-09T23:14:40Z
issue,82099,UserWarning: operator() sees varying value in profiling,https://github.com/pytorch/pytorch/issues/82099,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-09-09T23:14:39Z
issue,134303,[FakeTensor] Error: SymIntArrayRef expected to contain only concrete integers in UpSample2D block,https://github.com/pytorch/pytorch/issues/134303,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: onnx, oncall: export",categorize and decide next maintainer owner/action,2024-09-04T14:34:44Z
issue,134918,"DYNAMIC_TRT_MODEL_CONVERSION 0 INTERNAL ASSERT FAILED or, comfyui made me come here",https://github.com/pytorch/pytorch/issues/134918,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-09-03T19:15:39Z
issue,135003,"[BUG]: torch.profiler is blocked in ""_disable_profiler()"" when profiling cuda graph on single GPU",https://github.com/pytorch/pytorch/issues/135003,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: profiler,categorize and decide next maintainer owner/action,2024-09-03T19:06:46Z
issue,60252,UserWarning: Failed to initialize NumPy: No module named 'numpy',https://github.com/pytorch/pytorch/issues/60252,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: package/deploy,categorize and decide next maintainer owner/action,2024-09-03T15:05:55Z
issue,134849,torch::jit::load invalid_path error,https://github.com/pytorch/pytorch/issues/134849,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-08-30T15:18:54Z
issue,40870,[JIT] cannot statically infer the expected size of a list in this context for positional arguments to Tensor.view(),https://github.com/pytorch/pytorch/issues/40870,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-08-30T03:45:52Z
issue,134562,"ScriptModel Save ERROR, when I convert part of the model to TRT",https://github.com/pytorch/pytorch/issues/134562,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-08-27T14:20:53Z
issue,134559,[JIT][foreach_add] Arguments for call are not valid,https://github.com/pytorch/pytorch/issues/134559,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-08-27T14:20:46Z
issue,121191,Export-based GraphModules fail with deepcopy,https://github.com/pytorch/pytorch/issues/121191,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: export,categorize and decide next maintainer owner/action,2024-08-27T11:18:28Z
issue,133552,fx qat symmetric config zero_point not zero,https://github.com/pytorch/pytorch/issues/133552,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,categorize and decide next maintainer owner/action,2024-08-23T03:02:11Z
issue,133255,torch.export ExportedProgram from GPU cannot be loaded on a CPU only machine,https://github.com/pytorch/pytorch/issues/133255,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: export,categorize and decide next maintainer owner/action,2024-08-23T00:09:46Z
issue,123168,TorchScript: Union cannot be used as a tuple,https://github.com/pytorch/pytorch/issues/123168,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-08-21T15:24:45Z
issue,133678,[torch.jit.trace] error encountered when tracing model with weight clamping during forward,https://github.com/pytorch/pytorch/issues/133678,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-08-21T07:13:42Z
issue,133795,"RuntimeError: createStatus == pytorch_qnnp_status_success INTERNAL ASSERT FAILED at ""../aten/src/ATen/native/quantized/cpu/BinaryOps.cpp"":204, please report a bug to PyTorch. failed to create QNNPACK Add operator",https://github.com/pytorch/pytorch/issues/133795,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit, oncall: quantization",categorize and decide next maintainer owner/action,2024-08-20T14:30:36Z
issue,133847,"[Bug Report] when using % in torch.jit.script function, the output conforms to the form of C++ % operation, rather than Python % operation.",https://github.com/pytorch/pytorch/issues/133847,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-08-19T15:10:06Z
issue,133840,Pytorch Mobile - Different output on Android and PC with the same model,https://github.com/pytorch/pytorch/issues/133840,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit, oncall: mobile",categorize and decide next maintainer owner/action,2024-08-19T15:01:15Z
issue,132930,Treating SymNode compute more fungibly for the purpose of export flattener (or not?),https://github.com/pytorch/pytorch/issues/132930,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: dynamic shapes, oncall: export",categorize and decide next maintainer owner/action,2024-08-16T22:05:13Z
issue,133675,"[torch.jit] RuntimeError: false INTERNAL ASSERT FAILED at ""../torch/csrc/jit/ir/node_hashing.cpp"":148, please report a bug to PyTorch.",https://github.com/pytorch/pytorch/issues/133675,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-08-16T20:52:31Z
issue,133557,torch quantize error,https://github.com/pytorch/pytorch/issues/133557,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,categorize and decide next maintainer owner/action,2024-08-16T20:43:48Z
issue,133537,`torch.jit.script` fails on a simple function,https://github.com/pytorch/pytorch/issues/133537,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-08-16T13:26:14Z
issue,122934,Cannot export DETR model with torch.export,https://github.com/pytorch/pytorch/issues/122934,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: export,categorize and decide next maintainer owner/action,2024-08-15T18:18:01Z
issue,132738,Profiler getProfilerConfig interface can not support Ascend npu platform,https://github.com/pytorch/pytorch/issues/132738,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: profiler,categorize and decide next maintainer owner/action,2024-08-14T18:25:42Z
issue,116662,test/test_vulkan.py fails with vulkan_prepack::conv2d_clamp_prepack not found,https://github.com/pytorch/pytorch/issues/116662,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: vulkan, oncall: mobile",categorize and decide next maintainer owner/action,2024-08-14T15:21:29Z
issue,51328,Support torch::jit::script::Module converting to torch::nn::Module.,https://github.com/pytorch/pytorch/issues/51328,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-08-13T11:41:12Z
issue,132559,How to fix tensor.numpy() not supported for torch.export with strict=False,https://github.com/pytorch/pytorch/issues/132559,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: functionalization, module: numpy, oncall: export",categorize and decide next maintainer owner/action,2024-08-06T18:42:05Z
issue,132459,"[export] Implement a common ""convert dtype to int"" function",https://github.com/pytorch/pytorch/issues/132459,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: export,categorize and decide next maintainer owner/action,2024-08-06T16:39:55Z
issue,132385,crnn model from pth to torchscript problem,https://github.com/pytorch/pytorch/issues/132385,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-08-01T17:37:06Z
issue,132315,Support Enum for dictionaries in TorchScript,https://github.com/pytorch/pytorch/issues/132315,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-07-31T19:33:35Z
issue,128292,put profiling results from different ranks into one webpage,https://github.com/pytorch/pytorch/issues/128292,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: profiler,categorize and decide next maintainer owner/action,2024-07-30T17:19:28Z
issue,132029,torch.QUInt4x2Storage: RuntimeError cuda_dispatch_ptr INTERNAL ASSERT FAILED on torch 2.5.0.dev20240708+cu121,https://github.com/pytorch/pytorch/issues/132029,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,categorize and decide next maintainer owner/action,2024-07-29T16:26:38Z
issue,88617,"The libtorch tests Simplify.{SimplifySymbolicMinMax,SimplifyNestedMax,SimplifyNestedMin} fail on Apple Silicon",https://github.com/pytorch/pytorch/issues/88617,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: arm, module: macos, oncall: jit",categorize and decide next maintainer owner/action,2024-07-25T11:28:36Z
issue,18476,[CPP] Allow binding config structs into the Python front end,https://github.com/pytorch/pytorch/issues/18476,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: cpp, oncall: jit",categorize and decide next maintainer owner/action,2024-07-24T20:50:02Z
issue,130726,TorchDispatchMode fails on jit.trace,https://github.com/pytorch/pytorch/issues/130726,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-07-23T23:00:32Z
issue,131022,Code bug in call_prepare_scriptable_func_impl function.,https://github.com/pytorch/pytorch/issues/131022,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-07-18T19:55:06Z
issue,131055,PT2E XNNPACKQuantizer has unused operator_type_config and set_operator_type,https://github.com/pytorch/pytorch/issues/131055,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,categorize and decide next maintainer owner/action,2024-07-18T19:46:02Z
issue,130820,[Export] Mutating module attribute error during export on YOLOX,https://github.com/pytorch/pytorch/issues/130820,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: export,categorize and decide next maintainer owner/action,2024-07-17T06:48:47Z
issue,130735,"Problem with combination of jacref, vjp, tanh in torchscript",https://github.com/pytorch/pytorch/issues/130735,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-07-15T23:46:45Z
issue,67834,Model tracing not working,https://github.com/pytorch/pytorch/issues/67834,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-07-15T09:24:30Z
issue,130555,torch.export.export re-export preserving metadata for output node,https://github.com/pytorch/pytorch/issues/130555,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: export,categorize and decide next maintainer owner/action,2024-07-11T17:54:20Z
issue,104405,Detailed error: Tensor-likes are not close! When use torch.jit.trace,https://github.com/pytorch/pytorch/issues/104405,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-07-07T01:18:44Z
issue,76486,[Tracer] RuntimeError: _Map_base::at when tracing fake quantization,https://github.com/pytorch/pytorch/issues/76486,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-07-01T06:03:35Z
issue,48387,"Error occurred when trying to call ""torch::jit::load""method for loading "".pt"" file on IOS",https://github.com/pytorch/pytorch/issues/48387,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit, oncall: mobile",categorize and decide next maintainer owner/action,2024-06-28T23:44:50Z
issue,129633,pytorch_android-2.1.0.aar/jni/arm64-v8a/libpytorch_jni.so: invalid local symbol '__bss_end__' in global part of symbol table,https://github.com/pytorch/pytorch/issues/129633,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: mobile,categorize and decide next maintainer owner/action,2024-06-27T14:48:24Z
issue,129347,Error when loading a module file by calling torch::jit::load(..),https://github.com/pytorch/pytorch/issues/129347,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-06-24T15:47:21Z
issue,125272,Profiler Kineto Config: Warning: Failed to recover relationship between all profiler and kineto events: 51361 vs. 0 reassociated. (function reassociate),https://github.com/pytorch/pytorch/issues/125272,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: profiler,categorize and decide next maintainer owner/action,2024-06-24T15:22:35Z
issue,129282,Support for torch.Generator with JIT,https://github.com/pytorch/pytorch/issues/129282,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-06-23T16:50:57Z
issue,89064,UserWarning: The TorchScript type system doesn't support instance-level annotations on empty non-base types in `__init__`.,https://github.com/pytorch/pytorch/issues/89064,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-06-22T15:56:43Z
issue,123499,Support Exceptions in Pytorch Export,https://github.com/pytorch/pytorch/issues/123499,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: export",categorize and decide next maintainer owner/action,2024-06-17T23:48:10Z
issue,128491,Improve type inference in torchscript,https://github.com/pytorch/pytorch/issues/128491,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-06-13T15:02:40Z
issue,128585,TorchScript JIT UnicodeDecodeError: ‘utf-8’ codec can’t decode byte 0xe0 in position 0: unexpected end of data,https://github.com/pytorch/pytorch/issues/128585,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-06-13T15:02:01Z
issue,128496,torch.jit.script not work for ParameterDict.items,https://github.com/pytorch/pytorch/issues/128496,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-06-13T15:02:00Z
issue,128385,[export] Llama2 export with dynamic shapes fails using Torch 2.4 nightly,https://github.com/pytorch/pytorch/issues/128385,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: export,categorize and decide next maintainer owner/action,2024-06-12T18:45:55Z
issue,42207,torchscript does not support unicode strings,https://github.com/pytorch/pytorch/issues/42207,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-06-12T11:26:30Z
issue,126266,Quantization occurs with RuntimeError: `zero_point` must be between `quant_min` and `quant_max`.,https://github.com/pytorch/pytorch/issues/126266,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,categorize and decide next maintainer owner/action,2024-06-12T06:47:18Z
issue,127283,[BUG][JIT] `torch.jit.script` is not compatible with `DeprecationWarning` and `FutureWarning`,https://github.com/pytorch/pytorch/issues/127283,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-06-12T01:14:55Z
issue,116795,multiple output support of jit mobile model?,https://github.com/pytorch/pytorch/issues/116795,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: mobile,categorize and decide next maintainer owner/action,2024-06-10T19:33:04Z
issue,118069,Quantising a multimodal large language PyTorch model to run on Android,https://github.com/pytorch/pytorch/issues/118069,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: mobile,categorize and decide next maintainer owner/action,2024-06-10T19:31:00Z
issue,120976,RuntimeError: xnnpack::convolution not available！,https://github.com/pytorch/pytorch/issues/120976,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: mobile,categorize and decide next maintainer owner/action,2024-06-10T19:28:45Z
issue,128261,JIT script can cause unneeded `requires_grad` inside `no_grad` blocks,https://github.com/pytorch/pytorch/issues/128261,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-06-10T16:29:38Z
issue,128125,optional memory leak with torch.jit.script on models with fft,https://github.com/pytorch/pytorch/issues/128125,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-06-10T15:41:39Z
issue,126127,torch.export.export() fails with dynamic shapes when more than one shape is dynamic,https://github.com/pytorch/pytorch/issues/126127,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: dynamic shapes, oncall: export",categorize and decide next maintainer owner/action,2024-06-05T21:12:01Z
issue,67983,torch.jit.export does not working,https://github.com/pytorch/pytorch/issues/67983,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-06-05T20:20:54Z
issue,127357,torch.export.export() throws out an error when dealing weighttying model.,https://github.com/pytorch/pytorch/issues/127357,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: export,categorize and decide next maintainer owner/action,2024-06-05T15:25:57Z
issue,128019,"RuntimeError: r INTERNAL ASSERT FAILED at ""../aten/src/ATen/core/jit_type_base.h"":544 when using torch.jit.trace and torch._C._jit_interpret_graph",https://github.com/pytorch/pytorch/issues/128019,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-06-05T14:49:23Z
issue,46948,Record shaping assertions and use them for tracing / scripting optimization and codegen,https://github.com/pytorch/pytorch/issues/46948,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: codegen, oncall: jit",categorize and decide next maintainer owner/action,2024-06-04T19:35:57Z
issue,127679,RuntimeError: `jit.freeze` fails to find externally assigned attributes,https://github.com/pytorch/pytorch/issues/127679,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-06-03T17:52:32Z
issue,127643,[export] Errors out when unflattening TorchTitan,https://github.com/pytorch/pytorch/issues/127643,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: export,categorize and decide next maintainer owner/action,2024-06-03T17:46:32Z
issue,124697,[INDUCTOR] [CPU] [GPT-FAST-MOE] large perf regression with coordinate_descent_tuning disabled,https://github.com/pytorch/pytorch/issues/124697,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: cpu inductor, oncall: pt2",categorize and decide next maintainer owner/action,2024-06-03T02:20:40Z
issue,127062,T5 -small Dynamic quantization in graviton3,https://github.com/pytorch/pytorch/issues/127062,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,categorize and decide next maintainer owner/action,2024-05-30T09:36:26Z
issue,127076,[PT2E Quantization] `prepare_pt2e` produces inconsistent data types for primitive int,https://github.com/pytorch/pytorch/issues/127076,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,categorize and decide next maintainer owner/action,2024-05-30T01:18:40Z
issue,115985,Faster Pytorch dequantize() + matmul for quantized models,https://github.com/pytorch/pytorch/issues/115985,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,categorize and decide next maintainer owner/action,2024-05-29T19:02:28Z
issue,127215,No way to config profiling scope in torch.autograd.profile,https://github.com/pytorch/pytorch/issues/127215,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: profiler,categorize and decide next maintainer owner/action,2024-05-28T12:15:52Z
issue,126770,_convert_input_to_fake doesn't work when there are no inputs,https://github.com/pytorch/pytorch/issues/126770,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: export,categorize and decide next maintainer owner/action,2024-05-27T16:26:00Z
issue,126974,TypeError: slice indices must be integers or None or have an __index__ method,https://github.com/pytorch/pytorch/issues/126974,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,categorize and decide next maintainer owner/action,2024-05-24T06:14:06Z
issue,127013,[CI] Download android and ios dependencies from pip instead of conda,https://github.com/pytorch/pytorch/issues/127013,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: mobile,categorize and decide next maintainer owner/action,2024-05-23T22:35:45Z
issue,126843,Torchscript export of `Salesforce/blip2-opt-2.7b` model fails with Conv shape error,https://github.com/pytorch/pytorch/issues/126843,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-05-22T19:33:49Z
issue,118165,[Quantization-aware training] Can not quantize nn.MultiheadAttention module,https://github.com/pytorch/pytorch/issues/118165,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,categorize and decide next maintainer owner/action,2024-05-22T03:36:40Z
issue,126291,Pytorch -> ExecuTorch conversion fails with BatchNorm2d layer,https://github.com/pytorch/pytorch/issues/126291,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: export,categorize and decide next maintainer owner/action,2024-05-16T20:03:38Z
issue,125293,Need a C++ way to convert PyTorch model to TorchScript model,https://github.com/pytorch/pytorch/issues/125293,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-05-16T11:07:20Z
issue,126277,[JIT] Unexpected change in requires_grad attribute after calling JIT traced function with the profiling executor disabled,https://github.com/pytorch/pytorch/issues/126277,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-05-16T00:53:33Z
issue,126276,torch script not support ModuleList/Sequential ?,https://github.com/pytorch/pytorch/issues/126276,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-05-16T00:53:19Z
issue,126167,`aten::transpose.int` is acting errouneously,https://github.com/pytorch/pytorch/issues/126167,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-05-15T13:59:44Z
issue,125266,[export] linalg ops after core aten decomposition,https://github.com/pytorch/pytorch/issues/125266,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: decompositions, oncall: export",categorize and decide next maintainer owner/action,2024-05-14T20:25:16Z
issue,126140,[PT2E Quantization] _fuse_conv_bn_ | RuntimeError: !argument.default_value() INTERNAL ASSERT FAILED,https://github.com/pytorch/pytorch/issues/126140,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,categorize and decide next maintainer owner/action,2024-05-14T18:19:10Z
issue,125749,'MultiHeadAttention.attention' is being compiled since it was called from 'MultiHeadAttention.forward',https://github.com/pytorch/pytorch/issues/125749,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-05-08T09:58:40Z
issue,59612,Why Pytroch.distributed does not expose NCCL cuda stream?,https://github.com/pytorch/pytorch/issues/59612,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: c10d, oncall: distributed",categorize and decide next maintainer owner/action,2024-05-08T07:06:07Z
issue,125734,☂️ `torch.compile` generates slower code for LLMs than eager on ARM platform (M1/AARCH64),https://github.com/pytorch/pytorch/issues/125734,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: arm, module: performance, oncall: cpu inductor, oncall: pt2",categorize and decide next maintainer owner/action,2024-05-08T01:18:19Z
issue,110259,Cannot avoid kineto_LIBRARY-NOTFOUND error when using pre-built pytorch,https://github.com/pytorch/pytorch/issues/110259,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: binaries, oncall: profiler",categorize and decide next maintainer owner/action,2024-05-07T10:16:03Z
issue,125564,Issues in loading quantized weights with Version==None,https://github.com/pytorch/pytorch/issues/125564,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,categorize and decide next maintainer owner/action,2024-05-07T06:16:04Z
issue,123345,torch.export does not support Sparse CSR tensors,https://github.com/pytorch/pytorch/issues/123345,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: export,categorize and decide next maintainer owner/action,2024-05-06T20:28:02Z
issue,125003,bool as scalar in jit ir API not working as expected.,https://github.com/pytorch/pytorch/issues/125003,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-05-06T07:09:10Z
issue,66623,Constant folding in symbolic shape inference fails: expected scalar type Long but found Float,https://github.com/pytorch/pytorch/issues/66623,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-05-05T14:35:07Z
issue,125004,RuntimeError: Unsupported qscheme: per_channel_affine. When using quantise_fx(,https://github.com/pytorch/pytorch/issues/125004,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,categorize and decide next maintainer owner/action,2024-05-01T03:03:22Z
issue,125245,[export] Add customization API for unflattener to control restored module types.,https://github.com/pytorch/pytorch/issues/125245,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: export,categorize and decide next maintainer owner/action,2024-04-30T17:12:08Z
issue,124458,Understand the oneDNN graph fusion with torch script,https://github.com/pytorch/pytorch/issues/124458,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-04-30T03:33:17Z
issue,125007,Support exporting the compute graph JiT compiled by inductor?,https://github.com/pytorch/pytorch/issues/125007,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: export,categorize and decide next maintainer owner/action,2024-04-29T16:02:19Z
issue,76881,RecursionError when running torch.jit.script inside JitTestCase,https://github.com/pytorch/pytorch/issues/76881,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-04-29T03:06:09Z
issue,125033,Tensorboard's SummaryWriter.add_graph() doesn't work with packedSequence,https://github.com/pytorch/pytorch/issues/125033,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: tensorboard, oncall: visualization",categorize and decide next maintainer owner/action,2024-04-26T22:01:38Z
issue,124674,Improving format of communication metadata in PyTorch Execution Trace.,https://github.com/pytorch/pytorch/issues/124674,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: profiler,categorize and decide next maintainer owner/action,2024-04-25T19:37:27Z
issue,80861,"jit.freeze throws RuntimeError: stack_out && stack_out->size() == 1 INTERNAL ASSERT FAILED at ""../torch/csrc/jit/passes/frozen_conv_folding.cpp"":281",https://github.com/pytorch/pytorch/issues/80861,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-04-25T13:37:38Z
issue,59530,Batch size is hardcoded using torch.jit.trace with LSTMCell,https://github.com/pytorch/pytorch/issues/59530,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-04-25T07:22:44Z
issue,124796,Dynamo Export Support for Qwen/Qwen-7B-Chat: Mutating module attribute _ntk_alpha_cached_list during export,https://github.com/pytorch/pytorch/issues/124796,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: export, oncall: pt2",categorize and decide next maintainer owner/action,2024-04-25T02:21:03Z
issue,124507,torch.export of ResNet with dynamic height fails due to constraint violation,https://github.com/pytorch/pytorch/issues/124507,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: export,categorize and decide next maintainer owner/action,2024-04-21T16:46:40Z
issue,124461,Support constant value in torch.jit.script,https://github.com/pytorch/pytorch/issues/124461,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-04-19T15:23:51Z
issue,76807,There is a bug with latest stable torch version and the following Nightly versions related to `optimize_for_mobile`,https://github.com/pytorch/pytorch/issues/76807,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: mobile,categorize and decide next maintainer owner/action,2024-04-19T14:03:06Z
issue,123961,JIT reports RuntimeError when decorating forward with torch.jit.script,https://github.com/pytorch/pytorch/issues/123961,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-04-13T03:26:59Z
issue,52363,RuntimeError: Can't redefine method: forward on class: __torch__.torch.nn.quantized.modules.conv.Conv1d (of Python compilation unit at: 0x5571a223a770),https://github.com/pytorch/pytorch/issues/52363,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-04-12T19:41:05Z
issue,123824,Precision problem in `quantize_fx.convert_fx` and question about the `example_inputs` for `quantize_fx.prepare_fx`.,https://github.com/pytorch/pytorch/issues/123824,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,categorize and decide next maintainer owner/action,2024-04-12T19:17:06Z
issue,69425,Integrate Libtorch into Unreal Engine 4: _ivalue_INTERNAL ASSERT FAILED,https://github.com/pytorch/pytorch/issues/69425,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-04-11T22:20:10Z
issue,84318,[Profiler] Generic Tensor summary,https://github.com/pytorch/pytorch/issues/84318,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, oncall: profiler",categorize and decide next maintainer owner/action,2024-04-11T04:14:15Z
issue,123464,torch._C.JitType annotation is busted,https://github.com/pytorch/pytorch/issues/123464,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: typing, oncall: jit",categorize and decide next maintainer owner/action,2024-04-09T18:54:23Z
issue,58845,Use jit in flask,https://github.com/pytorch/pytorch/issues/58845,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-04-07T08:58:45Z
issue,123066,[JIT archive] Add a flag to not include debug files,https://github.com/pytorch/pytorch/issues/123066,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-04-01T16:05:16Z
issue,122303,How to exclude some modules from quantization?,https://github.com/pytorch/pytorch/issues/122303,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,categorize and decide next maintainer owner/action,2024-03-27T08:22:57Z
issue,66017,Torchscript `jit.script` breaks with OSError if dataclass used to pass results during model execution,https://github.com/pytorch/pytorch/issues/66017,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-03-26T16:22:01Z
issue,120437,fx::GraphModule::to_folder doesn't seem to handle _PyTreeCodeGen correctly.,https://github.com/pytorch/pytorch/issues/120437,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: fx, oncall: export",categorize and decide next maintainer owner/action,2024-03-23T04:28:23Z
issue,121230,INTERNAL ASSERT FAILED in torch.pow,https://github.com/pytorch/pytorch/issues/121230,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,categorize and decide next maintainer owner/action,2024-03-22T21:51:04Z
issue,122183,[FX Graph Mode Quantization] Prepare model fails with Linear-BN when `affine=False`,https://github.com/pytorch/pytorch/issues/122183,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,categorize and decide next maintainer owner/action,2024-03-22T21:48:09Z
issue,122224,support QAT fusion for conv transpose + bn + relu,https://github.com/pytorch/pytorch/issues/122224,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,categorize and decide next maintainer owner/action,2024-03-19T22:26:36Z
issue,117239,```torch.nn.Upsample``` class: Internal assert failed,https://github.com/pytorch/pytorch/issues/117239,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,categorize and decide next maintainer owner/action,2024-03-19T21:52:07Z
issue,87145,Unrecognized data format when using release libtorch libraries in debug build,https://github.com/pytorch/pytorch/issues/87145,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-03-19T16:04:31Z
issue,122020,torch.jit.script(model) and torch.jit.trace(model) performance degradation.,https://github.com/pytorch/pytorch/issues/122020,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-03-18T21:52:38Z
issue,69782,Torch Profiler does not count FLOPs for backward pass,https://github.com/pytorch/pytorch/issues/69782,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: profiler,categorize and decide next maintainer owner/action,2024-03-18T14:16:18Z
issue,121597,LibTorch-Lite 2.1.0 Libs missing x86_64,https://github.com/pytorch/pytorch/issues/121597,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: binaries, oncall: mobile",categorize and decide next maintainer owner/action,2024-03-11T20:15:23Z
issue,75556,[JIT] `torch.jit.ignore` is not working on hooks,https://github.com/pytorch/pytorch/issues/75556,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-03-07T05:50:54Z
issue,120705,Trying doing static_quantization with Pytorch 2.1.0 torch.ao.quantization,https://github.com/pytorch/pytorch/issues/120705,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,categorize and decide next maintainer owner/action,2024-03-05T09:02:43Z
issue,96639,Completely different output between .pt and .ptl,https://github.com/pytorch/pytorch/issues/96639,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: mobile,categorize and decide next maintainer owner/action,2024-03-04T17:52:59Z
issue,63929,[Poll] Support higher-order gradient computation in DDP,https://github.com/pytorch/pytorch/issues/63929,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: c10d, module: ddp, oncall: distributed",categorize and decide next maintainer owner/action,2024-03-02T00:29:57Z
issue,70051,[feature request] Support python decorators in TorchScript,https://github.com/pytorch/pytorch/issues/70051,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-02-29T09:58:29Z
issue,59690,Inference ran on new thread leak memory on Android,https://github.com/pytorch/pytorch/issues/59690,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: android, module: memory usage, oncall: mobile",categorize and decide next maintainer owner/action,2024-02-26T03:13:33Z
issue,120485,Try to make `jit.trace` working with dynamic control flow but it does not work,https://github.com/pytorch/pytorch/issues/120485,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-02-23T14:26:07Z
issue,43388,"Abort message: ‘terminating with uncaught exception of type c10::Error: _ivalue_ INTERNAL ASSERT FAILED at ../torch/csrc/jit/api/object.cpp:19, please report a bug to PyTorch. (_ivalue at ../torch/csrc/jit/api/object.cpp:19)",https://github.com/pytorch/pytorch/issues/43388,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit, oncall: mobile",categorize and decide next maintainer owner/action,2024-02-23T09:11:01Z
issue,89386,How can i patch the torch.jit in the second solution? Could not figure out entrypoint ?,https://github.com/pytorch/pytorch/issues/89386,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-02-22T17:10:32Z
issue,119484,Follow-up #119356: Preserve node names for torch.export.export,https://github.com/pytorch/pytorch/issues/119484,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: export,categorize and decide next maintainer owner/action,2024-02-12T17:45:48Z
issue,119454,Python 3.12 CPU test_fx.py::TestFunctionalTracing::test_nn_functional_adaptive_max_pool is failing,https://github.com/pytorch/pytorch/issues/119454,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: fx, module: tests, oncall: fx",categorize and decide next maintainer owner/action,2024-02-08T16:47:07Z
issue,119192,[JIT] Deserialized `RecursiveScriptFunction` forgets default values for arguments.,https://github.com/pytorch/pytorch/issues/119192,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-02-08T15:27:29Z
issue,118410,[export] Error loudly when tracing over a module not in the module hierarchy,https://github.com/pytorch/pytorch/issues/118410,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: export,categorize and decide next maintainer owner/action,2024-02-06T23:28:17Z
issue,14867,[C++/Pytorch] Get input shapes,https://github.com/pytorch/pytorch/issues/14867,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-02-06T01:41:33Z
issue,118463,Inconsistency in Output for `torch.argmin` in JIT Compiled Model,https://github.com/pytorch/pytorch/issues/118463,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-01-29T19:07:42Z
issue,118462,Inconsistency in Output for `torch.nn.functional.poisson_nll_loss` in JIT Compiled Model,https://github.com/pytorch/pytorch/issues/118462,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: loss, oncall: jit",categorize and decide next maintainer owner/action,2024-01-29T19:07:30Z
issue,118297,Inconsistent outputs of `cosine_similarity` and `linalg_svdvals` between eagermode and torch.jit.trace(),https://github.com/pytorch/pytorch/issues/118297,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-01-29T17:19:45Z
issue,118401,Inconsistency in Output When Using `torch.nn.functional.selu` with JIT Compilation,https://github.com/pytorch/pytorch/issues/118401,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-01-29T15:12:05Z
issue,118400,Inconsistent output of `torch.lodget` with JIT compilation,https://github.com/pytorch/pytorch/issues/118400,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-01-29T15:11:57Z
issue,118189,Cannot load a jit-traced model if it uses nan_to_num with too large a constant.,https://github.com/pytorch/pytorch/issues/118189,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-01-29T14:56:29Z
issue,118303,Inconsistent outputs of `mish` and `log10` between eagermode and torch.jit.trace,https://github.com/pytorch/pytorch/issues/118303,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-01-27T10:00:08Z
issue,118302,Inconsistent outputs of `torch.adaptive_avg_pool1d` and `torch.threshold_` between eagermode and torch.jit.trace,https://github.com/pytorch/pytorch/issues/118302,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-01-27T10:00:01Z
issue,118300,Inconsistent outputs in `linalg_ldl_factor_ex` and `special_spheriacal_bessel_j0` between eagermode and torch.jit.trace,https://github.com/pytorch/pytorch/issues/118300,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-01-27T09:59:47Z
issue,82417,[JIT] _unsafe_view returns alias when size(input) = size argument,https://github.com/pytorch/pytorch/issues/82417,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-01-25T04:48:42Z
issue,118086,Different behaviors with torch.isfinite and torch.tril between eagermode and JIT,https://github.com/pytorch/pytorch/issues/118086,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-01-24T09:24:08Z
issue,117920,Could not run 'quantized::embedding_byte' with arguments from the 'QuantizedCPU' backend.,https://github.com/pytorch/pytorch/issues/117920,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,categorize and decide next maintainer owner/action,2024-01-22T13:28:02Z
issue,113945,torch.jit.trace has incorrect execution for += operation during compilation,https://github.com/pytorch/pytorch/issues/113945,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-01-18T11:38:41Z
issue,117541,Jit model running on cuda:1 uses memory on cuda:0,https://github.com/pytorch/pytorch/issues/117541,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-01-16T14:38:04Z
issue,49538,jit tracer doesn't work with unflatten layer,https://github.com/pytorch/pytorch/issues/49538,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-01-12T15:06:42Z
issue,54140,[Torchscript] Error detected in torch::jit::(anonymous namespace)::DifferentiableGraphBackward on custom RNN model,https://github.com/pytorch/pytorch/issues/54140,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-01-12T03:08:58Z
issue,49444,[RFC] Add JIT graph fuser for oneDNN Graph API,https://github.com/pytorch/pytorch/issues/49444,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-01-12T03:07:52Z
issue,5489,"torch.jit.trace(network, data) fails if data is an OrderedDict",https://github.com/pytorch/pytorch/issues/5489,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, oncall: jit",categorize and decide next maintainer owner/action,2024-01-12T03:07:42Z
issue,48325,[JIT] Constant Propagation Shouldn't Run on Values that Escape the Graph,https://github.com/pytorch/pytorch/issues/48325,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-01-12T03:06:39Z
issue,43501,[jit] Shape hints for TorchScript,https://github.com/pytorch/pytorch/issues/43501,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-01-12T03:06:16Z
issue,41909,Dropout in TorchScript under torch.no_grad() produces incoherent results,https://github.com/pytorch/pytorch/issues/41909,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-01-12T03:06:12Z
issue,117229,```torch.nn.functional.pad```: Internal assert failed,https://github.com/pytorch/pytorch/issues/117229,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,categorize and decide next maintainer owner/action,2024-01-11T19:43:44Z
issue,117102,conver_fx report len(device) > 1 error,https://github.com/pytorch/pytorch/issues/117102,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,categorize and decide next maintainer owner/action,2024-01-11T15:09:57Z
issue,116992,torch.fake_quantize_per_tensor_affine with inf input provides wrong output value,https://github.com/pytorch/pytorch/issues/116992,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,categorize and decide next maintainer owner/action,2024-01-09T20:39:52Z
issue,116865,Optimizing Per-Channel Quantization for Improved Inference Performance,https://github.com/pytorch/pytorch/issues/116865,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,categorize and decide next maintainer owner/action,2024-01-09T02:05:44Z
issue,116253,`torch.Tensor.flipud`: Heap-buffer-overflow,https://github.com/pytorch/pytorch/issues/116253,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,categorize and decide next maintainer owner/action,2024-01-05T21:57:07Z
issue,116612,Loaded torch.jit traced module treats dtypes incorrectly in autograd,https://github.com/pytorch/pytorch/issues/116612,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-01-03T08:34:22Z
issue,82065,optimize_for_mobile vulkan_prepack::conv2d_clamp_prepack,https://github.com/pytorch/pytorch/issues/82065,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: vulkan, oncall: mobile",categorize and decide next maintainer owner/action,2024-01-02T21:58:06Z
issue,116544,[export] namedtuple inputs can't be retraced,https://github.com/pytorch/pytorch/issues/116544,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: export,categorize and decide next maintainer owner/action,2023-12-30T19:41:03Z
issue,116498,torch 1.8 cannot torch.jit.load for script model,https://github.com/pytorch/pytorch/issues/116498,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-12-29T00:50:59Z
issue,116240,[inductor][cpu] RuntimeError: quantized_resize_cpu_ does not have a deterministic implementation,https://github.com/pytorch/pytorch/issues/116240,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: cpu, oncall: quantization",categorize and decide next maintainer owner/action,2023-12-24T14:36:04Z
issue,116250,`torch.Tensor.rot90`: Heap-buffer-overflow,https://github.com/pytorch/pytorch/issues/116250,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,categorize and decide next maintainer owner/action,2023-12-21T16:07:05Z
issue,116252,`torch.Tensor.fliplr`: Heap-buffer-overflow,https://github.com/pytorch/pytorch/issues/116252,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,categorize and decide next maintainer owner/action,2023-12-21T16:06:58Z
issue,116251,`torch.Tensor.flip`: Heap-buffer-overflow,https://github.com/pytorch/pytorch/issues/116251,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,categorize and decide next maintainer owner/action,2023-12-21T16:06:51Z
issue,116254,C++ API `at::quantized_max_pool2d`: Heap-buffer-overflow,https://github.com/pytorch/pytorch/issues/116254,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,categorize and decide next maintainer owner/action,2023-12-21T16:06:37Z
issue,116010,Why has no torch::jit::load() in Pytorch Android's libtorch_cpu.so,https://github.com/pytorch/pytorch/issues/116010,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-12-19T05:16:21Z
issue,115914,torch.jit.script fails when parsing torch.SymInt,https://github.com/pytorch/pytorch/issues/115914,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-12-17T05:35:16Z
issue,115736,Precision Loss in JIT ScriptModule at float64 Tensor Init,https://github.com/pytorch/pytorch/issues/115736,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-12-15T10:34:26Z
issue,115733,RuntimeError: thread_predicates_.find(tv_inp) != thread_predicates_.end(),https://github.com/pytorch/pytorch/issues/115733,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-12-13T16:15:43Z
issue,115622,Torch jit load model with enums,https://github.com/pytorch/pytorch/issues/115622,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-12-13T16:15:19Z
issue,115639,"`torch.jit.save` fails for static function from C++ TorchScript extension with ""Failed to downcast a Function to a GraphFunction""",https://github.com/pytorch/pytorch/issues/115639,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-12-13T13:47:47Z
issue,115488,torch.export.export is overriding source_fn inside higher order op submodules,https://github.com/pytorch/pytorch/issues/115488,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: export,categorize and decide next maintainer owner/action,2023-12-11T18:23:38Z
issue,115055,Better test suite for torch.export.export,https://github.com/pytorch/pytorch/issues/115055,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: export,categorize and decide next maintainer owner/action,2023-12-07T20:25:56Z
issue,62962,[JIT] NVRTC unknown error,https://github.com/pytorch/pytorch/issues/62962,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-12-07T03:11:09Z
issue,55603,Faster Inference time for first .pt model prediction,https://github.com/pytorch/pytorch/issues/55603,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-12-05T17:42:07Z
issue,75788,`torch.jit.script` Script functions do return `requires_grad = False` if `torch.no_grad()` has been used,https://github.com/pytorch/pytorch/issues/75788,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-12-05T17:39:42Z
issue,114967,Inplace update to buffers doesn't work with `aot_compile`,https://github.com/pytorch/pytorch/issues/114967,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: export,categorize and decide next maintainer owner/action,2023-12-04T21:24:20Z
issue,114102,"""aten::prelu"" can not be used in the metal backend",https://github.com/pytorch/pytorch/issues/114102,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: mobile,categorize and decide next maintainer owner/action,2023-12-03T03:51:00Z
issue,52471,all reduce hangs and does not throw exception when CUDA_VISIBLE_DEVICES is not set properly using the NCCL backend,https://github.com/pytorch/pytorch/issues/52471,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: nccl, oncall: distributed",categorize and decide next maintainer owner/action,2023-11-30T13:38:05Z
issue,114543,[torch.jit.script] Torchscript produces incorrect result when argmax result is used in indexing,https://github.com/pytorch/pytorch/issues/114543,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-11-28T07:53:03Z
issue,114283,CppException in Android Studio with faster_rcnn_fbnetv3g_fpn.yaml Configuration,https://github.com/pytorch/pytorch/issues/114283,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: mobile,categorize and decide next maintainer owner/action,2023-11-21T23:13:24Z
issue,92073,compilig MultiHeadAttention,https://github.com/pytorch/pytorch/issues/92073,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-11-14T02:47:48Z
issue,90885,TorchScript with complex abs doesn't work in backward,https://github.com/pytorch/pytorch/issues/90885,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-11-13T17:21:49Z
issue,66930,JIT vs. eager mismatches for jit.traced `int8` to `int32` casting,https://github.com/pytorch/pytorch/issues/66930,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-11-11T09:33:32Z
issue,92987,Test Failure: TestUpgraders.test_aten_div_scalar_at_3 on a big-endian machine (issue in torch.jit.load()),https://github.com/pytorch/pytorch/issues/92987,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-11-08T10:39:06Z
issue,80561,Cannot run scripted BERT_Pytorch,https://github.com/pytorch/pytorch/issues/80561,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-11-03T20:06:16Z
issue,112425,torch.jit.trace output changes when I add comments to the file,https://github.com/pytorch/pytorch/issues/112425,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-11-02T14:18:40Z
issue,89114,jit.script() fails to resolve/cast Optional[Tensor] fields of sub-modules or base classes of the object being scripted,https://github.com/pytorch/pytorch/issues/89114,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-11-01T08:03:04Z
issue,76295,TorchScript: jit.script fails when using 'tolist' with int32 tensors.,https://github.com/pytorch/pytorch/issues/76295,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-11-01T04:05:47Z
issue,69609,[mobile_optimizer] RuntimeError: We don't have an op for metal_prepack::conv2d_prepack,https://github.com/pytorch/pytorch/issues/69609,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: mobile,categorize and decide next maintainer owner/action,2023-10-29T16:18:35Z
issue,111792,torch.jit.trace is not able to trace torch extension,https://github.com/pytorch/pytorch/issues/111792,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-10-24T02:31:17Z
issue,111863,Jit scripting support for `|` and mixing `typing`.,https://github.com/pytorch/pytorch/issues/111863,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-10-23T23:58:55Z
issue,97047,Optimize for mobile produces incorrect result with INSERT_FOLD_PREPACK_OPS optimization,https://github.com/pytorch/pytorch/issues/97047,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit, oncall: mobile",categorize and decide next maintainer owner/action,2023-10-13T09:55:31Z
issue,111135,"No op for aten::where with argument types: Tensor, Tensor, bool.",https://github.com/pytorch/pytorch/issues/111135,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-10-12T14:06:34Z
issue,98695,Torchscript: Name Mangling prevents Type Refinement,https://github.com/pytorch/pytorch/issues/98695,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-10-11T21:15:44Z
issue,68568,"About ModuleDict indexing when save model based on ""torch.jit.script()"" method",https://github.com/pytorch/pytorch/issues/68568,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-10-09T16:57:05Z
issue,103070,Unexpected failure in LLVM JIT when running TorchScript model in C++,https://github.com/pytorch/pytorch/issues/103070,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-10-08T03:05:12Z
issue,29843,torch.distributions.normal.Normal is not JIT supported,https://github.com/pytorch/pytorch/issues/29843,needs triage,47,triage signal present; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-10-06T18:35:26Z
issue,110379,Pytorch LoadNativeLibrary issue,https://github.com/pytorch/pytorch/issues/110379,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: mobile,categorize and decide next maintainer owner/action,2023-10-04T19:52:30Z
issue,42885,Unable to call `super` method with TorchScript,https://github.com/pytorch/pytorch/issues/42885,needs triage,47,triage signal present; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-10-04T17:26:32Z
issue,48932,Unknown builtin op: torchvision::nms when loading scripted FasterRCNN,https://github.com/pytorch/pytorch/issues/48932,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-10-03T17:35:04Z
issue,89185,[feature request] Add ability to preserve traced shape during torch.jit.save and torch.jit.load,https://github.com/pytorch/pytorch/issues/89185,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-09-30T04:35:18Z
issue,109488,[bug] FALLBACK path has been taken inside: runCudaFusionGroup,https://github.com/pytorch/pytorch/issues/109488,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-09-25T09:33:10Z
issue,67146,`torch.jit.is_scripting()` not set when scripting a Module,https://github.com/pytorch/pytorch/issues/67146,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-09-20T16:54:58Z
issue,109699,[TorchScript] Support ScriptFunction arguments in torch.jit.script calls.,https://github.com/pytorch/pytorch/issues/109699,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-09-20T15:02:17Z
issue,109719,Make torch.cuda.graphs.is_current_stream_capturing() available in TorchScript,https://github.com/pytorch/pytorch/issues/109719,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: cuda graphs, oncall: jit",categorize and decide next maintainer owner/action,2023-09-20T15:00:36Z
issue,109579,[Android: React Native] couldn't find DSO to load: libtorch-code-gen.so when loading model,https://github.com/pytorch/pytorch/issues/109579,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: mobile,categorize and decide next maintainer owner/action,2023-09-20T00:11:45Z
issue,109113,"The API ""torch::jit::_load_for_mobile"" is limited to create an object living on the stack.",https://github.com/pytorch/pytorch/issues/109113,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-09-12T15:19:06Z
issue,84831,INTERNAL ASSERT FAILED for _jit_pass_vulkan_optimize_for_mobile (Google Colab),https://github.com/pytorch/pytorch/issues/84831,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: vulkan, oncall: jit, oncall: mobile",categorize and decide next maintainer owner/action,2023-09-11T06:14:33Z
issue,93913,[BUG] jit.trace not working for torchvision ViT models,https://github.com/pytorch/pytorch/issues/93913,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-09-08T21:25:43Z
issue,108665,INTERNAL ASSERT FAILED in `shape_type_inference.cpp`,https://github.com/pytorch/pytorch/issues/108665,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-09-08T14:50:29Z
issue,108865,Heap buffer overflow with `torch::load` on fuzzy data,https://github.com/pytorch/pytorch/issues/108865,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-09-08T14:14:31Z
issue,108670,torch.jit.script produces incorrect gradients,https://github.com/pytorch/pytorch/issues/108670,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-09-07T01:17:37Z
issue,108332,RuntimeError: dims.value().size() == self->getMaybeRFactorDomain().size(),https://github.com/pytorch/pytorch/issues/108332,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-09-04T22:29:02Z
issue,105821,JIT input aliasing does not support aten::fill_,https://github.com/pytorch/pytorch/issues/105821,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-09-01T11:47:17Z
issue,41869,TorchScript pack_padded_sequence and pad_packed_sequence run time error,https://github.com/pytorch/pytorch/issues/41869,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-08-31T20:30:33Z
issue,52047,Detect OpenMP Loop and this application may hang warnings,https://github.com/pytorch/pytorch/issues/52047,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: multithreading, module: tests, oncall: jit",categorize and decide next maintainer owner/action,2023-08-31T08:14:27Z
issue,108226,torch.nn.functional.pad() with value type bool,https://github.com/pytorch/pytorch/issues/108226,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-08-30T14:42:47Z
issue,107771,libtorch infer error : CUDNN_STATUS_INTERNAL_ERROR,https://github.com/pytorch/pytorch/issues/107771,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-08-24T22:58:42Z
issue,60937,[jit] About traced script module using in c++,https://github.com/pytorch/pytorch/issues/60937,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-08-23T04:38:35Z
issue,40561,Cannot manually assign a tensor to .grad from TorchScript,https://github.com/pytorch/pytorch/issues/40561,needs triage,47,triage signal present; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-08-16T10:37:22Z
issue,106699,torch.jit.script: scripting doesn't work with wraps,https://github.com/pytorch/pytorch/issues/106699,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-08-10T11:19:47Z
issue,48082,Microsoft C++ exception: torch::jit::ErrorReport at memory location 0x0000001C0C3BBB40.,https://github.com/pytorch/pytorch/issues/48082,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-08-09T19:39:42Z
issue,32976,[JIT] pytorch 1.4 breaks torch.jit.script(LSTM/GRU),https://github.com/pytorch/pytorch/issues/32976,needs triage,47,triage signal present; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-08-09T02:25:54Z
issue,106764,[Feature request] Add new API Tensor.device_as,https://github.com/pytorch/pytorch/issues/106764,needs triage,47,"triage signal present; old update age or low-urgency/stale label; component/oncall labels: module: python frontend, oncall: jit",categorize and decide next maintainer owner/action,2023-08-08T23:56:04Z
issue,68014,Add TORCHELASTIC_RESTART_COUNT env variable to DDP logging,https://github.com/pytorch/pytorch/issues/68014,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, module: ddp, oncall: distributed",categorize and decide next maintainer owner/action,2023-08-07T13:48:54Z
issue,104450,"Numpy/scipy module works fine with Torch modules, but not TorchScript. How to torchscript a numpy/scipy module?",https://github.com/pytorch/pytorch/issues/104450,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-08-02T17:55:14Z
issue,105997,Flatbuffer torchscipt files don't load in PyTorch Android Lite 1.13.1,https://github.com/pytorch/pytorch/issues/105997,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-08-02T17:29:17Z
issue,106435,"[JIT] .item() dict keys cause `RuntimeError: Cannot create dict for key type 'Scalar', only int, float, complex, Tensor, device and string keys are supported`",https://github.com/pytorch/pytorch/issues/106435,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-08-02T13:51:35Z
issue,106128,Torchscript optimizer incorrectly applies constant propagation to convert prim::ListConstruct() into prim::Constant,https://github.com/pytorch/pytorch/issues/106128,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-08-01T16:17:05Z
issue,106129,Allow settingsetGraphExecutorOptimize default for all threads,https://github.com/pytorch/pytorch/issues/106129,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-08-01T16:17:00Z
issue,90365,[JIT] INTERNAL ASSERT FAILED when dispatching for `torch.Tensor.view`,https://github.com/pytorch/pytorch/issues/90365,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-07-27T21:51:50Z
issue,105917,Torch.jit.frontend.NotSupportedError: not supporting functions with variable number of arguments.,https://github.com/pytorch/pytorch/issues/105917,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-07-27T14:38:01Z
issue,105918,Torch.jit : RuntimeError: Unable to extract string literal index for ModuleDict,https://github.com/pytorch/pytorch/issues/105918,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-07-27T14:37:58Z
issue,105900,torch.jit.frontend.NotSupported when compiling stable-diffusion,https://github.com/pytorch/pytorch/issues/105900,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-07-27T14:27:36Z
issue,104959,Saving a LightningModule torch.jit.ScriptModule is incompatible with torch.amp.autocast,https://github.com/pytorch/pytorch/issues/104959,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision), oncall: jit",categorize and decide next maintainer owner/action,2023-07-20T02:32:21Z
issue,43520,[jit] cannot use functools.partial with TorchScript,https://github.com/pytorch/pytorch/issues/43520,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-07-17T13:22:03Z
issue,105196,Error loading TorchScript model with torchvision::nms operation in libtorch,https://github.com/pytorch/pytorch/issues/105196,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-07-14T02:58:47Z
issue,104906,torch.jit.frontend.NotSupportedError: keyword-arg expansion is not supported: for dgl.nn.HeteroGraphConv(),https://github.com/pytorch/pytorch/issues/104906,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-07-13T02:28:38Z
issue,51135,[TorchScript Usability] Behaviors of used-defined subclasses of torch.tensor are silently ignored,https://github.com/pytorch/pytorch/issues/51135,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-07-10T10:00:16Z
issue,104712,torch.jit slicing error (styleganv2),https://github.com/pytorch/pytorch/issues/104712,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-07-07T07:55:26Z
issue,104732,"torch.jit.trace says ""Arguments for call are invalid"" on torch.ops.aten.sub(3, x, alpha=3)",https://github.com/pytorch/pytorch/issues/104732,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-07-07T01:23:04Z
issue,92294,jit.fork stalls multiprocessing dataloader,https://github.com/pytorch/pytorch/issues/92294,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: data, module: dataloader, oncall: jit",categorize and decide next maintainer owner/action,2023-07-06T07:06:00Z
issue,92246,test_qnnpack_add fails,https://github.com/pytorch/pytorch/issues/92246,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: xnnpack, oncall: mobile",categorize and decide next maintainer owner/action,2023-06-30T20:58:12Z
issue,104458,[Feature Request] Add a new overload of torch::jit::load to restore traced shape and type,https://github.com/pytorch/pytorch/issues/104458,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-06-30T19:56:02Z
issue,103788,"[Torch Mlir] avg_pool1d function padding init value should be (0,)",https://github.com/pytorch/pytorch/issues/103788,needs triage,47,triage signal present; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-06-29T13:43:43Z
issue,51969,Support backward hooks in JIT,https://github.com/pytorch/pytorch/issues/51969,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-06-29T12:58:01Z
issue,104041,Scripted model is loaded on GPU but the inference seems to utilize the CPU with zero GPU utilization,https://github.com/pytorch/pytorch/issues/104041,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-06-22T20:56:21Z
issue,100730,`torch::jit::EliminateExceptions` lowering pass never completes on specific model,https://github.com/pytorch/pytorch/issues/100730,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-06-20T20:32:31Z
issue,62221,Torchscript Optimize for Mobile RuntimeError: Can't deepcopy IValue with tag: Enum,https://github.com/pytorch/pytorch/issues/62221,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-06-14T09:43:54Z
issue,103250,torch.jit.script mean(keepdim=True) segfaults on GPU,https://github.com/pytorch/pytorch/issues/103250,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-06-09T23:34:29Z
issue,47887,"ONNX export of a scripted submodule fails with ""Modules that are called during a trace must be registered as submodules of the thing being traced""",https://github.com/pytorch/pytorch/issues/47887,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: onnx, oncall: jit",categorize and decide next maintainer owner/action,2023-06-06T07:39:31Z
issue,102653,torchscript dataclasses have bad support for class types as fields,https://github.com/pytorch/pytorch/issues/102653,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-05-31T21:49:11Z
issue,33491,jit.trace checker fails for LSTM,https://github.com/pytorch/pytorch/issues/33491,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-05-30T09:27:03Z
issue,68340,Torch 1.10.0 - RuntimeError: requires_grad_ is not supported on ScriptModules,https://github.com/pytorch/pytorch/issues/68340,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-05-22T16:59:09Z
issue,101855,torch.jit.trace() Floating point exception,https://github.com/pytorch/pytorch/issues/101855,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-05-19T20:23:58Z
issue,58789,Type annotations for torch.jit.* decorators,https://github.com/pytorch/pytorch/issues/58789,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-05-18T19:13:35Z
issue,70058,Unknown builtin op: spconv::get_indice_pairs,https://github.com/pytorch/pytorch/issues/70058,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-05-18T06:15:05Z
issue,101055,DISABLE libtorch-2.0.0+cu117 destructor exception,https://github.com/pytorch/pytorch/issues/101055,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-05-11T03:52:30Z
issue,100636,tracing does not work when torch.distributions is involved,https://github.com/pytorch/pytorch/issues/100636,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-05-05T13:38:56Z
issue,99821,Tracer cannot infer type of Seq2SeqLMOutput,https://github.com/pytorch/pytorch/issues/99821,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-04-24T17:29:48Z
issue,99561,Pytorch mobile crashes on Android when loading a custom model,https://github.com/pytorch/pytorch/issues/99561,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: android, oncall: mobile",categorize and decide next maintainer owner/action,2023-04-21T17:12:56Z
issue,96185,`ld: error: unknown argument '-force_load'` when linking libtorch on Android,https://github.com/pytorch/pytorch/issues/96185,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: build, oncall: mobile",categorize and decide next maintainer owner/action,2023-04-19T04:30:46Z
issue,98888,Changes to TorchScript autodiff changing default behavior are no longer accepted,https://github.com/pytorch/pytorch/issues/98888,needs triage,47,triage signal present; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-04-12T14:44:00Z
issue,73142,Support passing Python code as a string to torch.jit.script,https://github.com/pytorch/pytorch/issues/73142,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-04-11T20:51:11Z
issue,98064,Module 'Sequential' has no attribute '_modules' :,https://github.com/pytorch/pytorch/issues/98064,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-04-10T13:36:29Z
issue,91387,getting issue 'typeindex' file not found in Littorch-Lite/install/include/ATen/core/custom_class.h,https://github.com/pytorch/pytorch/issues/91387,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: mobile,categorize and decide next maintainer owner/action,2023-04-07T02:12:54Z
issue,98422,[FX] Symbolic trace over `torch.Tensor.${fn}` APIs,https://github.com/pytorch/pytorch/issues/98422,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: fx,categorize and decide next maintainer owner/action,2023-04-05T20:28:09Z
issue,52312,TorchScript can't wrap Python constants in current module,https://github.com/pytorch/pytorch/issues/52312,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-04-05T01:16:56Z
issue,98210,torch.jit.script + legacy executor mode has diff in some pattern,https://github.com/pytorch/pytorch/issues/98210,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-04-05T00:40:48Z
issue,98124,PyTorch Profiler fails recording functions,https://github.com/pytorch/pytorch/issues/98124,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: profiler,categorize and decide next maintainer owner/action,2023-04-04T14:10:23Z
issue,86667,Adding a linear layer leads to failure of `optimize_for_mobile`,https://github.com/pytorch/pytorch/issues/86667,needs triage,47,"triage signal present; old update age or low-urgency/stale label; component/oncall labels: module: mkldnn, oncall: jit",categorize and decide next maintainer owner/action,2023-04-02T20:02:33Z
issue,97913,BUG torch.jit.annotate on List + torch.stack give wrong DTYPE,https://github.com/pytorch/pytorch/issues/97913,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-03-30T14:42:11Z
issue,94908,ASSERT(initialized()) Debug Error after JIT fusion on Windows,https://github.com/pytorch/pytorch/issues/94908,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: nvfuser, oncall: jit",categorize and decide next maintainer owner/action,2023-03-30T07:56:58Z
issue,52517,[doc] JIT-compatible c++ custom ops,https://github.com/pytorch/pytorch/issues/52517,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: docs, oncall: jit",categorize and decide next maintainer owner/action,2023-03-28T08:07:58Z
issue,82312,"A/libc: Fatal signal 6 (SIGABRT), code -1 (SI_QUEUE) in tid 9792 (Background), pid 9674 (ample.testtorch)",https://github.com/pytorch/pytorch/issues/82312,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: mobile,categorize and decide next maintainer owner/action,2023-03-27T05:33:20Z
issue,96448,Unable to move torch.jit.load-ed models to XLA devices,https://github.com/pytorch/pytorch/issues/96448,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: xla, oncall: jit",categorize and decide next maintainer owner/action,2023-03-22T23:13:19Z
issue,97343,Traced module shows non-deterministic behaviour on CUDA,https://github.com/pytorch/pytorch/issues/97343,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-03-22T20:04:12Z
issue,95858,[JIT] Support string type annotations in NamedTuples,https://github.com/pytorch/pytorch/issues/95858,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-03-22T15:25:28Z
issue,68921,[Bug] forward function error occurs when the scriptmodule (made in Python) including the hook function is loaded from C++(Libtorch1.8.2),https://github.com/pytorch/pytorch/issues/68921,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-03-18T02:49:36Z
issue,96735,when run python run_test.py -i test_ops_jit error like this. ValueError: option names {'--junit-xml-reruns'} already added,https://github.com/pytorch/pytorch/issues/96735,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-03-16T03:04:43Z
issue,69741,JIT: Support for `torch.autograd.functional.jacobian` in TorchScript,https://github.com/pytorch/pytorch/issues/69741,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-03-15T23:07:06Z
issue,96420,Add location information when exception are thrown in `torch.jit.annotations.try_ann_to_type`,https://github.com/pytorch/pytorch/issues/96420,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-03-14T21:13:44Z
issue,95960,`@torch.jit.unused` does not properly ignore unsupported function signature,https://github.com/pytorch/pytorch/issues/95960,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-03-14T18:11:14Z
issue,80007,when forward use **kwargs，how to construct the example_ Inputs parameter in jit.trace?,https://github.com/pytorch/pytorch/issues/80007,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-03-11T03:33:15Z
issue,79875,"A/libc: Fatal signal 6 (SIGABRT), code -6 in tid 11742 (objectdetection)",https://github.com/pytorch/pytorch/issues/79875,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit, oncall: mobile",categorize and decide next maintainer owner/action,2023-03-09T11:53:43Z
issue,76181,Allow `torch.fx` tracing on TorchScript models,https://github.com/pytorch/pytorch/issues/76181,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: fx, oncall: jit",categorize and decide next maintainer owner/action,2023-03-08T16:28:19Z
issue,95789,drastic speed regression of torch.jit.load starting with the 20230301 nightly,https://github.com/pytorch/pytorch/issues/95789,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-03-08T09:32:49Z
issue,96098,Error during inference on iOS: INTERNAL ASSERT FAILED at it_type_base.h:535,https://github.com/pytorch/pytorch/issues/96098,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-03-07T02:11:46Z
issue,66547,"destroy_process_group() does not clear worker count, leading to ""Timed out initializing process group in store based barrier""",https://github.com/pytorch/pytorch/issues/66547,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,categorize and decide next maintainer owner/action,2023-03-03T19:14:16Z
issue,75336,Libtorch Crash after Attempting to Evaluate Model after Opening an OpenGL Context,https://github.com/pytorch/pytorch/issues/75336,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-02-28T19:25:36Z
issue,95497,Import parameters from jit,https://github.com/pytorch/pytorch/issues/95497,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-02-28T16:25:08Z
issue,95560,torch.jit.load documentation doesn't specify if it is safe to load untrusted models or not,https://github.com/pytorch/pytorch/issues/95560,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: docs, oncall: jit",categorize and decide next maintainer owner/action,2023-02-27T16:28:02Z
issue,89868,No pytorch_jni.dll file in libtorch 1.13.0 lib folder,https://github.com/pytorch/pytorch/issues/89868,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: java,categorize and decide next maintainer owner/action,2023-02-25T00:26:03Z
issue,78483,is the issue resolved? windows not pytorch_jni in path,https://github.com/pytorch/pytorch/issues/78483,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: java,categorize and decide next maintainer owner/action,2023-02-24T22:00:50Z
issue,90374,torch.utils.tensorboard import fails if a new protobuf > 3.20 is installed (bug in tensorboard/tensorflow but better guard against it),https://github.com/pytorch/pytorch/issues/90374,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: visualization,categorize and decide next maintainer owner/action,2023-02-23T01:02:30Z
issue,95194,High Cuda Memory Consumption for Simple ResNet50 Inference,https://github.com/pytorch/pytorch/issues/95194,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-02-22T22:55:47Z
issue,36910,jit.script leads to RuntimeError: attribute lookup is not defined on python value of type in some cases,https://github.com/pytorch/pytorch/issues/36910,needs triage,47,triage signal present; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-02-15T14:25:30Z
issue,50533,Non-avairable nightly version for converting models to NNAPI,https://github.com/pytorch/pytorch/issues/50533,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: mobile,categorize and decide next maintainer owner/action,2023-02-08T17:03:06Z
issue,92740,[JIT] Applying `conv2d` over Constants Leads to Exception,https://github.com/pytorch/pytorch/issues/92740,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-02-08T06:56:27Z
issue,93350,Tensorboard SummaryWriter with cloud storage does not work on Mac,https://github.com/pytorch/pytorch/issues/93350,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: visualization,categorize and decide next maintainer owner/action,2023-02-07T10:06:51Z
issue,93826,torch.jit.script does not work with DataParallel,https://github.com/pytorch/pytorch/issues/93826,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-02-03T15:54:36Z
issue,90608,Operator overload priority should not rely on static initialization order,https://github.com/pytorch/pytorch/issues/90608,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-02-02T21:34:56Z
issue,92812,"torch.fx fails to trace through ""+"" op between torch.Size and torch.fx.proxy.Proxy",https://github.com/pytorch/pytorch/issues/92812,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: fx,categorize and decide next maintainer owner/action,2023-02-02T03:05:22Z
issue,93255,MaskRCNN model loaded fail with torch::jit::load(model_path) (C++ API),https://github.com/pytorch/pytorch/issues/93255,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-01-31T21:23:47Z
issue,72498,torch.jit.script + torch.split + onnxruntime incompatibility,https://github.com/pytorch/pytorch/issues/72498,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-01-30T11:37:34Z
issue,84340,Floordiv is deprecated.,https://github.com/pytorch/pytorch/issues/84340,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-01-27T20:27:16Z
issue,93002,Replace pattern fails on incompatible function arguments,https://github.com/pytorch/pytorch/issues/93002,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: fx,categorize and decide next maintainer owner/action,2023-01-27T09:17:40Z
issue,92910,torch.jit.save() generates different contents in a file among different endian machines,https://github.com/pytorch/pytorch/issues/92910,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: power, oncall: jit",categorize and decide next maintainer owner/action,2023-01-27T08:46:23Z
issue,91293,torch.fx tracer emits type error when tracing module that directly contains and uses the torch.cat() function,https://github.com/pytorch/pytorch/issues/91293,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: fx,categorize and decide next maintainer owner/action,2023-01-26T13:10:04Z
issue,92888,[LibTorch] pickle_save output cannot be reloaded using pickle_load in Windows,https://github.com/pytorch/pytorch/issues/92888,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-01-25T06:37:25Z
issue,92866,JIT Function Fails when run a second time,https://github.com/pytorch/pytorch/issues/92866,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-01-25T06:26:38Z
issue,92742,[JIT] Consecutive use of `addmm` Leads to Exception,https://github.com/pytorch/pytorch/issues/92742,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-01-23T15:22:14Z
issue,51445,Saving embedding vectors as TSV for Tensorboard is not efficient,https://github.com/pytorch/pytorch/issues/51445,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: visualization,categorize and decide next maintainer owner/action,2023-01-22T12:15:56Z
issue,92674,[JIT] `Linear` + `BatchNorm2d` Trigger Inconsistency between Eager Mode and JIT,https://github.com/pytorch/pytorch/issues/92674,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-01-20T17:21:23Z
issue,92375,operations failed in TorchScript interpreter,https://github.com/pytorch/pytorch/issues/92375,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-01-20T01:47:03Z
issue,92563,[JIT] INTERNAL ASSERT FAILED when `Conv2d` and `clamp` used together,https://github.com/pytorch/pytorch/issues/92563,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-01-20T01:42:14Z
issue,92548,[JIT] Inconsistency in tensor shape between eager mode and JIT,https://github.com/pytorch/pytorch/issues/92548,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-01-20T00:42:58Z
issue,92558,"[JIT][TracingCheckError] inplace ops incompatible with `contiguous(.., channels_last)`",https://github.com/pytorch/pytorch/issues/92558,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-01-20T00:28:22Z
issue,92654,Traced model output differs on C++ and Python,https://github.com/pytorch/pytorch/issues/92654,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-01-20T00:27:46Z
issue,92230,Add torch::jit::ScriptModule to the C++ API,https://github.com/pytorch/pytorch/issues/92230,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-01-18T13:11:31Z
issue,91396,[JIT] Zero-channel conv2d cannot be applied with `optimize_for_inference`,https://github.com/pytorch/pytorch/issues/91396,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-01-18T06:53:01Z
issue,92175,test_fx_passes generate bad test names,https://github.com/pytorch/pytorch/issues/92175,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: tests, oncall: fx",categorize and decide next maintainer owner/action,2023-01-13T21:46:46Z
issue,86662,libtorch throws `required keyword attribute 'profiled_view_size' has the wrong type` on Linux,https://github.com/pytorch/pytorch/issues/86662,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-01-11T19:02:43Z
issue,91098,[TorchScript] Failed to Forward Correct Number of Arguments to Different Functions,https://github.com/pytorch/pytorch/issues/91098,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-01-11T06:54:05Z
issue,91903,Profiling with stack enabled results in error when Python's cProfile is also running,https://github.com/pytorch/pytorch/issues/91903,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: profiler,categorize and decide next maintainer owner/action,2023-01-10T14:01:54Z
issue,91738,tracing torchvision detection model results in an error,https://github.com/pytorch/pytorch/issues/91738,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: vision, oncall: jit",categorize and decide next maintainer owner/action,2023-01-06T04:02:42Z
issue,91686,Sparse tensor not supported (Minkowski Engine),https://github.com/pytorch/pytorch/issues/91686,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-01-05T23:47:02Z
issue,64492,AttributeError: Can't pickle local object 'schedule.<locals>.schedule_fn',https://github.com/pytorch/pytorch/issues/64492,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: profiler,categorize and decide next maintainer owner/action,2023-01-04T16:15:19Z
issue,91358,[JIT] .backward() not supported by JIT trace,https://github.com/pytorch/pytorch/issues/91358,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-01-03T02:12:33Z
issue,91562,`torch::jit::optimize_for_inference` doesn't preserve exported methods when calling `freeze`,https://github.com/pytorch/pytorch/issues/91562,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-01-03T02:06:20Z
issue,91504,torch.jit.script ERR: RuntimeError: Can't redefine method: forward on class: __torch__.SoSadModule,https://github.com/pytorch/pytorch/issues/91504,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-01-03T01:42:58Z
issue,69352,I want to know how to read the LMDB file once when using DDP,https://github.com/pytorch/pytorch/issues/69352,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: dataloader, oncall: distributed",categorize and decide next maintainer owner/action,2022-12-29T14:32:17Z
issue,89241,"Can't use JIT modules traced with AMP autocast, with Triton Server (or any C++ environment) - freeze() issue ?",https://github.com/pytorch/pytorch/issues/89241,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-12-29T03:42:56Z
issue,91360,Offer `get_buffer` and `get_submodule` in `ScriptModule`?,https://github.com/pytorch/pytorch/issues/91360,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-12-27T23:29:05Z
issue,91229,JIT mishandles torch.round() in PyTorch 1.10.1,https://github.com/pytorch/pytorch/issues/91229,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-12-22T21:23:05Z
issue,91311,Profiler is not properly recording seq number when any key above autograd is used,https://github.com/pytorch/pytorch/issues/91311,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: profiler,categorize and decide next maintainer owner/action,2022-12-22T15:30:42Z
issue,69006,[JIT] allow `Optional[torch.Generator]` be tracable in Python,https://github.com/pytorch/pytorch/issues/69006,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-12-20T12:35:28Z
issue,90915,Torch SummaryWriter import fails with torch 2.0 with an error on numpy.object,https://github.com/pytorch/pytorch/issues/90915,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: tensorboard, oncall: visualization",categorize and decide next maintainer owner/action,2022-12-15T22:30:19Z
issue,90305,Inconsistent Hash of IValue between aten/src/ATen/core/ivalue.cpp and aten/src/ATen/core/Dict_inl.h,https://github.com/pytorch/pytorch/issues/90305,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-12-12T18:12:45Z
issue,90367,[JIT] INTERNAL ASSERT FAILED `torch.add` with boolean primitive constant,https://github.com/pytorch/pytorch/issues/90367,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-12-09T16:51:57Z
issue,90366,[JIT] INTERNAL ASSERT FAILED `torch.mul` with boolean primitive constant,https://github.com/pytorch/pytorch/issues/90366,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-12-09T16:51:52Z
issue,90369,[JIT] Wrong type inference leads to misleading error message,https://github.com/pytorch/pytorch/issues/90369,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-12-09T16:51:44Z
issue,90466,Saving a scripted module to a buffer does not work.,https://github.com/pytorch/pytorch/issues/90466,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-12-09T16:12:12Z
issue,67902,torch::jit::pickle_load is unable to read some files saved by torch.save,https://github.com/pytorch/pytorch/issues/67902,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-12-09T03:42:03Z
issue,90301,Unknown buildin op: aten::pad,https://github.com/pytorch/pytorch/issues/90301,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-12-07T03:15:47Z
issue,90181,Graph is renamed in torch.jit,https://github.com/pytorch/pytorch/issues/90181,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-12-06T18:41:26Z
issue,37883,Support @property decorator in TorchScript,https://github.com/pytorch/pytorch/issues/37883,needs triage,47,triage signal present; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-12-05T20:07:28Z
issue,21064,[JIT] Expose subgraph execution for intermediate output extraction,https://github.com/pytorch/pytorch/issues/21064,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-12-03T08:06:41Z
issue,58109,torch.jit.trace memory leak,https://github.com/pytorch/pytorch/issues/58109,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-12-02T16:23:39Z
issue,89076,Transformers model tracing not working,https://github.com/pytorch/pytorch/issues/89076,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-11-30T20:39:50Z
issue,68385,ImportError: cannot import name 'ProcessGroup' from 'torch.distributed',https://github.com/pytorch/pytorch/issues/68385,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: macos, oncall: distributed",categorize and decide next maintainer owner/action,2022-11-30T19:05:28Z
issue,71261,The jit model will fail when calling the torch.autograd.functional.jacobian with multiple inputs and setting the vectorize to true.,https://github.com/pytorch/pytorch/issues/71261,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, oncall: jit",categorize and decide next maintainer owner/action,2022-11-25T09:39:04Z
issue,89483,Unexpected behavior from torchscript (mixing trace with script),https://github.com/pytorch/pytorch/issues/89483,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-11-22T15:34:14Z
issue,89484,"kind_.is_prim() INTERNAL ASSERT FAILED at ""../torch/csrc/jit/ir/ir.cpp"":1098",https://github.com/pytorch/pytorch/issues/89484,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: cpp, oncall: jit",categorize and decide next maintainer owner/action,2022-11-22T15:13:35Z
issue,84304,op for aten::bitwise_and during torch.jit.trace,https://github.com/pytorch/pytorch/issues/84304,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-11-16T16:02:55Z
issue,89065,Improve clarity of meaning of `torch.jit.trace`'s `example_inputs`,https://github.com/pytorch/pytorch/issues/89065,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-11-15T23:08:28Z
issue,86791,We don't have an op for vulkan_prepack::conv2d_clamp_prepack but it isn't a special case.,https://github.com/pytorch/pytorch/issues/86791,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: vulkan, oncall: mobile",categorize and decide next maintainer owner/action,2022-11-14T08:54:08Z
issue,88765,torch.jit.trace() - AttributeError: 'NoneType' object has no attribute '__module__,https://github.com/pytorch/pytorch/issues/88765,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-11-14T01:37:48Z
issue,88735,RuntimeError: method '__torch__.___torch_mangle_0.MyModule.sin' already defined.,https://github.com/pytorch/pytorch/issues/88735,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: package/deploy,categorize and decide next maintainer owner/action,2022-11-10T16:42:08Z
issue,75673,Tensorboard Issue with visualizing the connections of encoder-decoder network,https://github.com/pytorch/pytorch/issues/75673,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: visualization,categorize and decide next maintainer owner/action,2022-11-10T08:10:23Z
issue,67972,Distributed.TCPStore,https://github.com/pytorch/pytorch/issues/67972,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,categorize and decide next maintainer owner/action,2022-11-08T16:59:29Z
issue,88574,Add support for `torch.Generator` in the FX IR,https://github.com/pytorch/pytorch/issues/88574,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: fx,categorize and decide next maintainer owner/action,2022-11-08T09:53:21Z
issue,39250,SummaryWriter `add_hparams` should support adding new hyperparameters,https://github.com/pytorch/pytorch/issues/39250,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: visualization,categorize and decide next maintainer owner/action,2022-11-06T21:31:17Z
issue,88491,torch.set_grad_enabled results in RuntimeError with torch.jit.script,https://github.com/pytorch/pytorch/issues/88491,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-11-04T21:30:15Z
issue,88448,linear mm weight and bias dtypes mismatch bypasses,https://github.com/pytorch/pytorch/issues/88448,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-11-04T21:28:16Z
issue,88301,1.12.1 incompatible with c++ built for 1.12.0 and vice versa,https://github.com/pytorch/pytorch/issues/88301,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: cpp, oncall: jit",categorize and decide next maintainer owner/action,2022-11-02T15:28:36Z
issue,51376,Huggingface model not being torch scriptable,https://github.com/pytorch/pytorch/issues/51376,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-10-31T09:49:48Z
issue,86572,Instructions for Selective Build for Mobile Linux Platform,https://github.com/pytorch/pytorch/issues/86572,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: mobile,categorize and decide next maintainer owner/action,2022-10-31T00:35:52Z
issue,71277,JIT support for `torch.__version__` & str comparison operations,https://github.com/pytorch/pytorch/issues/71277,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, oncall: jit",categorize and decide next maintainer owner/action,2022-10-29T20:26:17Z
issue,87781,`torchtyping` annotations make saving to Torchscript fail,https://github.com/pytorch/pytorch/issues/87781,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-10-27T14:24:31Z
issue,87787,`BatchNorm` a 0-shape tensor will crash in JIT trace w/o profiling executor on cuda,https://github.com/pytorch/pytorch/issues/87787,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-10-27T14:23:06Z
issue,48452,After updating to Xcode 12 and LibTorch to 1.7.0. Facing issue when running unit test.,https://github.com/pytorch/pytorch/issues/48452,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: arm, module: ios, oncall: jit, oncall: mobile",categorize and decide next maintainer owner/action,2022-10-27T12:19:33Z
issue,87792,`AvgPool` and `MaxPool` will crash in JIT w/o profiling executor,https://github.com/pytorch/pytorch/issues/87792,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-10-26T17:52:05Z
issue,87487,[JIT] Inconsistent handling of tracing dict output leads to assertion,https://github.com/pytorch/pytorch/issues/87487,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-10-23T19:26:17Z
issue,60333,Two-element ModuleList results in error in inference_mode when jit'ed,https://github.com/pytorch/pytorch/issues/60333,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-10-23T06:58:16Z
issue,87268,register_package has no further documentation,https://github.com/pytorch/pytorch/issues/87268,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: docs, oncall: package/deploy",categorize and decide next maintainer owner/action,2022-10-21T13:03:35Z
issue,87390,`chunk` a 0-dim tensor will crash in JIT script w/o profiling executor,https://github.com/pytorch/pytorch/issues/87390,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-10-21T12:55:27Z
issue,85513,"Measure impact of JIT decompositions, reconsider the design",https://github.com/pytorch/pytorch/issues/85513,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-10-20T21:29:31Z
issue,87289,'str' object has no attribute '__module__' in jit is_final,https://github.com/pytorch/pytorch/issues/87289,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-10-20T14:45:54Z
issue,77167,nn.functional.pad accepts bool values but raises internal assert when converted to JIT,https://github.com/pytorch/pytorch/issues/77167,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-10-20T05:50:10Z
issue,86818,Forward hooks for ScriptModules,https://github.com/pytorch/pytorch/issues/86818,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-10-18T01:45:48Z
issue,84321,[Profiler] Snapshot CudaCachingAllocator on profile begin,https://github.com/pytorch/pytorch/issues/84321,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, oncall: profiler",categorize and decide next maintainer owner/action,2022-10-15T17:56:39Z
issue,86816,JIT model returns different value on cpu with uniform-initialized input,https://github.com/pytorch/pytorch/issues/86816,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-10-13T17:17:32Z
issue,86804,JIT model will have a different jacobian after the first computation,https://github.com/pytorch/pytorch/issues/86804,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-10-13T17:16:31Z
issue,85877,JIT model could return 'NaN' gradient after the first execution,https://github.com/pytorch/pytorch/issues/85877,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-10-12T17:35:53Z
issue,86552,`max_unpool` and `max_pool` will trigger INTERNAL ASSERT FAIL in JIT,https://github.com/pytorch/pytorch/issues/86552,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-10-12T16:37:29Z
issue,63725,[Feature Request] add graph to hparams for tensorboard,https://github.com/pytorch/pytorch/issues/63725,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: tensorboard, oncall: visualization",categorize and decide next maintainer owner/action,2022-10-11T15:16:07Z
issue,86287,JIT `lgamma` will return `inf` only with dual input in forward mode,https://github.com/pytorch/pytorch/issues/86287,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-10-10T18:22:20Z
issue,86616,JIT returns different values for a model on cuda and returns a strange error message on cpu,https://github.com/pytorch/pytorch/issues/86616,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), oncall: jit",categorize and decide next maintainer owner/action,2022-10-10T17:22:53Z
issue,86598,`squeeze_` fails with JIT but succeeds without it,https://github.com/pytorch/pytorch/issues/86598,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-10-10T17:06:11Z
issue,86597,JIT returns different values for `cos + frac` on cpu,https://github.com/pytorch/pytorch/issues/86597,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), oncall: jit",categorize and decide next maintainer owner/action,2022-10-10T15:08:59Z
issue,86553,`topk` will return the wrong value and could read out-of-bound value after jit,https://github.com/pytorch/pytorch/issues/86553,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-10-10T14:52:35Z
issue,86455,(JIT) x:Optional[T] cannot not expect content type after `if x is None or x.shape[0]==1`,https://github.com/pytorch/pytorch/issues/86455,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-10-08T04:23:57Z
issue,86351,Jetson JIT: Memory Leak on inference after optimize_for_inference,https://github.com/pytorch/pytorch/issues/86351,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-10-06T05:51:11Z
issue,86276,JIT miss the argument `as_tuple` for API `nonzero`,https://github.com/pytorch/pytorch/issues/86276,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-10-05T16:09:48Z
issue,81337,JIT trace takes forever on a simple method,https://github.com/pytorch/pytorch/issues/81337,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-10-04T05:02:46Z
issue,86124,torch.jit.trace throwing Invalid name for qualified name eror,https://github.com/pytorch/pytorch/issues/86124,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-10-03T16:09:30Z
issue,47336,[jit] Properly support slicing of ModuleList,https://github.com/pytorch/pytorch/issues/47336,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-10-03T16:09:12Z
issue,86116,JIT fails to trace binary cross entropy with a strange error msg,https://github.com/pytorch/pytorch/issues/86116,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-10-03T15:18:54Z
issue,86112,`cdist` should succeed when `p` is integer in JIT,https://github.com/pytorch/pytorch/issues/86112,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-10-03T13:19:09Z
issue,86110,JIT return a tensor with different datatype from the tensor w/o gradient and normal function,https://github.com/pytorch/pytorch/issues/86110,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-10-03T12:50:01Z
issue,79003,Libtorch C++ mobile build linking error,https://github.com/pytorch/pytorch/issues/79003,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: build, oncall: mobile",categorize and decide next maintainer owner/action,2022-10-01T06:41:03Z
issue,85706,RuntimeError: [enforce fail at CPUAllocator.cpp:68] . DefaultCPUAllocator: can't allocate memory: you tried to allocate 4860032 bytes. Error code 12 (Cannot allocate memory),https://github.com/pytorch/pytorch/issues/85706,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-09-29T17:39:33Z
issue,85893,TorchScript does not recognize mix-in types with `Enum`,https://github.com/pytorch/pytorch/issues/85893,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-09-29T14:14:47Z
issue,85792,TorchScript causes range_check error after a few iterations of forward-backward passes,https://github.com/pytorch/pytorch/issues/85792,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-09-29T08:18:01Z
issue,85149,Pytorch on iOS (iPhone X & XR) throwing can't allocate memory exception. Ref Logs:,https://github.com/pytorch/pytorch/issues/85149,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: ios, oncall: mobile",categorize and decide next maintainer owner/action,2022-09-29T05:31:15Z
issue,84681,JIT will affect the gradient computation of forward mode,https://github.com/pytorch/pytorch/issues/84681,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-09-28T16:41:17Z
issue,85813,TorchScript error for `Enum` inside a module,https://github.com/pytorch/pytorch/issues/85813,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-09-28T15:16:04Z
issue,85805,"`vector_norm` will trigger ""Tracing failed sanity checks"" for JIT when ord is boolean tensor",https://github.com/pytorch/pytorch/issues/85805,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-09-28T14:08:50Z
issue,85804,JIT fails to trace `sparse.mm` with a strange error,https://github.com/pytorch/pytorch/issues/85804,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-09-28T14:01:03Z
issue,69616,Multi-GPU training stuck when using grad scaler,https://github.com/pytorch/pytorch/issues/69616,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision), module: cuda, oncall: distributed",categorize and decide next maintainer owner/action,2022-09-28T05:28:40Z
issue,44951,copy.deepcopy not working properly for jit.TopLevelTracedModule,https://github.com/pytorch/pytorch/issues/44951,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-09-27T15:02:27Z
issue,85588,`linalg.norm` cannot compute the grad in forward mode after script,https://github.com/pytorch/pytorch/issues/85588,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-09-26T19:52:07Z
issue,85505,The reload model has different (and strange) forward computation from original model with `LSTMCell`,https://github.com/pytorch/pytorch/issues/85505,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-09-26T15:11:34Z
issue,85514,The reload `MultiLabelMarginLoss` will have different gradients on cuda,https://github.com/pytorch/pytorch/issues/85514,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-09-26T15:11:20Z
issue,85585,`as_tensor` will return a different dtype with script,https://github.com/pytorch/pytorch/issues/85585,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-09-26T15:03:24Z
issue,85573,`jit` could make some undifferentiable APIs differentiable,https://github.com/pytorch/pytorch/issues/85573,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-09-23T23:38:23Z
issue,85570,`mvlgamma_` will fail when compiling with trace `jit`,https://github.com/pytorch/pytorch/issues/85570,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-09-23T21:29:50Z
issue,85568,`detach_` behaves differently when computing the gradients in forward mode w/ `jit`,https://github.com/pytorch/pytorch/issues/85568,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-09-23T20:50:28Z
issue,85342,"JIT fuser issues with {ceil,floor,round,trunc}(int8)",https://github.com/pytorch/pytorch/issues/85342,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-09-23T17:40:48Z
issue,83250,Using Pytorch and Mapbox in the same project,https://github.com/pytorch/pytorch/issues/83250,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: mobile,categorize and decide next maintainer owner/action,2022-09-15T21:03:18Z
issue,85072,Segmentation fault in `torch.jit.wait`,https://github.com/pytorch/pytorch/issues/85072,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-09-15T06:02:53Z
issue,61344,Get TensorType's device in python,https://github.com/pytorch/pytorch/issues/61344,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, oncall: jit",categorize and decide next maintainer owner/action,2022-09-14T01:20:02Z
issue,84685,[TensorExpr] applying `rfactor` for a Mul Reducer with init value different than 1 results in wrong results,https://github.com/pytorch/pytorch/issues/84685,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-09-13T21:46:41Z
issue,84495,functionalize: Does not compose cleanly with torch.jit.script/torch.jit.trace,https://github.com/pytorch/pytorch/issues/84495,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: functionalization, oncall: jit",categorize and decide next maintainer owner/action,2022-09-13T00:37:34Z
issue,84673,Autograd will take `init` module API into account when using `jit`,https://github.com/pytorch/pytorch/issues/84673,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-09-12T21:56:56Z
issue,84628,INTERNAL ASSERT when the type of argument is not considered in JIT,https://github.com/pytorch/pytorch/issues/84628,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-09-12T21:54:42Z
issue,84615,Serialize the warmed up torchscript module,https://github.com/pytorch/pytorch/issues/84615,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-09-12T20:41:09Z
issue,84588,torch.jit.script IndentationError: unexpected indent,https://github.com/pytorch/pytorch/issues/84588,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-09-11T16:03:15Z
issue,61966,Bootcamp Task: Add `prim::to_mkldnn` to convert from aten's nchw to mkldnn's nChw8c directly,https://github.com/pytorch/pytorch/issues/61966,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, oncall: jit",categorize and decide next maintainer owner/action,2022-09-10T18:56:54Z
issue,84646,JIT script calculation/dtype inconsistent depending on operator expression,https://github.com/pytorch/pytorch/issues/84646,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-09-07T16:59:45Z
issue,84055,scripted fasterRCNN model cannot be loaded with libtorch c++ API,https://github.com/pytorch/pytorch/issues/84055,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: cpp, oncall: jit",categorize and decide next maintainer owner/action,2022-08-31T11:09:21Z
issue,46375,JIT: error in LSTM with `flatten_parameters`,https://github.com/pytorch/pytorch/issues/46375,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-08-30T07:44:51Z
issue,84189,"RuntimeError: outputs_[i]->uses().empty() INTERNAL ASSERT FAILED at /pytorch/torch/csrc/jit/ir.cpp:1027, please report a bug to PyTorch. (eraseOutput at /pytorch/torch/csrc/jit/ir.cpp:1027)",https://github.com/pytorch/pytorch/issues/84189,needs triage,47,triage signal present; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-08-30T02:49:56Z
issue,81635,Windows Debug binaries crash on forward: assert fail on IListRefIterator destructor,https://github.com/pytorch/pytorch/issues/81635,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-08-29T11:21:57Z
issue,84187,[jit] WithInsertPoint can't get back to the prev_ node if the prev_ node has been destroyed,https://github.com/pytorch/pytorch/issues/84187,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-08-29T06:58:01Z
issue,84135,[jit] ignored method calling static method results in an error,https://github.com/pytorch/pytorch/issues/84135,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-08-26T18:52:02Z
issue,84049,TorchScript unsupport tuple unpacking as function inputs.,https://github.com/pytorch/pytorch/issues/84049,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-08-25T10:45:59Z
issue,82598,Deep copy models with `create_feature_extractor` produces different parameters,https://github.com/pytorch/pytorch/issues/82598,needs triage,47,"triage signal present; old update age or low-urgency/stale label; component/oncall labels: module: vision, oncall: fx",categorize and decide next maintainer owner/action,2022-08-22T20:02:20Z
issue,62381,torch.distributed and subprocess do not work together?,https://github.com/pytorch/pytorch/issues/62381,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,categorize and decide next maintainer owner/action,2022-08-22T13:10:12Z
issue,83769,[TorchTidy] Check if `set_to_none` would change optimizer semantics.,https://github.com/pytorch/pytorch/issues/83769,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: profiler,categorize and decide next maintainer owner/action,2022-08-19T20:49:21Z
issue,76013,test_jit.py TestWarn.test_warn and friends doesn't work under pytest,https://github.com/pytorch/pytorch/issues/76013,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-08-19T16:05:20Z
issue,83551,OpenJDK libtorch_cpu.so stack guard warning,https://github.com/pytorch/pytorch/issues/83551,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: java,categorize and decide next maintainer owner/action,2022-08-17T14:18:30Z
issue,83589,Speedup for adding images to tensorboard,https://github.com/pytorch/pytorch/issues/83589,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: visualization,categorize and decide next maintainer owner/action,2022-08-17T14:16:16Z
issue,83585,Segfault when profiling with_stack=True on model with jit.optimize_for_inference,https://github.com/pytorch/pytorch/issues/83585,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-08-17T11:43:27Z
issue,28206,torch.utils.tensorboard.SummaryWriter.add_graph do not support non-tensor inputs,https://github.com/pytorch/pytorch/issues/28206,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: visualization,categorize and decide next maintainer owner/action,2022-08-14T10:43:52Z
issue,83349,checkpoint function is not jit compatible,https://github.com/pytorch/pytorch/issues/83349,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-08-12T20:25:16Z
issue,83143,Bug in building pytorch deploy from source in macos USE_DEPLOY=1,https://github.com/pytorch/pytorch/issues/83143,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: package/deploy,categorize and decide next maintainer owner/action,2022-08-11T21:56:02Z
issue,83161,Optimize for mobile metal model,https://github.com/pytorch/pytorch/issues/83161,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: mobile,categorize and decide next maintainer owner/action,2022-08-11T21:54:06Z
issue,75903,torch.jit.script'd function very slow on first invocation on latest nightly,https://github.com/pytorch/pytorch/issues/75903,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: nvfuser, oncall: jit",categorize and decide next maintainer owner/action,2022-08-11T17:28:29Z
issue,82951,torch.profiler's FLOPs measure only counts operations involving '+' and '*' .,https://github.com/pytorch/pytorch/issues/82951,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: profiler,categorize and decide next maintainer owner/action,2022-08-10T13:23:47Z
issue,46460,C++ API: Training and inference of torchscript modules on multiple GPU,https://github.com/pytorch/pytorch/issues/46460,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-08-10T13:02:56Z
issue,83045,[JIT] Scripting modules fails for modules that contain nested NamedTuples,https://github.com/pytorch/pytorch/issues/83045,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-08-09T01:11:59Z
issue,82872,Unknown builtin op: torchvision::deform_conv2d,https://github.com/pytorch/pytorch/issues/82872,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-08-08T22:27:43Z
issue,82762,Fill in a bool Tensor not supported in jit,https://github.com/pytorch/pytorch/issues/82762,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-08-03T22:37:51Z
issue,81460,[JIT] Request Constant Propagation to keep fake_quantize_per_tensor_affine and fake_quantize_per_channel_affine on the graph,https://github.com/pytorch/pytorch/issues/81460,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-08-02T21:42:02Z
issue,82635,[Torchscript] torch.min returns wrong gradient when inputs are equal,https://github.com/pytorch/pytorch/issues/82635,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-08-02T05:16:58Z
issue,82634,[Torchscript] some activations backward are not fused when used with linear,https://github.com/pytorch/pytorch/issues/82634,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-08-02T05:11:23Z
issue,69931,[JIT] torch.exp roughly 20 times slower in TorchScript vs. PyTorch,https://github.com/pytorch/pytorch/issues/69931,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-08-01T14:03:29Z
issue,82451,add support for bitwise operations with floating point numbers,https://github.com/pytorch/pytorch/issues/82451,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-07-31T18:02:49Z
issue,62313,"On Triton inference server, the TorchScript .pt file, exported by 1.9.0, serves unsuccessfully",https://github.com/pytorch/pytorch/issues/62313,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-07-29T07:43:37Z
issue,68256,RuntimeError: CUDA error: initialization error when calling torch.distributed.init_process_group using torch multiprocessing,https://github.com/pytorch/pytorch/issues/68256,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,categorize and decide next maintainer owner/action,2022-07-29T00:16:24Z
issue,81651,optimize_for_mobile has an issue with constant operations at the end of a loop,https://github.com/pytorch/pytorch/issues/81651,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: mobile,categorize and decide next maintainer owner/action,2022-07-27T22:12:07Z
issue,82324,[JIT] SchemaInfo warning appears out in the wild,https://github.com/pytorch/pytorch/issues/82324,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-07-27T15:47:50Z
issue,30365,TorchScript Performance: 150x gap between TorchScript and Native Python,https://github.com/pytorch/pytorch/issues/30365,needs triage,47,triage signal present; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-07-26T23:00:36Z
issue,82216,iOS TestApp from mobile performance recipes tutorial doesn't build on macOS,https://github.com/pytorch/pytorch/issues/82216,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: ios, oncall: mobile",categorize and decide next maintainer owner/action,2022-07-26T16:49:11Z
issue,60844,DDP fails if you have multiple forward passes and a single backwards pass with `find_unused_parameters=True`,https://github.com/pytorch/pytorch/issues/60844,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: ddp, oncall: distributed",categorize and decide next maintainer owner/action,2022-07-26T14:55:15Z
issue,82084,Untangle TorchScript prim ops in aten namespace,https://github.com/pytorch/pytorch/issues/82084,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-07-25T03:30:32Z
issue,67570,RFC: Overlap optimizer computation with DDP/FSDP backward,https://github.com/pytorch/pytorch/issues/67570,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: ddp, oncall: distributed",categorize and decide next maintainer owner/action,2022-07-19T02:45:34Z
issue,81245,"Libtorch cannot load TrochScript Module correctly, when a network contains conv2d(inchannels=64, outchannels=128, kernelsize=1) .",https://github.com/pytorch/pytorch/issues/81245,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-07-12T16:48:50Z
issue,65305,Add support for HF NLP models in torch.package/deploy,https://github.com/pytorch/pytorch/issues/65305,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: package/deploy,categorize and decide next maintainer owner/action,2022-07-11T15:46:27Z
issue,81100,[jit] Failed to load a saved scripted function,https://github.com/pytorch/pytorch/issues/81100,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-07-08T09:47:59Z
issue,80939,[jit] script backward wrong gradient,https://github.com/pytorch/pytorch/issues/80939,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-07-06T05:53:23Z
issue,80829,Can torchscript dump backward graph?,https://github.com/pytorch/pytorch/issues/80829,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-07-06T00:08:47Z
issue,80863,SummaryWriter add_embedding issue with label_img,https://github.com/pytorch/pytorch/issues/80863,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: visualization,categorize and decide next maintainer owner/action,2022-07-05T17:46:36Z
issue,70511,Strange behavior of torch.jit.trace when moving parameters across device,https://github.com/pytorch/pytorch/issues/70511,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-07-04T06:44:43Z
issue,80606,[jit.script] jit.script give uncertain results using torch.half,https://github.com/pytorch/pytorch/issues/80606,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: nvfuser, oncall: jit",categorize and decide next maintainer owner/action,2022-06-30T22:04:37Z
issue,80321,java.lang.ExceptionInInitializerError at org.pytorch.NativePeer.initHybrid(Native Method),https://github.com/pytorch/pytorch/issues/80321,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: mobile,categorize and decide next maintainer owner/action,2022-06-29T06:25:27Z
issue,80142,CPUProfilingAllocator greedy allocation plan generation failed,https://github.com/pytorch/pytorch/issues/80142,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: mobile,categorize and decide next maintainer owner/action,2022-06-28T16:10:58Z
issue,61709,[pytorch android] use Vulkan backend crash,https://github.com/pytorch/pytorch/issues/61709,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: android, module: vulkan, oncall: mobile",categorize and decide next maintainer owner/action,2022-06-22T06:46:29Z
issue,67142,"enumerate(..., start=idx) is not working correctly while scripting",https://github.com/pytorch/pytorch/issues/67142,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-06-20T16:21:50Z
issue,75502,TensorBoard frontend fails to display embeddings when `add_embedding()` writes large `label_img`,https://github.com/pytorch/pytorch/issues/75502,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: tensorboard, oncall: visualization",categorize and decide next maintainer owner/action,2022-06-18T20:12:13Z
issue,79751,[JIT] failures with nested with blocks + loop continuation,https://github.com/pytorch/pytorch/issues/79751,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-06-16T21:48:21Z
issue,79684,Does Torch JIT Support Trace High-level Custom Op?,https://github.com/pytorch/pytorch/issues/79684,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-06-16T06:12:47Z
issue,79375,TorchScript bidirectional lnlstm from example doesn't work,https://github.com/pytorch/pytorch/issues/79375,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-06-14T00:23:31Z
issue,47162,Custon C++ classes are not traceable,https://github.com/pytorch/pytorch/issues/47162,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-06-13T08:47:22Z
issue,79275,testSerializationInterop in test/cpp/jit/torch_python_test.cpp has not run in over two years,https://github.com/pytorch/pytorch/issues/79275,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-06-10T13:50:42Z
issue,68871,[LibTorch-Lite] Add a custom flag to build LibTorch-Lite with LAPACK included,https://github.com/pytorch/pytorch/issues/68871,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: ios, oncall: mobile",categorize and decide next maintainer owner/action,2022-06-08T17:33:22Z
issue,75935,torch.jit.trace error when custom autograd function used in the model,https://github.com/pytorch/pytorch/issues/75935,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-06-08T12:22:54Z
issue,79008,torchscript jit trace support custom op without specific csrc and .so,https://github.com/pytorch/pytorch/issues/79008,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-06-07T11:43:03Z
issue,60971,jit scripting for the parametrizations,https://github.com/pytorch/pytorch/issues/60971,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: nn.utils.parametrize, oncall: jit",categorize and decide next maintainer owner/action,2022-06-06T17:07:00Z
issue,78920,Strange tracing result with torchscript,https://github.com/pytorch/pytorch/issues/78920,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-06-06T07:26:48Z
issue,62300,ProcessGroupGloo CUDA Comm Synchronization Looks Wrong,https://github.com/pytorch/pytorch/issues/62300,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: c10d, oncall: distributed",categorize and decide next maintainer owner/action,2022-06-03T20:47:17Z
issue,77837,TorchScript attempts to compile dead branch of torch.jit.is_scripting,https://github.com/pytorch/pytorch/issues/77837,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-06-03T19:49:44Z
issue,78812,TorchScript inference get intermediate result?,https://github.com/pytorch/pytorch/issues/78812,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-06-03T18:36:16Z
issue,78805,"torch.jit.script segmentation fault (pytorch debayer module) 1.10, 1.11 and nightly",https://github.com/pytorch/pytorch/issues/78805,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-06-03T13:43:08Z
issue,78486,Exception in torch.jit.script doesn't indicate where in the code the problem lies.,https://github.com/pytorch/pytorch/issues/78486,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-06-02T22:39:52Z
issue,78559,[JIT] autodiff implementation of rand_like function is outdated,https://github.com/pytorch/pytorch/issues/78559,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-05-31T19:09:29Z
issue,72220,[JIT] Assert that the autodiff implementation of backward() returns the correct number of values,https://github.com/pytorch/pytorch/issues/72220,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-05-31T18:16:36Z
issue,75280,[JIT] log_extract.py improvements,https://github.com/pytorch/pytorch/issues/75280,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-05-31T17:39:19Z
issue,68987,Operator is not supported in mobile module.,https://github.com/pytorch/pytorch/issues/68987,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: mobile,categorize and decide next maintainer owner/action,2022-05-29T19:33:22Z
issue,64419,Windows Torch.distributed Multi-GPU training with Gloo backend not working,https://github.com/pytorch/pytorch/issues/64419,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,categorize and decide next maintainer owner/action,2022-05-28T11:32:36Z
issue,78172,Allow specifying pickle module for torch.package,https://github.com/pytorch/pytorch/issues/78172,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: package/deploy,categorize and decide next maintainer owner/action,2022-05-25T18:40:03Z
issue,61221,Unable to add a scalar to an MKLDNN tensor,https://github.com/pytorch/pytorch/issues/61221,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: intel, module: mkldnn, oncall: jit",categorize and decide next maintainer owner/action,2022-05-24T11:36:28Z
issue,78133,Enhancements to AliasDB to handle in-place operations,https://github.com/pytorch/pytorch/issues/78133,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-05-23T22:45:24Z
issue,50714,[PyTorch Mobile] Android speed benchmark binary crashes when using vulkan,https://github.com/pytorch/pytorch/issues/50714,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: mobile,categorize and decide next maintainer owner/action,2022-05-23T15:04:37Z
issue,77840,Legacy model format is not supported on mobile,https://github.com/pytorch/pytorch/issues/77840,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: mobile,categorize and decide next maintainer owner/action,2022-05-23T03:27:04Z
issue,76672,C++ CUDA assign existing memory to forward method.,https://github.com/pytorch/pytorch/issues/76672,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-05-18T16:45:39Z
issue,77354,"outputs_[i]->uses().empty()INTERNAL ASSERT FAILED at ""/opt/conda/conda-bld/pytorch_1646755853042/work/torch/csrc/jit/ir/ir.cpp"":1314, please report a bug to PyTorch.",https://github.com/pytorch/pytorch/issues/77354,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-05-12T21:21:43Z
issue,73608,`torch.jit.load` fails when function parameters use non-ASCII characters,https://github.com/pytorch/pytorch/issues/73608,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-05-11T02:38:14Z
issue,67241,"init_process_group hangs for multi-node, Pytorch > v1.3.1 and file init_method",https://github.com/pytorch/pytorch/issues/67241,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: deadlock, module: nccl, oncall: distributed",categorize and decide next maintainer owner/action,2022-05-10T13:25:04Z
issue,54133,[TorchScript] kind_.is_prim() INTERNAL ASSERT FAILED,https://github.com/pytorch/pytorch/issues/54133,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-05-09T15:46:06Z
issue,76578,[JIT][Autocast] Batchnorm folding pass during freezing doesn't preserve types,https://github.com/pytorch/pytorch/issues/76578,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-05-08T20:29:50Z
issue,77027,Kill use of TensorImpl::ShareExternalPointer in torch/csrc/jit/tensorexpr/external_functions.cpp,https://github.com/pytorch/pytorch/issues/77027,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-05-07T17:18:48Z
issue,74799,Irrelevant warning during ONNX export of torch.jit.ScriptFunction: Model has no forward function,https://github.com/pytorch/pytorch/issues/74799,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: onnx, oncall: jit",categorize and decide next maintainer owner/action,2022-05-04T20:00:21Z
issue,76686,[torch::deploy] Remove `manager_` from the constructor and deconstructor of `InterpreterSession`,https://github.com/pytorch/pytorch/issues/76686,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: deploy, oncall: package/deploy",categorize and decide next maintainer owner/action,2022-05-03T20:01:05Z
issue,76683,[torch::deploy] remove reliance on manager_ for unload,https://github.com/pytorch/pytorch/issues/76683,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: package/deploy,categorize and decide next maintainer owner/action,2022-05-03T08:48:11Z
issue,76682,[torch::deploy] Remove manager_ from AquireSession,https://github.com/pytorch/pytorch/issues/76682,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: package/deploy,categorize and decide next maintainer owner/action,2022-05-03T08:48:05Z
issue,76685,[torch::deploy] move create_movable to interpreter_manager,https://github.com/pytorch/pytorch/issues/76685,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: package/deploy,categorize and decide next maintainer owner/action,2022-05-03T08:47:49Z
issue,75956,[JIT] [Autocast] JIT Autocast Pass operations' list should be extendable and consistent with imperative path,https://github.com/pytorch/pytorch/issues/75956,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-05-02T15:24:45Z
issue,76309,More understandable name column of the table of the profiling result.,https://github.com/pytorch/pytorch/issues/76309,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: profiler,categorize and decide next maintainer owner/action,2022-04-30T00:09:36Z
issue,76410,TorchFunction handling and overload resolution very slow in `torch.ops`,https://github.com/pytorch/pytorch/issues/76410,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-04-27T23:14:42Z
issue,76503,[JIT] make IRAttributeError extend jit::Error,https://github.com/pytorch/pytorch/issues/76503,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-04-27T23:11:04Z
issue,72330,Add and Mul torch tensors on Metal (IOS),https://github.com/pytorch/pytorch/issues/72330,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: ios, oncall: mobile",categorize and decide next maintainer owner/action,2022-04-27T18:25:54Z
issue,50943,Exporting opnames fails when using custom classes,https://github.com/pytorch/pytorch/issues/50943,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: mobile,categorize and decide next maintainer owner/action,2022-04-23T18:06:33Z
issue,75794,Different result with JIT,https://github.com/pytorch/pytorch/issues/75794,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-04-19T23:23:39Z
issue,75655,`jit(Function)` results in double execution,https://github.com/pytorch/pytorch/issues/75655,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-04-19T17:43:27Z
issue,76029,Mobile assets upload could break third party mirrors due to binary data size,https://github.com/pytorch/pytorch/issues/76029,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: mobile,categorize and decide next maintainer owner/action,2022-04-19T17:35:28Z
issue,68330,"isObject()INTERNAL ASSERT FAILED at ""../aten/src/ATen/core/ivalue_inl.h"":115, please report a bug to PyTorch. Expected Object but got Tensor",https://github.com/pytorch/pytorch/issues/68330,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: android, oncall: jit, oncall: mobile",categorize and decide next maintainer owner/action,2022-04-16T12:29:45Z
issue,75864,"INTERNAL ASSERT FAILED at ""vulkan_rewrite.cpp"":272",https://github.com/pytorch/pytorch/issues/75864,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: mobile,categorize and decide next maintainer owner/action,2022-04-15T21:45:32Z
issue,58744,Using pytorch 1.1 model on pytorch 1.8 environment,https://github.com/pytorch/pytorch/issues/58744,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-04-13T02:00:28Z
issue,75654,jit fails when trying to assign values to model via hook,https://github.com/pytorch/pytorch/issues/75654,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-04-12T09:17:58Z
issue,72516,test_del (jit.test_builtins.TestBuiltins) fails due to highlight assertions,https://github.com/pytorch/pytorch/issues/72516,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-04-10T00:15:35Z
issue,75309,[jit] failing OpInfo JIT tests for conv1d and complex input,https://github.com/pytorch/pytorch/issues/75309,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-04-08T18:30:13Z
issue,75308,Using Adaptive Model for Pytorch Mobile,https://github.com/pytorch/pytorch/issues/75308,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: mobile,categorize and decide next maintainer owner/action,2022-04-08T04:12:00Z
issue,75467,torch.package: log.info when loading packages w/ externed modules + documentation,https://github.com/pytorch/pytorch/issues/75467,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: package/deploy,categorize and decide next maintainer owner/action,2022-04-07T22:17:28Z
issue,75217,Storing LTC tensor shape information in jit::Value,https://github.com/pytorch/pytorch/issues/75217,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-04-06T18:13:31Z
issue,72112,Unable to parse IR generated by Lazy Tensor Core with TorchScript Backend,https://github.com/pytorch/pytorch/issues/72112,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: lazy, oncall: jit",categorize and decide next maintainer owner/action,2022-04-06T17:25:20Z
issue,75005,Cannot create TorchScript file after converting into a quantized model,https://github.com/pytorch/pytorch/issues/75005,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-04-06T15:28:38Z
issue,46907,SummaryWriter deletes data automatically when there are too many,https://github.com/pytorch/pytorch/issues/46907,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: tensorboard, oncall: visualization",categorize and decide next maintainer owner/action,2022-04-05T16:23:29Z
issue,75038,Generate source location information in TS LTC backend,https://github.com/pytorch/pytorch/issues/75038,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-04-04T00:39:16Z
issue,74964,Add Unit Tests for Torch.Package Subclasses [Waiting on changes to land],https://github.com/pytorch/pytorch/issues/74964,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: package/deploy,categorize and decide next maintainer owner/action,2022-04-01T14:57:34Z
issue,74956,Allow `torch.package` to accept `*.pyc` files,https://github.com/pytorch/pytorch/issues/74956,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: package/deploy,categorize and decide next maintainer owner/action,2022-04-01T14:57:11Z
issue,75002,JIT autocasting fails on Optional[Tensor],https://github.com/pytorch/pytorch/issues/75002,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-03-31T21:44:07Z
issue,65813,[RFC] Low-level speed optimizations for PowerSGD,https://github.com/pytorch/pytorch/issues/65813,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: ddp, oncall: distributed",categorize and decide next maintainer owner/action,2022-03-30T04:15:49Z
issue,74909,torch.package fails to import if dataclass and __future__.annotations present,https://github.com/pytorch/pytorch/issues/74909,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: package/deploy,categorize and decide next maintainer owner/action,2022-03-29T21:29:48Z
issue,73972,Forward_AD and Torchscript Functions results in Nones or wrong values.,https://github.com/pytorch/pytorch/issues/73972,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: forward ad, oncall: jit",categorize and decide next maintainer owner/action,2022-03-29T18:50:49Z
issue,70159,[Android] Unknown builtin op: aten::reflection_pad3d,https://github.com/pytorch/pytorch/issues/70159,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: mobile,categorize and decide next maintainer owner/action,2022-03-27T22:59:52Z
issue,74605,torch.profiler.profile does't work well for CPU model when not using torch.profiler.schedule,https://github.com/pytorch/pytorch/issues/74605,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: profiler,categorize and decide next maintainer owner/action,2022-03-26T00:06:21Z
issue,72659,Add a unit test which uses an extension module + ordered importer,https://github.com/pytorch/pytorch/issues/72659,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: package/deploy,categorize and decide next maintainer owner/action,2022-03-23T16:09:09Z
issue,74589,Test test_ops_jit is taking too much time on ASAN,https://github.com/pytorch/pytorch/issues/74589,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-03-23T00:19:00Z
issue,36718,Support alpha channel in tensorboard.add_figure,https://github.com/pytorch/pytorch/issues/36718,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: tensorboard, oncall: visualization",categorize and decide next maintainer owner/action,2022-03-21T17:07:47Z
issue,74272,[PyTorch] Lightweight dispatch,https://github.com/pytorch/pytorch/issues/74272,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: mobile,categorize and decide next maintainer owner/action,2022-03-16T00:59:01Z
issue,74221,`package.PackageExporter` does not actually appear to have a `file_structure` method,https://github.com/pytorch/pytorch/issues/74221,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: docs, oncall: package/deploy",categorize and decide next maintainer owner/action,2022-03-15T15:30:51Z
issue,72110,Conversion Error in pytorch mobile with metal,https://github.com/pytorch/pytorch/issues/72110,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: memory format, oncall: mobile",categorize and decide next maintainer owner/action,2022-03-15T06:28:17Z
issue,62821,Enable fft support for mobile builds,https://github.com/pytorch/pytorch/issues/62821,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: android, module: arm, module: fft, oncall: mobile",categorize and decide next maintainer owner/action,2022-03-14T11:16:48Z
issue,74095,torchscript RNN modules cannot move between GPU,https://github.com/pytorch/pytorch/issues/74095,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-03-11T18:53:01Z
issue,64688,Backend capability chart for nccl GPU send/recv may be stale,https://github.com/pytorch/pytorch/issues/64688,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: nccl, oncall: distributed",categorize and decide next maintainer owner/action,2022-03-11T17:01:05Z
issue,71929,[JIT] Results are different when saving in TorchScript Format through train/eval mode,https://github.com/pytorch/pytorch/issues/71929,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-03-11T07:20:23Z
issue,69434,"JIT RuntimeError: 'Union[Tensor, List[float], List[int]]' object is not subscriptable",https://github.com/pytorch/pytorch/issues/69434,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-03-10T17:18:48Z
issue,74030,"[JIT] tuple of float is accepted in List[float] but not in Union[float, List[float]]",https://github.com/pytorch/pytorch/issues/74030,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-03-10T16:17:08Z
issue,41417,[jit] support class polymorphism,https://github.com/pytorch/pytorch/issues/41417,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-03-09T20:25:25Z
issue,69308,Enable DDP checkpointing tests for Gloo backend,https://github.com/pytorch/pytorch/issues/69308,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: ddp, oncall: distributed",categorize and decide next maintainer owner/action,2022-03-02T16:54:47Z
issue,73571,conv with padding='same' fails on lazy torchscript nvfuser,https://github.com/pytorch/pytorch/issues/73571,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-03-01T16:49:10Z
issue,70533,JIT / TorchScript should support for sum(List[torch.Tensor]) like non-JIT PyTorch already does,https://github.com/pytorch/pytorch/issues/70533,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-03-01T10:03:29Z
issue,73419,test_jit_cuda_fuser fails on non-CUDA node for 1.11.0rc3,https://github.com/pytorch/pytorch/issues/73419,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-02-25T10:43:25Z
issue,70240,JIT: wrong list/tuple length if using Union,https://github.com/pytorch/pytorch/issues/70240,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-02-19T12:03:23Z
issue,70223,"jit: Confusing behavior w/ torch.autograd.grad, iterative loop, and printing?",https://github.com/pytorch/pytorch/issues/70223,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-02-19T05:02:48Z
issue,73072,Exporting model to TorchScript fails once I load the saved torchscript file,https://github.com/pytorch/pytorch/issues/73072,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-02-18T10:13:16Z
issue,72906,Replace deprecated distutils package in torch/utils/tensorboard/__init__.py,https://github.com/pytorch/pytorch/issues/72906,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: visualization,categorize and decide next maintainer owner/action,2022-02-16T23:39:18Z
issue,69347,Memory leak when moving tensors between Metal and CPU on iOS,https://github.com/pytorch/pytorch/issues/69347,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: mobile,categorize and decide next maintainer owner/action,2022-02-16T14:22:53Z
issue,70099,"Question: what is ""Parameter indices""?",https://github.com/pytorch/pytorch/issues/70099,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,categorize and decide next maintainer owner/action,2022-02-15T15:02:44Z
issue,70102,torch jit script segm fault,https://github.com/pytorch/pytorch/issues/70102,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-02-15T12:03:02Z
issue,70037,Question about collect tensor in distributed dataparallel?,https://github.com/pytorch/pytorch/issues/70037,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,categorize and decide next maintainer owner/action,2022-02-14T10:02:55Z
issue,71912,`torch._sources.normalize_source_lines()` and consequently `torch.jit.script` fail with lambda functions,https://github.com/pytorch/pytorch/issues/71912,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-02-14T08:04:31Z
issue,69972,torchscript does not work with `SyncBatchNorm` layers,https://github.com/pytorch/pytorch/issues/69972,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-02-13T22:02:38Z
issue,69830,Conversion error from pytorch model to libtorch model,https://github.com/pytorch/pytorch/issues/69830,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-02-13T06:03:01Z
issue,72746,CPU execution/dispatch time dominates and slows down small TorchScript GPU models,https://github.com/pytorch/pytorch/issues/72746,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-02-11T21:35:24Z
issue,69814,JIT / TorchScript docs missing any mention of typing.cast,https://github.com/pytorch/pytorch/issues/69814,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-02-11T09:02:58Z
issue,72295,[JIT][Autocasting] Add autocasting & constant propagation as part of freezing,https://github.com/pytorch/pytorch/issues/72295,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-02-10T21:36:02Z
issue,55211,[TorchScript] Can we support type refinement by Python type annotation?,https://github.com/pytorch/pytorch/issues/55211,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-02-10T19:16:12Z
issue,72617,TorchScript assertion failure for a `List[...]` inside a `NamedTuple`,https://github.com/pytorch/pytorch/issues/72617,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-02-09T21:06:41Z
issue,72252,Mobile: minimize inferencing latency by having better control over available CPU cores,https://github.com/pytorch/pytorch/issues/72252,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: mobile,categorize and decide next maintainer owner/action,2022-02-08T23:25:23Z
issue,69591,Make DDP + Zero have the same communication volume as regular DDP,https://github.com/pytorch/pytorch/issues/69591,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: ddp, oncall: distributed",categorize and decide next maintainer owner/action,2022-02-07T00:06:17Z
issue,69594,RuntimeError: Operation not supported,https://github.com/pytorch/pytorch/issues/69594,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: c10d, oncall: distributed",categorize and decide next maintainer owner/action,2022-02-07T00:06:16Z
issue,69563,[Distributed Tests] Print stacktraces of all processes when one fails,https://github.com/pytorch/pytorch/issues/69563,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,categorize and decide next maintainer owner/action,2022-02-05T22:03:29Z
issue,69506,[performance] a profiler util to show a rough break-down of the types of ops used by a model,https://github.com/pytorch/pytorch/issues/69506,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: profiler,categorize and decide next maintainer owner/action,2022-02-05T18:02:46Z
issue,71849,"addcmul, byte_channels_last fail test_nnc_correctness opinfo tests under UBSAN",https://github.com/pytorch/pytorch/issues/71849,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: sanitizers,categorize and decide next maintainer owner/action,2022-02-05T02:26:09Z
issue,72338,[JIT] Channels last optimization pass,https://github.com/pytorch/pytorch/issues/72338,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-02-04T19:29:43Z
issue,69435,"torch.is_tensor(obj) doesn't work with JIT, despite the fact that isinstance(obj, Tensor) already works with JIT",https://github.com/pytorch/pytorch/issues/69435,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-02-03T20:02:46Z
issue,69354,Missing instruction in recipe for Andoid/mobile interpreter,https://github.com/pytorch/pytorch/issues/69354,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: mobile,categorize and decide next maintainer owner/action,2022-02-01T16:03:08Z
issue,72029,JIT function crashing or failing (depending on profiling),https://github.com/pytorch/pytorch/issues/72029,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-02-01T00:39:37Z
issue,69084,Unicode Normalization with Torchscript,https://github.com/pytorch/pytorch/issues/69084,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-01-29T17:05:16Z
issue,68984,speed_benchmark_torch crashes when trying to run a model using Vulkan backend on Android,https://github.com/pytorch/pytorch/issues/68984,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: mobile,categorize and decide next maintainer owner/action,2022-01-29T02:10:23Z
issue,68940,Breakpoint training in a Tensorboard log,https://github.com/pytorch/pytorch/issues/68940,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: visualization,categorize and decide next maintainer owner/action,2022-01-28T23:02:44Z
issue,68968,torch.jit.trace with pack_padded_sequence cannot do dynamic batch,https://github.com/pytorch/pytorch/issues/68968,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-01-28T02:13:26Z
issue,71784,[JIT] addmm differs from eager mode,https://github.com/pytorch/pytorch/issues/71784,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-01-26T05:15:47Z
issue,71286,"[JIT][tensorexpr] cat, batch_norm opinfo tests failing",https://github.com/pytorch/pytorch/issues/71286,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-01-25T23:26:49Z
issue,65643,[JIT] Start Using and Optimizing on Symbolic Shape Inference in Freezing,https://github.com/pytorch/pytorch/issues/65643,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-01-24T21:23:24Z
issue,68800,[1.10 regression][jit][cuda fuse] cat codegen wrong var name,https://github.com/pytorch/pytorch/issues/68800,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-01-23T04:02:53Z
issue,68759,`test_variant_consistency_jit` errors for `torch.nn.functional.prelu`,https://github.com/pytorch/pytorch/issues/68759,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-01-21T22:03:01Z
issue,71548,Support freezing of models containing ModuleContainerIndex,https://github.com/pytorch/pytorch/issues/71548,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-01-20T17:16:40Z
issue,59905,[JIT] batch operators in training/inference,https://github.com/pytorch/pytorch/issues/59905,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-01-20T03:01:00Z
issue,54612,Exceptions thrown in JIT interpreter always are translated into RuntimeError,https://github.com/pytorch/pytorch/issues/54612,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-01-19T21:08:06Z
issue,71470,torch.jit.script failed to compile nn.MultiheadAttention when specifying the kdim and vdim parameters.,https://github.com/pytorch/pytorch/issues/71470,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-01-19T11:26:47Z
issue,65983,"traced EmbeddingBag cannot be used with different batch size, functional.embedding_bag is not de",https://github.com/pytorch/pytorch/issues/65983,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-01-18T16:25:59Z
issue,71151,The implement of `containsTensorType(const TypePtr& t)` in jit pass `PropagateInputShapes` ignore some situation?,https://github.com/pytorch/pytorch/issues/71151,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-01-17T02:56:02Z
issue,71165,traced module fails on second execution,https://github.com/pytorch/pytorch/issues/71165,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-01-15T04:11:07Z
issue,71288,`memory_format` argument isn't supported in torchscript for tensor.is_contiguous(),https://github.com/pytorch/pytorch/issues/71288,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-01-13T23:51:07Z
issue,70951,A diagnostics mode to report when constraints aren't being met for optimal performance,https://github.com/pytorch/pytorch/issues/70951,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: profiler,categorize and decide next maintainer owner/action,2022-01-07T04:42:29Z
issue,52271,torchscript should work with type: ignore comments from mypy even if they are not at the end of the line,https://github.com/pytorch/pytorch/issues/52271,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-01-06T13:09:46Z
issue,68785,[elastic/distributed] API to retrieve consolidated worker information,https://github.com/pytorch/pytorch/issues/68785,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: distributed, oncall: r2p",categorize and decide next maintainer owner/action,2022-01-06T09:01:19Z
issue,70267,Inconsistent multi-node latency with NCCL and OpenMPI,https://github.com/pytorch/pytorch/issues/70267,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: nccl, module: openmp, oncall: distributed",categorize and decide next maintainer owner/action,2022-01-05T23:50:06Z
issue,70155,RuntimeError: Tensor must be CUDA and dense when calling all_gather_object even though there is no tensor in the object.,https://github.com/pytorch/pytorch/issues/70155,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: c10d, oncall: distributed",categorize and decide next maintainer owner/action,2022-01-05T23:36:03Z
issue,70627,Torchscript Compiled functions don't support keyword-only arguments with defaults,https://github.com/pytorch/pytorch/issues/70627,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-01-04T19:45:53Z
issue,70573,JIT magic method which returns class instance fails.,https://github.com/pytorch/pytorch/issues/70573,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-01-03T09:52:35Z
issue,41492,Make torch.iinfo/torch.finfo torchscriptable,https://github.com/pytorch/pytorch/issues/41492,needs triage,47,triage signal present; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-12-31T13:58:43Z
issue,65490,optimize_for_mobile() assert fail: missing op for prepacked::conv2d_clamp_prepack,https://github.com/pytorch/pytorch/issues/65490,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: mobile,categorize and decide next maintainer owner/action,2021-12-29T16:58:16Z
issue,41353,[jit] support for torch.distributed primitives,https://github.com/pytorch/pytorch/issues/41353,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-12-24T06:43:40Z
issue,41416,[jit] support for generators and `yield`,https://github.com/pytorch/pytorch/issues/41416,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-12-23T15:31:54Z
issue,51673,[JIT/Futures] Future callbacks/then() APIs should throw the correct exception,https://github.com/pytorch/pytorch/issues/51673,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, oncall: jit",categorize and decide next maintainer owner/action,2021-12-23T01:19:29Z
issue,69608,Optimize For Inference - add warning/error upon serialization,https://github.com/pytorch/pytorch/issues/69608,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-12-22T20:42:45Z
issue,70054,[c10d] have a way to determine two ranks on the same hosts or not,https://github.com/pytorch/pytorch/issues/70054,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: c10d, oncall: distributed",categorize and decide next maintainer owner/action,2021-12-21T20:10:38Z
issue,69938,"JIT is overriding variables of type List[int] to type Tuple[int, int] for an unknown reason during function calls",https://github.com/pytorch/pytorch/issues/69938,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-12-20T21:30:09Z
issue,69320,[JIT] Break up JIT Opinfo Tests,https://github.com/pytorch/pytorch/issues/69320,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-12-16T17:18:27Z
issue,69522,PyTorch 1.9.1 incorrect result for all-reduce,https://github.com/pytorch/pytorch/issues/69522,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,categorize and decide next maintainer owner/action,2021-12-16T02:28:13Z
issue,52214,`hasattr` check not supported on objects other than `self`,https://github.com/pytorch/pytorch/issues/52214,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-12-14T16:08:19Z
issue,49782,TorchScript model's output different on Mobile and PC,https://github.com/pytorch/pytorch/issues/49782,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: mobile,categorize and decide next maintainer owner/action,2021-12-05T13:49:13Z
issue,68917,"Bugs related to NCCL on RTX 6000, code freezes with no output when using DistributedDataParallel",https://github.com/pytorch/pytorch/issues/68917,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: ddp, oncall: distributed",categorize and decide next maintainer owner/action,2021-11-30T18:20:08Z
issue,68862,find_unused_parameters of DDP,https://github.com/pytorch/pytorch/issues/68862,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: ddp, oncall: distributed",categorize and decide next maintainer owner/action,2021-11-30T18:06:38Z
issue,68927,Thank you for the diagnostic script.,https://github.com/pytorch/pytorch/issues/68927,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: visualization,categorize and decide next maintainer owner/action,2021-11-29T23:50:23Z
issue,63395,How to efficiently (without looping) get data from tensor predicted by a torchscript in C++?,https://github.com/pytorch/pytorch/issues/63395,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-11-28T03:54:22Z
issue,68536,Exporting a model with `_save_for_lite_interpreter()` fails with Segmentation fault,https://github.com/pytorch/pytorch/issues/68536,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: mobile,categorize and decide next maintainer owner/action,2021-11-24T12:53:27Z
issue,56595,"TorchScript breaks (""internal assert failed"") on `torch.set_grad_enabled`",https://github.com/pytorch/pytorch/issues/56595,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-11-20T20:20:56Z
issue,57457,Valgrind Shows Potential Memory Leaks,https://github.com/pytorch/pytorch/issues/57457,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: autograd, oncall: jit",categorize and decide next maintainer owner/action,2021-11-18T15:40:22Z
issue,43237,[jit] remove shape/device specialization from tracing after ONNX done with its own shape inference,https://github.com/pytorch/pytorch/issues/43237,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-11-16T23:27:16Z
issue,68049,Request to revise the Pytorch tutorial.,https://github.com/pytorch/pytorch/issues/68049,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: docs, oncall: jit",categorize and decide next maintainer owner/action,2021-11-16T19:06:05Z
issue,17932,JIT torch.ones_like with dtype starts failing on master,https://github.com/pytorch/pytorch/issues/17932,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-11-14T21:51:32Z
issue,14455,[JIT] Tracing a script function/module where not all args are Tensors,https://github.com/pytorch/pytorch/issues/14455,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-11-14T21:51:30Z
issue,68297,[DDP] Debug mode should ensure reduction is finished in backward pass,https://github.com/pytorch/pytorch/issues/68297,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: ddp, oncall: distributed",categorize and decide next maintainer owner/action,2021-11-12T23:18:06Z
issue,68104,DDP is not working with Pytorch Lightning,https://github.com/pytorch/pytorch/issues/68104,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,categorize and decide next maintainer owner/action,2021-11-11T18:56:46Z
issue,68088,[distributed] Restructure our distributed collective tests,https://github.com/pytorch/pytorch/issues/68088,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: c10d, oncall: distributed",categorize and decide next maintainer owner/action,2021-11-09T23:02:27Z
issue,68065,Incorrect return type annotation for _get_mobile_model_contained_types,https://github.com/pytorch/pytorch/issues/68065,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-11-09T16:43:56Z
issue,68064,In place `or` operator doesn't work in torchscript,https://github.com/pytorch/pytorch/issues/68064,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-11-09T16:25:43Z
issue,59551,"error when trying to call ""torch::jit::load""method, use Metal backend in PyTorch Mobile",https://github.com/pytorch/pytorch/issues/59551,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: ios, oncall: mobile",categorize and decide next maintainer owner/action,2021-11-08T23:59:23Z
issue,61268,Unknown type name '__torch__.torch.classes.metal.Conv2dOpContext',https://github.com/pytorch/pytorch/issues/61268,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: ios, module: macos, oncall: mobile",categorize and decide next maintainer owner/action,2021-11-08T23:36:20Z
issue,67603,The minimum supported version of pytorch when using toch-mobile？,https://github.com/pytorch/pytorch/issues/67603,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: mobile,categorize and decide next maintainer owner/action,2021-11-08T23:11:10Z
issue,60518,different gpus to train,https://github.com/pytorch/pytorch/issues/60518,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,categorize and decide next maintainer owner/action,2021-11-08T02:57:52Z
issue,53451,Model loaded from full model state checkpoint does not run on multiple GPUs (device mismatch),https://github.com/pytorch/pytorch/issues/53451,needs triage,47,"triage signal present; old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: nn, module: serialization, oncall: distributed",categorize and decide next maintainer owner/action,2021-11-05T14:23:32Z
issue,67889,[libtorch][torch.jit.script][input dynamic shape]Error in loading “. Pt” file on C + +: index out of range,https://github.com/pytorch/pytorch/issues/67889,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-11-05T02:31:42Z
issue,67838,`test_variant_consistency_jit_contiguous_cpu_float32` from `test_ops` fails,https://github.com/pytorch/pytorch/issues/67838,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-11-04T11:14:35Z
issue,54821,Decouple TorchScript compiler from operator library,https://github.com/pytorch/pytorch/issues/54821,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-11-03T18:07:00Z
issue,65814,INTERNAL ASSERT FAILED when reusing JIT compiled module,https://github.com/pytorch/pytorch/issues/65814,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-11-03T13:56:40Z
issue,67599,"Mobile builds need to make custom build of libtorch to run, let's just include it to make it easier",https://github.com/pytorch/pytorch/issues/67599,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: mobile,categorize and decide next maintainer owner/action,2021-11-02T11:40:24Z
issue,67659,`test_variant_consistency_jit` failing for `max_unpool2d`,https://github.com/pytorch/pytorch/issues/67659,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-11-02T03:31:51Z
issue,58456,TSAN issues in torch::jit::ProfilingRecord,https://github.com/pytorch/pytorch/issues/58456,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-11-02T02:28:00Z
issue,65058,Add nnapi serialization for module components in Wav2Vec2Model,https://github.com/pytorch/pytorch/issues/65058,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: android, oncall: mobile",categorize and decide next maintainer owner/action,2021-10-30T14:36:08Z
issue,65662,mobile deployment: optimized_for_mobile---wrap_cpp_module error,https://github.com/pytorch/pytorch/issues/65662,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: mobile,categorize and decide next maintainer owner/action,2021-10-29T15:02:54Z
issue,67542,Segfault from ONNX exporting code where torch.jit.script and torch.no_grad are used together,https://github.com/pytorch/pytorch/issues/67542,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-10-29T10:42:06Z
issue,67473,"[Feature] Separate, Queryable Alias Parser",https://github.com/pytorch/pytorch/issues/67473,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-10-29T03:42:24Z
issue,65734,pytorch mobile: some quant models on android cause memory leak,https://github.com/pytorch/pytorch/issues/65734,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: android, oncall: mobile",categorize and decide next maintainer owner/action,2021-10-28T22:25:44Z
issue,66545,Forward method slows down at some point if it is executed repeatedly on Android with NNAPI,https://github.com/pytorch/pytorch/issues/66545,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: android, oncall: mobile",categorize and decide next maintainer owner/action,2021-10-28T22:24:21Z
issue,67474,[JIT][Feature] Runtime Verifier of Alias Annotations,https://github.com/pytorch/pytorch/issues/67474,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-10-28T20:51:55Z
issue,67458,OpInfo JIT tests fail for torch.nonzero,https://github.com/pytorch/pytorch/issues/67458,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-10-28T18:48:38Z
issue,67420,Avoid warnings when jitting pytorch modules,https://github.com/pytorch/pytorch/issues/67420,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-10-28T05:21:36Z
issue,67392,Missing doc for torch.jit functions,https://github.com/pytorch/pytorch/issues/67392,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: docs, oncall: jit",categorize and decide next maintainer owner/action,2021-10-27T22:44:58Z
issue,66816,Jit fuser half precision support,https://github.com/pytorch/pytorch/issues/66816,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-10-27T19:42:09Z
issue,67160,"Include function name in ""This Python function is annotated to be ignored and cannot be run"" error message",https://github.com/pytorch/pytorch/issues/67160,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-10-26T15:34:22Z
issue,65824,Failing to compile /torch/csrc/jit/frontend/ir_emitter.cpp with pointer is null warning,https://github.com/pytorch/pytorch/issues/65824,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-10-25T23:14:39Z
issue,56703,[Pytorch Mobile] Error running build_pytorch_android.sh,https://github.com/pytorch/pytorch/issues/56703,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: android, oncall: mobile",categorize and decide next maintainer owner/action,2021-10-25T22:30:23Z
issue,57271,RuntimeError after optimize_for_mobile,https://github.com/pytorch/pytorch/issues/57271,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: android, oncall: mobile",categorize and decide next maintainer owner/action,2021-10-25T22:30:12Z
issue,60172,[Android] Upgrading to 1.9.0 causes NNAPI model loading to fail,https://github.com/pytorch/pytorch/issues/60172,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: android, oncall: mobile",categorize and decide next maintainer owner/action,2021-10-25T22:27:27Z
issue,61693,"Pytorch Mobile on Android, LAPACK library not found in compilation",https://github.com/pytorch/pytorch/issues/61693,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: android, oncall: mobile",categorize and decide next maintainer owner/action,2021-10-25T22:26:16Z
issue,61951,Vulkan: error -5 when trying to use it with Nvidia card,https://github.com/pytorch/pytorch/issues/61951,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: android, module: vulkan, oncall: mobile",categorize and decide next maintainer owner/action,2021-10-25T22:24:50Z
issue,62165,"RuntimeError: xnn_status_success == run_status INTERNAL ASSERT FAILED at ""/pytorch/aten/src/ATen/native/xnnpack/Linear.cpp"":158",https://github.com/pytorch/pytorch/issues/62165,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: android, module: ios, oncall: mobile",categorize and decide next maintainer owner/action,2021-10-25T22:24:25Z
issue,62676,pytorch_mobile custom build Module.forward null reference.,https://github.com/pytorch/pytorch/issues/62676,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: android, oncall: mobile",categorize and decide next maintainer owner/action,2021-10-25T22:23:49Z
issue,64333,When compiling TAG 1.9.0 or master for android，An error occurred and the compilation failed！ABIS_LIST=arm64-v8a,https://github.com/pytorch/pytorch/issues/64333,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: android, oncall: mobile",categorize and decide next maintainer owner/action,2021-10-25T22:21:35Z
issue,66768,Error: `copy_to_metal_ is implemented only for float dtype`,https://github.com/pytorch/pytorch/issues/66768,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: ios, oncall: mobile",categorize and decide next maintainer owner/action,2021-10-25T22:17:59Z
issue,66627,Overhaul error handling in `TCPStore`,https://github.com/pytorch/pytorch/issues/66627,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,categorize and decide next maintainer owner/action,2021-10-25T18:03:52Z
issue,67111,[doc] Can we update TorchScript union support?,https://github.com/pytorch/pytorch/issues/67111,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: docs, oncall: jit",categorize and decide next maintainer owner/action,2021-10-24T19:58:49Z
issue,67099,JIT: inconsistency in LSTM.forward,https://github.com/pytorch/pytorch/issues/67099,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-10-22T16:38:50Z
issue,51941,RuntimeError: torch.dtype passed as int from JIT to ignored function,https://github.com/pytorch/pytorch/issues/51941,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-10-19T19:38:37Z
issue,66841,"[Torchscript] Comparison between list and int returns true [1] == (1, ) == 1",https://github.com/pytorch/pytorch/issues/66841,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-10-19T02:17:21Z
issue,51154,[JIT] Rough edges found while exploring expressions supported by TorchScript,https://github.com/pytorch/pytorch/issues/51154,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-10-18T18:01:28Z
issue,66791,jit tracer：Lost imported pre-model parameters,https://github.com/pytorch/pytorch/issues/66791,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-10-18T14:42:34Z
issue,51892,[cuda jit extentions] global shared destination conflicts with virtual envs,https://github.com/pytorch/pytorch/issues/51892,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-10-14T20:48:07Z
issue,57929,[JIT] provide a pass for alias reduction,https://github.com/pytorch/pytorch/issues/57929,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-10-14T20:25:10Z
issue,66651,[JIT] Profile optional tensor,https://github.com/pytorch/pytorch/issues/66651,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-10-14T20:23:51Z
issue,66624,Separate libtorch and non-libtorch specific files under `torch/csrc/distributed`,https://github.com/pytorch/pytorch/issues/66624,needs triage,47,"triage signal present; old update age or low-urgency/stale label; component/oncall labels: module: rpc, oncall: distributed",categorize and decide next maintainer owner/action,2021-10-14T14:58:41Z
issue,65907,Inplace error of BatchNorm layer in DistributedDataParallel module,https://github.com/pytorch/pytorch/issues/65907,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,categorize and decide next maintainer owner/action,2021-10-12T06:09:27Z
issue,66150,JIT: empty_like doesn't support requires_grad,https://github.com/pytorch/pytorch/issues/66150,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-10-05T18:47:38Z
issue,66094,[JIT][Sym Shape Analysis] Extend analysis to extract output shape logic,https://github.com/pytorch/pytorch/issues/66094,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, oncall: jit",categorize and decide next maintainer owner/action,2021-10-05T01:03:21Z
issue,66065,torch.jit.script fails to cast explicit Optional parameter to bool,https://github.com/pytorch/pytorch/issues/66065,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-10-04T06:41:31Z
issue,65985,torch.cuda.power_usage (ESG 🌱),https://github.com/pytorch/pytorch/issues/65985,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: cuda, oncall: profiler",categorize and decide next maintainer owner/action,2021-10-01T17:48:03Z
issue,65918,C++ extensions don't correctly set USE_C10D_FOO preprocessor macros,https://github.com/pytorch/pytorch/issues/65918,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,categorize and decide next maintainer owner/action,2021-09-30T12:29:28Z
issue,65915,"I am getting undefined symbol: _ZN5torch3jit17parseSchemaOrNameERKNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEEE error. This I am getting when I am trying to ""import torch from nemo.collections import nlp"". I am trying to use pytorch ngc container 21.05. I tried to import torch before the nemo extension. Please suggest how I can resolve this.",https://github.com/pytorch/pytorch/issues/65915,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-09-30T12:17:54Z
issue,63610,Enable some amount of CI tests on lazy_tensor_staging branch,https://github.com/pytorch/pytorch/issues/63610,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: lazy, oncall: jit",categorize and decide next maintainer owner/action,2021-09-29T06:21:09Z
issue,65776,`torch.jit.save` fails on Windows,https://github.com/pytorch/pytorch/issues/65776,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-09-28T20:42:09Z
issue,65630,Constants in torch.Jit.script,https://github.com/pytorch/pytorch/issues/65630,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-09-24T20:50:23Z
issue,65588,Make torch.utils.tensorboard.add_scalar accepts float global_step,https://github.com/pytorch/pytorch/issues/65588,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: visualization,categorize and decide next maintainer owner/action,2021-09-24T15:39:49Z
issue,65611,"`torch.jit.trace` fails claiming ""forward method already defined""",https://github.com/pytorch/pytorch/issues/65611,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-09-24T14:33:06Z
issue,64462,torch.package: re-export docs suggestions don't work + other sharp edges,https://github.com/pytorch/pytorch/issues/64462,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: package/deploy,categorize and decide next maintainer owner/action,2021-09-23T23:45:09Z
issue,64986,RandomSampler / DistributedSampler does not seem really random,https://github.com/pytorch/pytorch/issues/64986,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: dataloader, oncall: distributed",categorize and decide next maintainer owner/action,2021-09-23T20:30:16Z
issue,65397,Vague linkage when using libcxxabi built torch in torch_glow,https://github.com/pytorch/pytorch/issues/65397,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-09-23T03:19:40Z
issue,65449,Make SummaryWriter.add_image supports PIL image directly,https://github.com/pytorch/pytorch/issues/65449,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: visualization,categorize and decide next maintainer owner/action,2021-09-22T14:55:28Z
issue,65398,[JIT] JIT does not support non-persistent buffers,https://github.com/pytorch/pytorch/issues/65398,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-09-21T11:56:03Z
issue,64795,profiler crashes the program when trying to stop on different threads without context manager,https://github.com/pytorch/pytorch/issues/64795,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: multithreading, oncall: distributed, oncall: profiler",categorize and decide next maintainer owner/action,2021-09-20T18:01:05Z
issue,47565,Keyword-only function not allowed with TorchScript,https://github.com/pytorch/pytorch/issues/47565,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, oncall: jit",categorize and decide next maintainer owner/action,2021-09-20T16:27:03Z
issue,65297,[JIT] TorchScript not works as expected when ModuleDict has modules with different signatures,https://github.com/pytorch/pytorch/issues/65297,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-09-18T10:48:30Z
issue,65098,Bug: torch.jit.load cannot load from IOBuffer,https://github.com/pytorch/pytorch/issues/65098,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-09-17T13:58:29Z
issue,65052,Use hyphens to separate long CLI parameters instead of underscores,https://github.com/pytorch/pytorch/issues/65052,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,categorize and decide next maintainer owner/action,2021-09-17T07:39:22Z
issue,65144,jit.trace's returned module returns a tuple instead of a named tuple,https://github.com/pytorch/pytorch/issues/65144,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-09-16T15:04:47Z
issue,63445,"getNcclVersion() in NCCLUtils.cpp handles ""2.10.3"" incorrectly",https://github.com/pytorch/pytorch/issues/63445,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: ddp, module: nccl, oncall: distributed",categorize and decide next maintainer owner/action,2021-09-14T23:11:36Z
issue,64799,Add support to profile all RPC worker threads that are run on a process,https://github.com/pytorch/pytorch/issues/64799,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,categorize and decide next maintainer owner/action,2021-09-14T21:51:07Z
issue,44077,[jit] Better type refinement for class attributes which are class types,https://github.com/pytorch/pytorch/issues/44077,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-09-14T17:31:56Z
issue,18520,Jit fail with TracingCheckError with tracing model with layers created after init.,https://github.com/pytorch/pytorch/issues/18520,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-09-14T17:17:53Z
issue,49643,Tracing doesn't work with spectral normalization,https://github.com/pytorch/pytorch/issues/49643,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-09-14T17:14:01Z
issue,64896,pytorch1.9.0 can not use torch.jit.load() to load model which save in pytorch1.3,https://github.com/pytorch/pytorch/issues/64896,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-09-14T16:05:57Z
issue,64905,[docs] [feature request] Tutorial on implementing correctly distributed dataset evaluators,https://github.com/pytorch/pytorch/issues/64905,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,categorize and decide next maintainer owner/action,2021-09-14T06:28:59Z
issue,64957,`INTERNAL ASSERT FAILED` after `torch.jit.freeze`,https://github.com/pytorch/pytorch/issues/64957,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-09-13T22:32:50Z
issue,64918,TorchScript error loading module with float literal default argument,https://github.com/pytorch/pytorch/issues/64918,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-09-13T10:36:55Z
issue,64488,It' so strange!!!,https://github.com/pytorch/pytorch/issues/64488,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: android, module: serialization, oncall: mobile",categorize and decide next maintainer owner/action,2021-09-10T20:31:32Z
issue,64775,[JIT] Canonicalize aten::rsub,https://github.com/pytorch/pytorch/issues/64775,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-09-09T20:03:58Z
issue,64490,Load torchscript model with custom operators (GNN) on Windows,https://github.com/pytorch/pytorch/issues/64490,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-09-09T09:41:08Z
issue,34713,torch.jit.script report error when using index to subscript nn.ModuleList,https://github.com/pytorch/pytorch/issues/34713,needs triage,47,triage signal present; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-09-09T07:22:39Z
issue,60143,"torch.jit.save gives error - RuntimeError: Could not export Python function call 'NMSop'. Remove calls to Python functions before export. Did you forget to add @script or @script_method annotation? If this is a nn.ModuleList, add it to __constants__",https://github.com/pytorch/pytorch/issues/60143,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-09-09T03:06:59Z
issue,64150,Enhance DDP doc to have a complete example including data loading,https://github.com/pytorch/pytorch/issues/64150,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,categorize and decide next maintainer owner/action,2021-09-08T15:26:06Z
issue,64628,[jit] cannot compile staticmethod of classes when they have a unscriptable __init__,https://github.com/pytorch/pytorch/issues/64628,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-09-08T06:46:32Z
issue,64487,Error in torchScript based on Pytorch tutorial,https://github.com/pytorch/pytorch/issues/64487,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-09-03T19:27:03Z
issue,63740,DDP Gradient reduction is not triggered after calling autograd.backward(),https://github.com/pytorch/pytorch/issues/63740,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,categorize and decide next maintainer owner/action,2021-09-03T18:04:14Z
issue,57893,Android API to run model on Vulkan backend,https://github.com/pytorch/pytorch/issues/57893,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: mobile,categorize and decide next maintainer owner/action,2021-09-02T07:33:22Z
issue,64130,Every process takes a tiny memory on each nvidia GPU card when using DDP.,https://github.com/pytorch/pytorch/issues/64130,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,categorize and decide next maintainer owner/action,2021-09-01T18:57:44Z
issue,64258,Incubation of Graphical Representation in TF2 of HParams,https://github.com/pytorch/pytorch/issues/64258,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: tensorboard, oncall: visualization",categorize and decide next maintainer owner/action,2021-09-01T18:54:24Z
issue,64334,How to add nan value judgment for variable t0_1 in fused_clamp kernel generated by torch/csrc/jit/tensorexpr/cuda_codegen.cpp.,https://github.com/pytorch/pytorch/issues/64334,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-09-01T02:53:24Z
issue,64260,Using Cloudpickle for pickling jitted functions,https://github.com/pytorch/pytorch/issues/64260,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-08-31T11:33:46Z
issue,63648,[JIT] Typing on math.ceil is inaccurate for Scalar input,https://github.com/pytorch/pytorch/issues/63648,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-08-31T11:13:09Z
issue,64247,How to optimize jit-script model performance (backend device is gpu),https://github.com/pytorch/pytorch/issues/64247,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-08-31T05:31:46Z
issue,64005,[LTC] Find a way to convert at::Generator into torch::jit::NamedValue,https://github.com/pytorch/pytorch/issues/64005,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: lazy, oncall: jit",categorize and decide next maintainer owner/action,2021-08-30T17:00:16Z
issue,58507,OpInfo JIT tests do not work with Tensor kwarg arguments,https://github.com/pytorch/pytorch/issues/58507,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: tests, oncall: jit",categorize and decide next maintainer owner/action,2021-08-30T02:55:37Z
issue,42847,Get test_jit.py below 10k lines,https://github.com/pytorch/pytorch/issues/42847,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: tests, oncall: jit",categorize and decide next maintainer owner/action,2021-08-30T02:44:31Z
issue,64119,Cleanup tests in rpc_test.py,https://github.com/pytorch/pytorch/issues/64119,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: tests, oncall: distributed",categorize and decide next maintainer owner/action,2021-08-30T02:39:22Z
issue,64096,String frontend doesn't support tuple unpacking,https://github.com/pytorch/pytorch/issues/64096,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-08-27T16:05:19Z
issue,64060,Consolidating NCCL version parsing functions,https://github.com/pytorch/pytorch/issues/64060,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, module: c10d, oncall: distributed",categorize and decide next maintainer owner/action,2021-08-27T14:37:39Z
issue,62885,Type of first constructor parameter of ElasticDistributedSampler and DistributedSampler,https://github.com/pytorch/pytorch/issues/62885,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: dataloader, oncall: distributed",categorize and decide next maintainer owner/action,2021-08-27T09:37:35Z
issue,53421,[bug][JIT] Error when tracing model with partial hooks in PyTorch v1.8.0,https://github.com/pytorch/pytorch/issues/53421,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-08-26T16:09:53Z
issue,55749,Fail to introduce torch::jit::Module as a parameter of a cusomized operator.,https://github.com/pytorch/pytorch/issues/55749,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-08-26T16:09:48Z
issue,60089,torch.package: TypeError: 'NoneType' object is not iterable,https://github.com/pytorch/pytorch/issues/60089,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: package/deploy,categorize and decide next maintainer owner/action,2021-08-26T16:08:29Z
issue,62542,3 Times memory cost when loading the model to torch.nn.parallel.DistributedDataParallel,https://github.com/pytorch/pytorch/issues/62542,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: ddp, oncall: distributed",categorize and decide next maintainer owner/action,2021-08-25T20:35:09Z
issue,63832,[RFC] Add `torch.distributed.run` as a console script in pytorch's setup.py,https://github.com/pytorch/pytorch/issues/63832,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: elastic, oncall: distributed",categorize and decide next maintainer owner/action,2021-08-25T19:44:57Z
issue,63455,distributed_error,https://github.com/pytorch/pytorch/issues/63455,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,categorize and decide next maintainer owner/action,2021-08-25T17:09:30Z
issue,63405,torch.arange has issue tracing with bool input,https://github.com/pytorch/pytorch/issues/63405,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-08-20T17:05:37Z
issue,63403,Consolidate distributed benchmark folders,https://github.com/pytorch/pytorch/issues/63403,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, module: ddp, oncall: distributed",categorize and decide next maintainer owner/action,2021-08-17T21:13:59Z
issue,62325,[jit]][script] handling of 'void' returns,https://github.com/pytorch/pytorch/issues/62325,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-08-17T14:37:05Z
issue,62775,[TorchScript] Support for registering builtin operators with control flow,https://github.com/pytorch/pytorch/issues/62775,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-08-16T19:10:12Z
issue,63247,Add a timeout argument to RPC shutdown(),https://github.com/pytorch/pytorch/issues/63247,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, oncall: distributed",categorize and decide next maintainer owner/action,2021-08-13T18:50:29Z
issue,63124,torch.jit.trace quantized bigbird leads to 0INTERNAL ASSERT FAILED runtime error,https://github.com/pytorch/pytorch/issues/63124,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-08-13T17:17:06Z
issue,57559,[JIT] Investigate cloning/copying object before tracing it,https://github.com/pytorch/pytorch/issues/57559,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-08-12T21:08:39Z
issue,63171,[BE] Clean up DDP Single-Program Multi-Device Vestige in reducer.cpp,https://github.com/pytorch/pytorch/issues/63171,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, oncall: distributed",categorize and decide next maintainer owner/action,2021-08-12T19:15:01Z
issue,63145,torch.jit.trace does not work if there is an autograd in the function,https://github.com/pytorch/pytorch/issues/63145,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-08-12T11:50:15Z
issue,54937,[JIT] export fast-transformer to TorchScript,https://github.com/pytorch/pytorch/issues/54937,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-08-12T06:22:51Z
issue,55742,[JIT] TorchScript to represent typed tensor annotations in IR graph,https://github.com/pytorch/pytorch/issues/55742,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-08-12T06:10:33Z
issue,62154,AttributeError: module 'torch.jit' has no attribute '_script_if_tracing',https://github.com/pytorch/pytorch/issues/62154,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-08-12T05:34:30Z
issue,62394,crash in libtorch jit,https://github.com/pytorch/pytorch/issues/62394,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-08-12T05:30:57Z
issue,54593,Multiple refinement with `isinstance` doesn't work,https://github.com/pytorch/pytorch/issues/54593,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-08-11T14:38:11Z
issue,42592,Lower performance when forwarding JIT model with libTorch 1.6 compared with 1.5,https://github.com/pytorch/pytorch/issues/42592,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-08-11T06:22:53Z
issue,53744,TorchScript divison by zero fails to error out,https://github.com/pytorch/pytorch/issues/53744,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-08-11T05:55:58Z
issue,56660,consistency jit tests gives error for gradient operator,https://github.com/pytorch/pytorch/issues/56660,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-08-11T05:14:48Z
issue,60953,Add support for Python's optimized mode to torchscript,https://github.com/pytorch/pytorch/issues/60953,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-08-11T04:55:34Z
issue,62854,"aten::pixel_shuffle(Tensor self, int upscale_factor) -> (Tensor): Argument self not provided.",https://github.com/pytorch/pytorch/issues/62854,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-08-05T23:28:27Z
issue,62283,c10d and discontiguous tensors with mismatch strides,https://github.com/pytorch/pytorch/issues/62283,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: c10d, oncall: distributed",categorize and decide next maintainer owner/action,2021-08-04T13:51:08Z
issue,62314,dist.destroy_process_group() don't release master_port when there're at least 2 process groups,https://github.com/pytorch/pytorch/issues/62314,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: c10d, oncall: distributed",categorize and decide next maintainer owner/action,2021-08-04T13:49:12Z
issue,62403,DDP,https://github.com/pytorch/pytorch/issues/62403,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: ddp, oncall: distributed",categorize and decide next maintainer owner/action,2021-08-04T13:32:43Z
issue,8473,RuntimeError: /pytorch/torch/csrc/jit/tracer.h:117: getTracingState: Assertion `var_state == state` failed.,https://github.com/pytorch/pytorch/issues/8473,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-08-04T06:27:58Z
issue,62543,FAIL: test_threshold (test_jit_fuser.TestFuser),https://github.com/pytorch/pytorch/issues/62543,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-08-03T06:18:48Z
issue,62603,Unexpected `is_training` after using `train` in a JIT Module,https://github.com/pytorch/pytorch/issues/62603,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-08-02T20:29:08Z
issue,62556,[TorchScript] Inconsistency between Python and PyTorch when raising 0 to a negative power,https://github.com/pytorch/pytorch/issues/62556,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-08-02T05:40:55Z
issue,54728,JIT prim ops v.s. dispatcher,https://github.com/pytorch/pytorch/issues/54728,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: dispatch, oncall: jit, oncall: mobile",categorize and decide next maintainer owner/action,2021-08-02T05:26:32Z
issue,54982,[RFC] A PyTorch Tensor Shape DSL For Symbolic Shape Inference,https://github.com/pytorch/pytorch/issues/54982,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-07-27T20:55:34Z
issue,62007,Add Future return from default process group collective communications,https://github.com/pytorch/pytorch/issues/62007,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: c10d, oncall: distributed",categorize and decide next maintainer owner/action,2021-07-23T02:28:50Z
issue,61765,How to save tensors on mobile (lite interpreter)?,https://github.com/pytorch/pytorch/issues/61765,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: mobile,categorize and decide next maintainer owner/action,2021-07-21T10:06:17Z
issue,51574,[Pipe] conversion to Sequential: various complex scenarios,https://github.com/pytorch/pytorch/issues/51574,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,categorize and decide next maintainer owner/action,2021-07-13T19:16:46Z
issue,50734,[TorchScript] Document TorchScript IR,https://github.com/pytorch/pytorch/issues/50734,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-07-13T15:24:02Z
issue,61267,Incorrect trace with MKLDNN (adding scalar),https://github.com/pytorch/pytorch/issues/61267,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: mkldnn, oncall: jit",categorize and decide next maintainer owner/action,2021-07-12T20:07:38Z
issue,60847,How to release CPU memory cache in Libtorch JIT ?,https://github.com/pytorch/pytorch/issues/60847,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-07-07T03:01:06Z
issue,57894,"JIT ""optimizations"" take way too long",https://github.com/pytorch/pytorch/issues/57894,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-07-07T00:50:10Z
issue,60640,Building fails on a node with libibverbs installed with no Infiniband hardware present on RHEL8.,https://github.com/pytorch/pytorch/issues/60640,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: tensorpipe, oncall: distributed",categorize and decide next maintainer owner/action,2021-07-02T07:51:14Z
issue,60440,DDP with cuda rpc failed with DistributedOptimizer Adagrad,https://github.com/pytorch/pytorch/issues/60440,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: optimizer, oncall: distributed",categorize and decide next maintainer owner/action,2021-07-01T22:50:11Z
issue,56969,[JIT / NNC] Incomplete Ops Coverage,https://github.com/pytorch/pytorch/issues/56969,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-06-29T21:41:46Z
issue,60732,Potential memory leaks reported by static analysis tool,https://github.com/pytorch/pytorch/issues/60732,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-06-28T08:12:03Z
issue,59865,Custom build of c++ libtorch for Android program or dynamic so lib does not reduce program or lib size,https://github.com/pytorch/pytorch/issues/59865,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: android, oncall: mobile",categorize and decide next maintainer owner/action,2021-06-25T11:47:24Z
issue,60359,Expose the .pt model file path during TorchScript custom class Serialization/Deserialization,https://github.com/pytorch/pytorch/issues/60359,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-06-22T01:40:34Z
issue,60189,[package] Tutorial not accessible,https://github.com/pytorch/pytorch/issues/60189,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: package/deploy,categorize and decide next maintainer owner/action,2021-06-19T10:02:03Z
issue,60062,[c10d] Make `broadcast_object_list` accept a device parameter,https://github.com/pytorch/pytorch/issues/60062,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, oncall: distributed",categorize and decide next maintainer owner/action,2021-06-16T01:42:44Z
issue,59441,distributed/test_jit_c10d.py fails with RuntimeError: Address already in use,https://github.com/pytorch/pytorch/issues/59441,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,categorize and decide next maintainer owner/action,2021-06-14T23:54:07Z
issue,58907,"JIT fails when calling an exported method inside another one, if kwargs with defaults are not passed",https://github.com/pytorch/pytorch/issues/58907,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-06-14T10:05:06Z
issue,59823,print_regression should handle sharded test properly,https://github.com/pytorch/pytorch/issues/59823,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,categorize and decide next maintainer owner/action,2021-06-11T21:58:47Z
issue,59806,Context manager for torch.jit.ignore is not covered by test coverage,https://github.com/pytorch/pytorch/issues/59806,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-06-10T18:28:57Z
issue,59652,pin_memory mutates its input but this is not reflected in JIT schema,https://github.com/pytorch/pytorch/issues/59652,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-06-08T19:57:51Z
issue,59552,test_batch_isend_irecv_nccl fails with NCCL 2.8.3,https://github.com/pytorch/pytorch/issues/59552,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,categorize and decide next maintainer owner/action,2021-06-07T20:20:10Z
issue,59153,JIT compiling a ParameterList raises an unnecessary warning,https://github.com/pytorch/pytorch/issues/59153,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-06-07T19:04:15Z
issue,59528,AttributeError: module 'torch.jit' has no attribute '_script_if_tracing',https://github.com/pytorch/pytorch/issues/59528,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-06-07T16:56:06Z
issue,59114,[JIT] OpInfo / Autodiff tests dont seem to actually invoke symbolic backwards,https://github.com/pytorch/pytorch/issues/59114,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-06-06T17:27:38Z
issue,59257,Optimizer step stuck during distributed training,https://github.com/pytorch/pytorch/issues/59257,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision), module: data parallel, oncall: distributed",categorize and decide next maintainer owner/action,2021-06-04T21:18:02Z
issue,35372,How to support single-process-multiple-devices in DistributedDataParallel other than CUDA device,https://github.com/pytorch/pytorch/issues/35372,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,categorize and decide next maintainer owner/action,2021-06-04T13:57:34Z
issue,37442,Tensor.is_distributed not documented,https://github.com/pytorch/pytorch/issues/37442,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: docs, oncall: distributed",categorize and decide next maintainer owner/action,2021-06-03T02:07:06Z
issue,57716,Unknown builtin op: torchvision::nms in LibTorch,https://github.com/pytorch/pytorch/issues/57716,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: serialization, module: vision, oncall: jit",categorize and decide next maintainer owner/action,2021-06-02T00:14:23Z
issue,59174,[jit] Error: Results of original model and exported/imported version of model differed,https://github.com/pytorch/pytorch/issues/59174,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-06-02T00:12:28Z
issue,42267,assertRaisesRegxWithHighlight isn't triggered correctly,https://github.com/pytorch/pytorch/issues/42267,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-06-01T21:50:51Z
issue,41242,"test_backward_deadlock fails with ""Directory not empty""",https://github.com/pytorch/pytorch/issues/41242,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: power, oncall: jit",categorize and decide next maintainer owner/action,2021-06-01T16:46:34Z
issue,58982,Can't re-init rpc with a different rank after rpc shutdown,https://github.com/pytorch/pytorch/issues/58982,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,categorize and decide next maintainer owner/action,2021-06-01T06:24:09Z
issue,37387,The pytorch's graph is lack of common names for nodes,https://github.com/pytorch/pytorch/issues/37387,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, module: tensorboard, oncall: visualization",categorize and decide next maintainer owner/action,2021-05-25T21:52:01Z
issue,58836,Latest nightly produces CUDA torchscript that cannot be run,https://github.com/pytorch/pytorch/issues/58836,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-05-24T21:04:03Z
issue,58857,Call non class methods from `torch::jit::Module` in C++,https://github.com/pytorch/pytorch/issues/58857,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-05-24T20:40:39Z
issue,57887,[Mobile] Valid scale ratio with invalid error,https://github.com/pytorch/pytorch/issues/57887,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: mobile,categorize and decide next maintainer owner/action,2021-05-24T17:58:00Z
issue,58804,A Null pointer dereference bug,https://github.com/pytorch/pytorch/issues/58804,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-05-24T16:09:50Z
issue,58516,Unhelpful error message when type annotation is violated.,https://github.com/pytorch/pytorch/issues/58516,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-05-18T22:58:32Z
issue,57639,"After tracing, the device of the named buffer is still `cuda` although it was moved to `cpu `",https://github.com/pytorch/pytorch/issues/57639,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-05-18T16:56:06Z
issue,53472,After upgrade to 1.8.0 memory leak in libTorch after using torch::jit::load(),https://github.com/pytorch/pytorch/issues/53472,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit, oncall: mobile",categorize and decide next maintainer owner/action,2021-05-18T14:50:15Z
issue,57739,object with __torch_functon__ does not work with jit function,https://github.com/pytorch/pytorch/issues/57739,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-05-11T18:03:57Z
issue,57793,"LibTorch inference script model is slower than PyTorch, and the speed is not stable. Why?",https://github.com/pytorch/pytorch/issues/57793,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-05-07T23:38:04Z
issue,57717,TorchScript JIT Compiler Can't Recognize Repeated Branches (Scripting),https://github.com/pytorch/pytorch/issues/57717,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-05-07T12:50:30Z
issue,57592,`jit.trace` fails to capture single-element tensor passed into op requiring scalar-type argument,https://github.com/pytorch/pytorch/issues/57592,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-05-05T05:59:28Z
issue,51931,[Pipe] Tied layers,https://github.com/pytorch/pytorch/issues/51931,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,categorize and decide next maintainer owner/action,2021-05-05T00:30:15Z
issue,51932,[Pipe] Memory-Efficient Model Construction,https://github.com/pytorch/pytorch/issues/51932,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,categorize and decide next maintainer owner/action,2021-05-05T00:30:09Z
issue,43765,JIT fails sanity checks during tracing torch.rand_like,https://github.com/pytorch/pytorch/issues/43765,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-05-04T21:27:59Z
issue,55999,[JIT][ProfilingExecutor] A reliable and consistent API/protocol to query & propagate profiling data,https://github.com/pytorch/pytorch/issues/55999,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-05-04T19:45:54Z
issue,57313,[jit] pickling custom classes may lead to invalid pickle files,https://github.com/pytorch/pytorch/issues/57313,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-05-04T17:12:33Z
issue,48066,[JIT] Saving and Loading JIT Model Does Not Preserve __dir__,https://github.com/pytorch/pytorch/issues/48066,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-05-04T17:02:42Z
issue,53904,[JIT] Support `torch.ones([])`,https://github.com/pytorch/pytorch/issues/53904,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-05-01T17:23:03Z
issue,57163,Segmentation fault when using add_graph from tensorboard in combination with try except in forward pass,https://github.com/pytorch/pytorch/issues/57163,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: visualization,categorize and decide next maintainer owner/action,2021-04-29T16:36:39Z
issue,57035,`typing.List` need to be on the same line as the variable declaration?,https://github.com/pytorch/pytorch/issues/57035,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-04-27T16:51:17Z
issue,56006,Unable to get a Vulkan tensor using `to('vulkan')`,https://github.com/pytorch/pytorch/issues/56006,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: vulkan, oncall: mobile",categorize and decide next maintainer owner/action,2021-04-26T17:04:38Z
issue,56244,Cuda RPC error when using then(),https://github.com/pytorch/pytorch/issues/56244,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: tensorpipe, oncall: distributed",categorize and decide next maintainer owner/action,2021-04-26T08:03:37Z
issue,56571,Python static variable access for TorchScript's custom C++ classes,https://github.com/pytorch/pytorch/issues/56571,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-04-22T16:30:29Z
issue,52655,Optimize `einsum` in TorchScript profile guided optimization,https://github.com/pytorch/pytorch/issues/52655,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: linear algebra, module: performance, oncall: jit",categorize and decide next maintainer owner/action,2021-04-20T21:14:25Z
issue,56096,[JIT] Merge/reconcile `__jit_unused_properties` and `__jit_ignored_attributes__`,https://github.com/pytorch/pytorch/issues/56096,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-04-20T17:35:25Z
issue,56159,Addition/subtraction/multiplication/division of bool variables are not supported in JIT,https://github.com/pytorch/pytorch/issues/56159,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-04-20T17:19:35Z
issue,43249,[JIT] Cannot iterate over nn.ModuleList in JITted code when accessed through one level of indirection,https://github.com/pytorch/pytorch/issues/43249,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-04-20T16:34:41Z
issue,49614,Improve error on missing/problematic Module member type annotation,https://github.com/pytorch/pytorch/issues/49614,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-04-20T16:32:35Z
issue,50231,"Tracing with autocast failed with error: ""Cannot insert a Tensor that requires grad as a constant""",https://github.com/pytorch/pytorch/issues/50231,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision), oncall: jit",categorize and decide next maintainer owner/action,2021-04-20T16:32:24Z
issue,51018,len of ModuleList is incorrect for jitted models,https://github.com/pytorch/pytorch/issues/51018,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-04-20T16:32:13Z
issue,49261,[TorchScipt] Custom Objects C++ API,https://github.com/pytorch/pytorch/issues/49261,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-04-20T16:31:56Z
issue,49820,[JIT] Don't support math.exp and math.log in the Script,https://github.com/pytorch/pytorch/issues/49820,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-04-20T16:31:34Z
issue,51465,torch::jit::Module cannot be forward declared and used into a `unique_ptr`,https://github.com/pytorch/pytorch/issues/51465,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-04-20T16:30:49Z
issue,51137,[TorchScript Usability] Earlier type checking needed for interface type,https://github.com/pytorch/pytorch/issues/51137,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-04-20T16:29:28Z
issue,51942,Overwritten method has wrong self when JIT'ed,https://github.com/pytorch/pytorch/issues/51942,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-04-20T16:28:45Z
issue,51994,Allow Custom Classes to register a handler for .to operations,https://github.com/pytorch/pytorch/issues/51994,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: custom-operators, oncall: jit",categorize and decide next maintainer owner/action,2021-04-20T16:28:22Z
issue,52309,[JIT] __getitem__ on scripted ModuleDict returns unscripted module,https://github.com/pytorch/pytorch/issues/52309,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-04-20T16:28:02Z
issue,52920,[TorchScript Usability] More intuitive error message when passing wrong types to `torch.jit.script()`,https://github.com/pytorch/pytorch/issues/52920,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-04-20T16:27:11Z
issue,53344,[JIT] raise unittest.SkipTest and self.skipTest cannot be used to skip tests in JitTestCase,https://github.com/pytorch/pytorch/issues/53344,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-04-20T16:26:22Z
issue,56314,test_variant_consistency_jit fails for torch.tensordot with dtype float32 with error INTERNAL ASSERT FAILED,https://github.com/pytorch/pytorch/issues/56314,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: tests, oncall: jit",categorize and decide next maintainer owner/action,2021-04-20T03:54:54Z
issue,55779,Request setNumThreads API for Java,https://github.com/pytorch/pytorch/issues/55779,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: mobile,categorize and decide next maintainer owner/action,2021-04-19T17:38:19Z
issue,56007,NNAPI support for torch.cat([...]) when dim=-1,https://github.com/pytorch/pytorch/issues/56007,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: mobile,categorize and decide next maintainer owner/action,2021-04-19T17:37:52Z
issue,56111,ProcessGroupGlooTest and other distributed tests don't capture stdout/stderr in test UI view,https://github.com/pytorch/pytorch/issues/56111,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,categorize and decide next maintainer owner/action,2021-04-16T15:34:22Z
issue,51055,Native code assert or hang on process exit with a torch.jit.script model on multi-GPU,https://github.com/pytorch/pytorch/issues/51055,needs triage,47,triage signal present; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-04-16T15:05:53Z
issue,34682,Pytorch not compatible with react native android,https://github.com/pytorch/pytorch/issues/34682,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: mobile,categorize and decide next maintainer owner/action,2021-04-15T23:32:03Z
issue,48108,JIT should not force users to write ugly code,https://github.com/pytorch/pytorch/issues/48108,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-04-15T18:14:14Z
issue,56010,[tensorboard] Expected data-type of images in tensorboard SummaryWriter,https://github.com/pytorch/pytorch/issues/56010,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: visualization,categorize and decide next maintainer owner/action,2021-04-15T10:36:09Z
issue,56024,Support for mkldnn + ddp,https://github.com/pytorch/pytorch/issues/56024,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: mkldnn, oncall: distributed",categorize and decide next maintainer owner/action,2021-04-14T15:46:05Z
issue,55541,`torch.jit.interface` does not understand dunder methods (e.g. `__call__`),https://github.com/pytorch/pytorch/issues/55541,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-04-13T17:17:39Z
issue,35215,Get hold of Backward graph from a C++ module.,https://github.com/pytorch/pytorch/issues/35215,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-04-13T08:12:03Z
issue,55782,YoloV4 Pytorch model inference fails on NNAPI Android,https://github.com/pytorch/pytorch/issues/55782,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: mobile,categorize and decide next maintainer owner/action,2021-04-13T00:38:19Z
issue,44469,Error in the converting to pytorch mobile..,https://github.com/pytorch/pytorch/issues/44469,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: mobile,categorize and decide next maintainer owner/action,2021-04-07T17:42:20Z
issue,55194,Tracing `len(tensor)` gets static constant value in result graph,https://github.com/pytorch/pytorch/issues/55194,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-04-02T17:36:10Z
issue,32901,QNNPACK linear doesn't preserve dimensions,https://github.com/pytorch/pytorch/issues/32901,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: mobile,categorize and decide next maintainer owner/action,2021-04-01T00:13:19Z
issue,55072,[JIT][Testing] Replace use of `assertRaisesRegex` with `assertRaisesRegexWithHilight`,https://github.com/pytorch/pytorch/issues/55072,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-03-31T18:30:08Z
issue,54912,[autocast] DataParallel in a single process - possibly outdated docs,https://github.com/pytorch/pytorch/issues/54912,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,categorize and decide next maintainer owner/action,2021-03-31T18:29:22Z
issue,54598,Type refinement with `isinstance` only works with generics,https://github.com/pytorch/pytorch/issues/54598,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-03-30T17:14:19Z
issue,54599,Better documentation for `torch.jit.isinstance`,https://github.com/pytorch/pytorch/issues/54599,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: docs, oncall: jit",categorize and decide next maintainer owner/action,2021-03-30T17:13:27Z
issue,50933,[JIT] Type refinement failure in if-elif-else structure,https://github.com/pytorch/pytorch/issues/50933,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-03-30T15:55:58Z
issue,54625,"[torchscript] nn.Embedding(_weight), tensor.masked_fill_, and torch.autograd.grad causes wrong gradient",https://github.com/pytorch/pytorch/issues/54625,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-03-30T05:24:19Z
issue,45524,[JIT] JIT Autograd is saving more results than are necessary for bprop like the result of RELU when it is an intermediate,https://github.com/pytorch/pytorch/issues/45524,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-03-26T17:55:21Z
issue,13164,Tracing custom ops,https://github.com/pytorch/pytorch/issues/13164,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-03-26T10:15:40Z
issue,54348,[JIT] TorchScript does not correctly handle `torch.` namespace functions called with `kwargs`,https://github.com/pytorch/pytorch/issues/54348,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-03-23T21:06:14Z
issue,54435,[JIT] prim::ModuleContainerIndex doesnt work with freezing,https://github.com/pytorch/pytorch/issues/54435,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-03-23T17:38:25Z
issue,54365,Support `Generic` Type in TorchScript,https://github.com/pytorch/pytorch/issues/54365,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-03-23T17:16:40Z
issue,54146,Conversion of Pytorch model to Torchscript for model LaneATT(Lane Detection),https://github.com/pytorch/pytorch/issues/54146,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-03-23T04:08:24Z
issue,46944,Export TorchScript Classes as TorchScript code,https://github.com/pytorch/pytorch/issues/46944,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-03-18T01:30:33Z
issue,53921,[JIT] TorchScript infers incorrect type for `or` in non-boolean context,https://github.com/pytorch/pytorch/issues/53921,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-03-16T17:41:00Z
issue,53862,"After the source code is compiled, an error will be reported fvcore",https://github.com/pytorch/pytorch/issues/53862,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-03-16T17:27:20Z
issue,53878,[JIT] TorchScript errors out when assigning a value to an existing variable that's defined in outer block,https://github.com/pytorch/pytorch/issues/53878,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-03-16T17:26:22Z
issue,53877,[JIT] Bad interaction between if-else-style and assert-style type refinement,https://github.com/pytorch/pytorch/issues/53877,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-03-16T17:22:40Z
issue,53753,TorchScript cannot handle declared module attributes of Python types,https://github.com/pytorch/pytorch/issues/53753,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-03-16T17:20:14Z
issue,50971,Failing to trace Huggingface models using Trace API from Torchscript,https://github.com/pytorch/pytorch/issues/50971,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-03-15T12:13:17Z
issue,41512,Tensor.new_tensor is not supported,https://github.com/pytorch/pytorch/issues/41512,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-03-14T23:44:03Z
issue,47692,pyre not able to infer the type of torch.jit.script,https://github.com/pytorch/pytorch/issues/47692,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-03-12T21:27:40Z
issue,53824,[JIT] torch.jit.optimized_execution(True) greatly slows down some operations in PyTorch 1.8.0,https://github.com/pytorch/pytorch/issues/53824,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-03-12T05:50:45Z
issue,53264,How to load trained . torch from conversion to .mlmodel,https://github.com/pytorch/pytorch/issues/53264,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: mobile,categorize and decide next maintainer owner/action,2021-03-12T05:28:21Z
issue,53867,[JIT][Static Runtime] Memory optimization for output tensors,https://github.com/pytorch/pytorch/issues/53867,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-03-12T02:17:33Z
issue,53414,[JIT] Metaprogram non-type based expression trees,https://github.com/pytorch/pytorch/issues/53414,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-03-09T18:26:34Z
issue,52589,[JIT-autodiff] batch_norm isn't differentiable,https://github.com/pytorch/pytorch/issues/52589,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-03-09T07:25:09Z
issue,53140,[JIT] Document __jit_ignored_attributes__,https://github.com/pytorch/pytorch/issues/53140,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-03-09T02:24:27Z
issue,52768,[nnc] Optimize tiny convolutions,https://github.com/pytorch/pytorch/issues/52768,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-03-07T03:25:44Z
issue,52772,[nnc] Optimize 1x1 convolutions,https://github.com/pytorch/pytorch/issues/52772,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-03-07T03:25:37Z
issue,52777,[nnc] Placeholder::load braced init list fails to deduce,https://github.com/pytorch/pytorch/issues/52777,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-03-07T03:25:25Z
issue,52821,[nnc] Compute expressions with > 4 axes is pretty tedious,https://github.com/pytorch/pytorch/issues/52821,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-03-07T03:24:20Z
issue,46995,max_pool2d always fails on native Android app,https://github.com/pytorch/pytorch/issues/46995,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: mobile,categorize and decide next maintainer owner/action,2021-03-05T16:23:39Z
issue,52599,torch::jit::Node::isNondeterministic() does not include aten::feature_dropout or inline kinds,https://github.com/pytorch/pytorch/issues/52599,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-03-03T16:53:04Z
issue,53017,[TorchScript Usability] Optional type does not support `x != None`,https://github.com/pytorch/pytorch/issues/53017,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-03-02T18:11:08Z
issue,53022,[TorchScript Usability] Instance attribute annotation does not seem to take effect,https://github.com/pytorch/pytorch/issues/53022,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-03-02T18:10:18Z
issue,52970,Wrong link in readme for the android nightly version,https://github.com/pytorch/pytorch/issues/52970,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: android, module: docs, oncall: mobile",categorize and decide next maintainer owner/action,2021-03-01T16:26:37Z
issue,50444,[TorchScript] Rough edges discovered during type system rewrite,https://github.com/pytorch/pytorch/issues/50444,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-03-01T15:56:34Z
issue,52788,distributed/optim/test_zero_redundancy_optimizer uses unittest.main() instead of common_utils.run_tests(),https://github.com/pytorch/pytorch/issues/52788,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,categorize and decide next maintainer owner/action,2021-02-25T17:06:54Z
issue,52795,Add format indexed argument support in JIT,https://github.com/pytorch/pytorch/issues/52795,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-02-25T02:07:16Z
issue,52720,[JIT] Finish migrating torch._C._jit_ bindings to torch._C._jit,https://github.com/pytorch/pytorch/issues/52720,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-02-24T03:02:05Z
issue,49935,[FX] intermediate types of empty lists/dicts not preserved during torch.fx tracing,https://github.com/pytorch/pytorch/issues/49935,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-02-23T05:51:09Z
issue,52607,Enable timed barrier in c10d,https://github.com/pytorch/pytorch/issues/52607,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: c10d, oncall: distributed",categorize and decide next maintainer owner/action,2021-02-22T18:44:55Z
issue,46946,Changing order of field assignments breaks TorchScript classes,https://github.com/pytorch/pytorch/issues/46946,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-02-17T23:07:36Z
issue,39690,[Quantization] Output tensor type is lost after serializing and loading back a quantized model,https://github.com/pytorch/pytorch/issues/39690,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-02-15T12:48:57Z
issue,51835,methods decorated with torch.jit._overload_method are not accessible,https://github.com/pytorch/pytorch/issues/51835,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-02-11T22:37:46Z
issue,51379,TorchScript class compilation failure with pybind11-2.6.2,https://github.com/pytorch/pytorch/issues/51379,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-02-11T18:34:25Z
issue,52147,Pointer passed where number is expected for PYTORCH_CUDA_FUSER_JIT_OPT_LEVEL leading to crash,https://github.com/pytorch/pytorch/issues/52147,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-02-11T17:44:14Z
issue,52144,Concatenation of Java tensors (Android for mobile),https://github.com/pytorch/pytorch/issues/52144,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: mobile,categorize and decide next maintainer owner/action,2021-02-11T17:37:40Z
issue,51409,C++ load model error,https://github.com/pytorch/pytorch/issues/51409,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-02-09T22:23:24Z
issue,48703,Caffe2 upsample do not support float times output_size,https://github.com/pytorch/pytorch/issues/48703,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: mobile,categorize and decide next maintainer owner/action,2021-02-08T23:34:44Z
issue,51808,[jit] Keyword-arg expansion for constants,https://github.com/pytorch/pytorch/issues/51808,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-02-08T17:32:28Z
issue,45465,torchscript model size increases when loading and saving it again,https://github.com/pytorch/pytorch/issues/45465,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-02-05T19:38:15Z
issue,40914,Inconsistent handling of torch.Size.__add__,https://github.com/pytorch/pytorch/issues/40914,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-02-04T21:20:37Z
issue,41513,Python object None check not supported,https://github.com/pytorch/pytorch/issues/41513,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-02-04T20:58:41Z
issue,41415,[jit] `set` builtin support,https://github.com/pytorch/pytorch/issues/41415,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-02-04T20:51:26Z
issue,41046,Human-readable names and operations for TorchScript model graphs,https://github.com/pytorch/pytorch/issues/41046,needs triage,47,triage signal present; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-02-04T20:48:14Z
issue,43873,PackedSequence objects that are created within jit.script are not PackedSequence,https://github.com/pytorch/pytorch/issues/43873,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-02-04T20:33:23Z
issue,45199,JIT doesn't support rfloordiv with scalars correctly,https://github.com/pytorch/pytorch/issues/45199,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-02-04T17:48:05Z
issue,48549,JIT script has wrong signature for .norm,https://github.com/pytorch/pytorch/issues/48549,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-02-04T17:26:04Z
issue,48191,[JIT] Contained Module Attributes should be recursively compiled,https://github.com/pytorch/pytorch/issues/48191,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-02-04T01:24:40Z
issue,51136,[TorchScript Usability] error message when invoking __init__ of module types inside TorchScript,https://github.com/pytorch/pytorch/issues/51136,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-02-03T08:08:11Z
issue,51616,[RPC] rref.get_type() should have default return type of future,https://github.com/pytorch/pytorch/issues/51616,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: rpc, oncall: distributed",categorize and decide next maintainer owner/action,2021-02-03T05:31:37Z
issue,50834,Precision of scalars is lost in script mode for binary operations with tensors,https://github.com/pytorch/pytorch/issues/50834,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-02-02T18:06:53Z
issue,51563,[TorchScript Usability] Type-checker too strict for Optional type initialization pattern,https://github.com/pytorch/pytorch/issues/51563,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-02-02T15:29:28Z
issue,48408,TorchScript: Optional sized lists cannot be called with optional scalar values,https://github.com/pytorch/pytorch/issues/48408,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-02-01T17:58:27Z
issue,29092,[FR] script returns subclass of the original module class,https://github.com/pytorch/pytorch/issues/29092,needs triage,47,triage signal present; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-01-28T00:40:30Z
issue,51140,[TorchScript Usability] Non-intuitive error messages for user errors,https://github.com/pytorch/pytorch/issues/51140,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-01-26T20:40:31Z
issue,48928,"When multiple GPUs run multiple processes, it is found that any process not running in GPU 0 will have some more memory (such as 200m) in GPU 0. What is the cause of this?（多个GPU跑多进程时候，发现只要不在0号GPU跑的进程都会在0号GPU多出一些内存(如200M)，请问这是什么情况导致的？）",https://github.com/pytorch/pytorch/issues/48928,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,categorize and decide next maintainer owner/action,2021-01-21T06:46:15Z
issue,49650,Annotating submodule with correct type fails scripting,https://github.com/pytorch/pytorch/issues/49650,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-01-21T00:07:05Z
issue,47964,TorchScript is unable to distinguish int and ScalarType,https://github.com/pytorch/pytorch/issues/47964,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-01-21T00:04:42Z
issue,42673,"Unexpected behavior of ""to"" method inside a torch.jit.script decorated function",https://github.com/pytorch/pytorch/issues/42673,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-01-21T00:04:10Z
issue,50772,[TorchScript Usability] Incorrect default types,https://github.com/pytorch/pytorch/issues/50772,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-01-20T01:31:04Z
issue,50678,[PyTorch Mobile] NNAPI prototype installer no longer available,https://github.com/pytorch/pytorch/issues/50678,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: mobile,categorize and decide next maintainer owner/action,2021-01-18T23:20:51Z
issue,50531,"PyTorch mobile perf recipes, fails when building android",https://github.com/pytorch/pytorch/issues/50531,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: android, oncall: mobile",categorize and decide next maintainer owner/action,2021-01-15T03:00:18Z
issue,49372,[jit] Support `torch.Tensor.numel()` and `torch.Size.numel()` properly in tracing,https://github.com/pytorch/pytorch/issues/49372,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-01-13T01:42:28Z
issue,49569,Update jit/OVERVIEW.md,https://github.com/pytorch/pytorch/issues/49569,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-01-12T00:48:31Z
issue,49585,[JIT] Serialization forward compatibility tests,https://github.com/pytorch/pytorch/issues/49585,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-01-08T22:57:10Z
issue,50127,JIT does not correctly compile custom classes derived from torch.Tensor,https://github.com/pytorch/pytorch/issues/50127,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-01-06T05:21:23Z
issue,16602,Error while installing pytorch,https://github.com/pytorch/pytorch/issues/16602,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: build,categorize and decide next maintainer owner/action,2021-01-04T18:46:29Z
issue,49823,Why trace method used insted of script method in graph creation?,https://github.com/pytorch/pytorch/issues/49823,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: visualization,categorize and decide next maintainer owner/action,2020-12-29T21:35:50Z
issue,8818,MPI causing job to hang --- unresponsive to external (termination) signals,https://github.com/pytorch/pytorch/issues/8818,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,categorize and decide next maintainer owner/action,2020-12-29T18:21:19Z
issue,49536,Pytorch mobile Vulkan API bug during call to ReLU,https://github.com/pytorch/pytorch/issues/49536,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: java, oncall: mobile",categorize and decide next maintainer owner/action,2020-12-21T06:00:55Z
issue,43488,"c10::Error: Couldn’t find an operator for aten::dropout(Tensor input, float p, bool train)",https://github.com/pytorch/pytorch/issues/43488,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: mobile,categorize and decide next maintainer owner/action,2020-12-20T09:16:16Z
issue,43261,OSError: could not find class definition when export torchscript module,https://github.com/pytorch/pytorch/issues/43261,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2020-12-18T07:27:34Z
issue,48869,Tracing fails sanity check,https://github.com/pytorch/pytorch/issues/48869,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2020-12-17T22:15:13Z
issue,49253,Differentiate between objects with a true `Tensor` type and objects with a default `Tensor` type,https://github.com/pytorch/pytorch/issues/49253,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2020-12-15T19:14:41Z
issue,49377,[JIT-DEBUG] Suppress GRAPH dump in profile-graph-executor,https://github.com/pytorch/pytorch/issues/49377,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2020-12-15T00:23:03Z
issue,40696,"Trace model with floor, Inconsistent performance",https://github.com/pytorch/pytorch/issues/40696,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2020-12-10T05:06:05Z
issue,48984,Scripting silently ignores methods of class derived from `NamedTuple`,https://github.com/pytorch/pytorch/issues/48984,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2020-12-10T02:22:45Z
issue,40906,will you support SRU on mobile?,https://github.com/pytorch/pytorch/issues/40906,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: android, oncall: mobile",categorize and decide next maintainer owner/action,2020-12-03T05:14:19Z
issue,48710,RuntimeError: CUDNN_STATUS_BAD_PARAM when testing environment,https://github.com/pytorch/pytorch/issues/48710,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,categorize and decide next maintainer owner/action,2020-12-03T00:11:43Z
issue,48478,Callbacks of Futures shouldn't wait inline on another future,https://github.com/pytorch/pytorch/issues/48478,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: ddp, oncall: distributed",categorize and decide next maintainer owner/action,2020-12-02T01:22:57Z
issue,41419,[jit] Support NamedTuple in tracing,https://github.com/pytorch/pytorch/issues/41419,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, oncall: jit",categorize and decide next maintainer owner/action,2020-12-01T18:19:30Z
issue,48307,Provide torchscript support for staticmethods for classes derived from Enum,https://github.com/pytorch/pytorch/issues/48307,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2020-12-01T08:37:43Z
issue,48410,PyTorch Mobile speed_benchmark_torch crashes on mobile optimized model,https://github.com/pytorch/pytorch/issues/48410,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: mobile,categorize and decide next maintainer owner/action,2020-11-26T01:14:43Z
issue,48045,"[NNC] Improve ""UNSUPPORTED DTYPE"" error messages",https://github.com/pytorch/pytorch/issues/48045,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, oncall: jit",categorize and decide next maintainer owner/action,2020-11-17T15:43:17Z
issue,48094,[JIT][PE] Profiling Executor should re-profile `DifferentiableGraph`,https://github.com/pytorch/pytorch/issues/48094,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2020-11-17T13:37:45Z
issue,48090,[JIT] Constant folding on no-op `aten::mul`,https://github.com/pytorch/pytorch/issues/48090,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2020-11-17T12:47:11Z
issue,48089,[JIT] Optimization pass in profiling executor to fold away conditional branches,https://github.com/pytorch/pytorch/issues/48089,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2020-11-17T12:22:41Z
issue,47342,My demo app uses other package name will crash but org.pytorch.demo,https://github.com/pytorch/pytorch/issues/47342,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: mobile,categorize and decide next maintainer owner/action,2020-11-13T03:29:25Z
issue,47883,`.modules()` is not callable in TorchScript when any submodule is a module interface type (i.e. a class decorated by `@torch.jit.interface`),https://github.com/pytorch/pytorch/issues/47883,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2020-11-13T01:39:08Z
issue,47335,[jit] support length inference for list comprehension,https://github.com/pytorch/pytorch/issues/47335,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2020-11-10T20:03:56Z
issue,47682,Common dunder methods fail when called directly,https://github.com/pytorch/pytorch/issues/47682,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2020-11-10T18:35:56Z
issue,47265,[Feature Request] tensorboard add_graph support for multiple method,https://github.com/pytorch/pytorch/issues/47265,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: visualization,categorize and decide next maintainer owner/action,2020-11-04T16:06:18Z
issue,47177,LayerNorm on Android will cause very high cpu usage,https://github.com/pytorch/pytorch/issues/47177,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: mobile,categorize and decide next maintainer owner/action,2020-11-03T05:26:35Z
issue,47010,[JIT] Tracing turns None Device Input into a Constant,https://github.com/pytorch/pytorch/issues/47010,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2020-11-02T23:26:38Z
issue,46979,[JIT] scripting torchaudio.transforms.MFCC() error,https://github.com/pytorch/pytorch/issues/46979,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: mobile,categorize and decide next maintainer owner/action,2020-11-02T23:23:06Z
issue,46654,Converting object detector model to TorchScript,https://github.com/pytorch/pytorch/issues/46654,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2020-10-27T07:33:43Z
issue,19053,RNN module weights not compacted,https://github.com/pytorch/pytorch/issues/19053,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2020-10-22T03:02:26Z
issue,46505,Cannot script when a static method or cross module function calls custom op,https://github.com/pytorch/pytorch/issues/46505,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: cpp-extensions, module: torchbind, oncall: jit",categorize and decide next maintainer owner/action,2020-10-20T17:53:10Z
issue,46368,[JIT] Inconsistent behavior of torchScript parser in python and c++,https://github.com/pytorch/pytorch/issues/46368,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2020-10-20T17:31:33Z
issue,46574,Mismatch between the documentation and implementation of caffe2 PReLU operator,https://github.com/pytorch/pytorch/issues/46574,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: docs,categorize and decide next maintainer owner/action,2020-10-20T15:32:12Z
issue,43125,"Non-conforming variable identifiers in JIT code errors / printouts: e.g. ""Tensor 0""",https://github.com/pytorch/pytorch/issues/43125,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2020-10-08T02:10:35Z
issue,40500,Not able to launch tensorboard using pytorch,https://github.com/pytorch/pytorch/issues/40500,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: tensorboard, oncall: visualization",categorize and decide next maintainer owner/action,2020-10-07T09:04:03Z
issue,45327,Tensorboard summary_iterator,https://github.com/pytorch/pytorch/issues/45327,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: tensorboard, oncall: visualization",categorize and decide next maintainer owner/action,2020-10-05T16:54:58Z
issue,17126,Support callables in scripted functions,https://github.com/pytorch/pytorch/issues/17126,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2020-10-05T12:21:07Z
issue,45208,Adding @torch.no_grad() to forward () causes undefined value in torch.jit.script(),https://github.com/pytorch/pytorch/issues/45208,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2020-09-29T17:12:13Z
issue,43323,Set SummaryWriter step globally,https://github.com/pytorch/pytorch/issues/43323,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: visualization,categorize and decide next maintainer owner/action,2020-09-24T13:52:43Z
issue,44608,building torchscript extension results in INVALID TYPE: Only int64_t and bool are supported as an integral argument type custom_class,https://github.com/pytorch/pytorch/issues/44608,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2020-09-23T04:26:11Z
issue,44768,[JIT] function decorated with @torch.no_grad() can not be exported when another function is called inside it.,https://github.com/pytorch/pytorch/issues/44768,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2020-09-21T22:12:30Z
issue,44992,[jit] cann't pass class objects across boundary of tracing/scripting,https://github.com/pytorch/pytorch/issues/44992,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2020-09-21T22:11:07Z
issue,45012,torch.jit saves nonpersistent buffers,https://github.com/pytorch/pytorch/issues/45012,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2020-09-21T21:57:50Z
issue,44800,torchscript continue training,https://github.com/pytorch/pytorch/issues/44800,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2020-09-21T21:16:56Z
issue,39240,[mobile] Running on armeabi-v7a is inconsistent with arm64-v8a,https://github.com/pytorch/pytorch/issues/39240,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: mobile,categorize and decide next maintainer owner/action,2020-09-20T11:54:32Z
issue,44503,TestJitGeneratedModule.test_nn_CTCLoss_lengths_intlists fails if not skipped,https://github.com/pytorch/pytorch/issues/44503,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2020-09-16T14:17:39Z
issue,44446,[jit] class type optional type annotation not working properly,https://github.com/pytorch/pytorch/issues/44446,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2020-09-14T21:58:21Z
issue,44630,No pytorch_jni in StartLocal's zip,https://github.com/pytorch/pytorch/issues/44630,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: mobile,categorize and decide next maintainer owner/action,2020-09-14T17:03:46Z
issue,43119,Tracing can leak CUDA memory when using tensor constants,https://github.com/pytorch/pytorch/issues/43119,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2020-08-25T21:27:00Z
issue,43378,[JIT] Using Any type variable inside can produce incorrect results due to type unification problems,https://github.com/pytorch/pytorch/issues/43378,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2020-08-24T22:21:35Z
issue,42947,[jit] std::exception_ptr can't be torchbind/convert to ivalue,https://github.com/pytorch/pytorch/issues/42947,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2020-08-24T20:35:50Z
issue,42948,[jit] TorchBind std::chrono::milliseconds,https://github.com/pytorch/pytorch/issues/42948,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2020-08-24T20:35:08Z
issue,41683,'torch::jit::script::ErrorReport' from 'torch::jit::load('modelpath') in Visual Studio 2017,https://github.com/pytorch/pytorch/issues/41683,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2020-08-22T06:22:36Z
issue,43409,Switch C10_EXPORT_CAFFE2_OP_TO_C10 to new operator registration API,https://github.com/pytorch/pytorch/issues/43409,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: bootcamp,categorize and decide next maintainer owner/action,2020-08-21T22:37:37Z
issue,43328,torch.utils.tensorboard.SummaryWriter.add_histogram() execution time explodes with each epoch.,https://github.com/pytorch/pytorch/issues/43328,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: visualization,categorize and decide next maintainer owner/action,2020-08-20T17:05:11Z
issue,43134,[jit] need a better way to handle mix CPU/GPU (Inference/Training) for tracing,https://github.com/pytorch/pytorch/issues/43134,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2020-08-19T17:49:31Z
issue,41291,Return list of tuples from custom operator,https://github.com/pytorch/pytorch/issues/41291,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2020-08-10T21:40:53Z
issue,38744,[jit] `jit.annotate` and multiple compilation of functions with different types,https://github.com/pytorch/pytorch/issues/38744,needs triage,47,triage signal present; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2020-08-08T21:30:39Z
issue,42170,Tracing a[b] records converting b to fixed device type of a,https://github.com/pytorch/pytorch/issues/42170,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2020-08-07T22:44:13Z
issue,42039,Segmentation fault (core dumped) when running optimize_for_mobile,https://github.com/pytorch/pytorch/issues/42039,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: mobile,categorize and decide next maintainer owner/action,2020-08-03T14:48:41Z
issue,42355,nn::Sequential Link error in Android version,https://github.com/pytorch/pytorch/issues/42355,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: android, oncall: mobile",categorize and decide next maintainer owner/action,2020-08-01T08:18:51Z
issue,42379,TensorBoard Summaries Written with `SummaryWriter.add_scalars()` are not Purged,https://github.com/pytorch/pytorch/issues/42379,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: tensorboard, oncall: visualization",categorize and decide next maintainer owner/action,2020-07-31T21:43:32Z
issue,42308,Not able to run quantized model on android,https://github.com/pytorch/pytorch/issues/42308,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: mobile,categorize and decide next maintainer owner/action,2020-07-30T19:50:57Z
issue,42145,Support Python type() in TorchScript,https://github.com/pytorch/pytorch/issues/42145,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2020-07-28T18:56:04Z
issue,42081,Document optimizations enabled by JIT scripting and tracing (TorchScript),https://github.com/pytorch/pytorch/issues/42081,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2020-07-28T17:51:37Z
issue,39724,pickle_save on mobile (no longer works with 1.5.0 release),https://github.com/pytorch/pytorch/issues/39724,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: mobile,categorize and decide next maintainer owner/action,2020-07-24T06:53:58Z
issue,41433,[JIT][to-backend] `selective_to_backend` infra,https://github.com/pytorch/pytorch/issues/41433,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2020-07-14T23:31:51Z
issue,41418,[jit] support `rpc_remote` and `rpc_sync`,https://github.com/pytorch/pytorch/issues/41418,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2020-07-14T18:41:18Z
issue,41141,Sum of two tuples becomes a list in torchscript,https://github.com/pytorch/pytorch/issues/41141,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2020-07-09T14:24:06Z
issue,39569,[JIT] OrderedDict doesn't support custom objects,https://github.com/pytorch/pytorch/issues/39569,needs triage,47,triage signal present; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2020-06-05T19:04:50Z
issue,39539,[JIT] Print out mutation in IR Dumps,https://github.com/pytorch/pytorch/issues/39539,needs triage,47,triage signal present; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2020-06-04T21:44:11Z
issue,38614,`SummaryWriter.add_graph` borks with simple example,https://github.com/pytorch/pytorch/issues/38614,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: visualization,categorize and decide next maintainer owner/action,2020-06-04T17:19:57Z
issue,39309,[JIT] nn.ModuleList loses None objects inside it after scripting,https://github.com/pytorch/pytorch/issues/39309,needs triage,47,triage signal present; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2020-06-04T15:29:30Z
issue,38312,Status of support for training on mobile,https://github.com/pytorch/pytorch/issues/38312,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: mobile,categorize and decide next maintainer owner/action,2020-06-02T16:31:05Z
issue,39292,Tensorboard，graph in pytorch 1.4 is more complicated than pytorch 1.1？,https://github.com/pytorch/pytorch/issues/39292,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: tensorboard, oncall: visualization",categorize and decide next maintainer owner/action,2020-06-01T01:33:45Z
issue,36211,[JIT] support self.named_buffers and self.named_parameters in TorchScript,https://github.com/pytorch/pytorch/issues/36211,needs triage,47,triage signal present; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2020-05-31T07:17:24Z
issue,38273,Adding model metadata in TorchScript model file,https://github.com/pytorch/pytorch/issues/38273,needs triage,47,"triage signal present; old update age or low-urgency/stale label; component/oncall labels: module: serialization, oncall: jit",categorize and decide next maintainer owner/action,2020-05-20T23:56:56Z
issue,37740,torch Summary writer does not display torchvision.io.read_video output,https://github.com/pytorch/pytorch/issues/37740,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: tensorboard, oncall: visualization",categorize and decide next maintainer owner/action,2020-05-04T22:01:25Z
issue,36926,JIT string ops and other miscellaneous ops probably shouldn't be in aten namespace,https://github.com/pytorch/pytorch/issues/36926,needs triage,47,triage signal present; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2020-04-27T14:50:42Z
issue,37016,Guidance on implementing a new backend,https://github.com/pytorch/pytorch/issues/37016,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: mobile,categorize and decide next maintainer owner/action,2020-04-22T18:34:20Z
issue,36603,TorchScript Support for Named Tensors,https://github.com/pytorch/pytorch/issues/36603,needs triage,47,triage signal present; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2020-04-17T20:29:15Z
issue,33281,"Uninitialised value was created by a stack allocation, reported by valgrind",https://github.com/pytorch/pytorch/issues/33281,needs triage,47,triage signal present; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2020-04-10T07:20:36Z
issue,36182,## 🐛 Bug: QNNPACK tests failing on master on Nexus 6,https://github.com/pytorch/pytorch/issues/36182,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: mobile,categorize and decide next maintainer owner/action,2020-04-07T23:33:04Z
issue,35526,RuntimeError during converting Reformer model to TorchScript,https://github.com/pytorch/pytorch/issues/35526,needs triage,47,triage signal present; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2020-04-03T21:16:59Z
issue,35606,"Could not find any similar ops to ""foo..."" in the `Libtorch`",https://github.com/pytorch/pytorch/issues/35606,needs triage,47,triage signal present; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2020-04-01T18:39:32Z
issue,35095,"tensorboard add_graph 's ""operator_export_type""",https://github.com/pytorch/pytorch/issues/35095,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: tensorboard, oncall: visualization",categorize and decide next maintainer owner/action,2020-03-23T18:18:27Z
issue,33631,bitmapToFloat32Tensor() 1 channel Tensor [feature] [mobile],https://github.com/pytorch/pytorch/issues/33631,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: java, oncall: mobile",categorize and decide next maintainer owner/action,2020-03-22T17:41:31Z
issue,33124,QNNPACK: GNU aarch64 assembler does not support 4s on neon mov,https://github.com/pytorch/pytorch/issues/33124,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: mobile,categorize and decide next maintainer owner/action,2020-03-21T16:50:07Z
issue,34851,[jit] `__hash__` magic method is missing,https://github.com/pytorch/pytorch/issues/34851,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2020-03-17T18:25:21Z
issue,29125,_compared_saved_loaded doesn't work with torch.tensor constants,https://github.com/pytorch/pytorch/issues/29125,needs triage,47,triage signal present; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2020-03-16T19:13:35Z
issue,33993,ModuleList doesn't support slicing in TorchScript,https://github.com/pytorch/pytorch/issues/33993,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2020-02-29T01:24:55Z
issue,29128,JIT should respect SKIP_PYTHON_BINDINGS and SKIP_PYTHON_BINDINGS_SIGNATURES,https://github.com/pytorch/pytorch/issues/29128,needs triage,47,triage signal present; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2020-02-28T18:52:17Z
issue,28267,Problem with jit TorchScript while copying data between GRUs,https://github.com/pytorch/pytorch/issues/28267,needs triage,47,triage signal present; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2020-02-28T18:52:15Z
issue,28191,[JIT] Allow user to provide aliasing information on input tensors and model parameters,https://github.com/pytorch/pytorch/issues/28191,needs triage,47,triage signal present; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2020-02-28T18:52:15Z
issue,33810,Wrongly detected: Division of ints in TorchScript uses Python 3 true division semantics,https://github.com/pytorch/pytorch/issues/33810,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2020-02-26T11:33:19Z
issue,33285,iOS libtorch superpoint model bug,https://github.com/pytorch/pytorch/issues/33285,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: ios, module: macos, oncall: mobile",categorize and decide next maintainer owner/action,2020-02-22T18:41:35Z
issue,33110,InlineAutodiffSubgraphs in JIT inlines non-differentiable custom groups unexpectedly.,https://github.com/pytorch/pytorch/issues/33110,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: custom-operators, oncall: jit",categorize and decide next maintainer owner/action,2020-02-14T20:00:30Z
issue,9595,[JIT] Unify python -> SugaredValue construction + fix inlining of graphs with Tuple-typed inputs,https://github.com/pytorch/pytorch/issues/9595,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2020-02-11T02:11:14Z
issue,30863,"[TensorBoard] Installed tensorboard 1.14.0, TestTensorBoardWriter.test_writer failed.",https://github.com/pytorch/pytorch/issues/30863,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: visualization,categorize and decide next maintainer owner/action,2020-01-11T12:19:32Z
issue,31603,Typo in `torch.utils.tensorboard.add_image`,https://github.com/pytorch/pytorch/issues/31603,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: visualization,categorize and decide next maintainer owner/action,2020-01-11T12:11:40Z
issue,7904,TracedModules don't support parameter sharing between modules,https://github.com/pytorch/pytorch/issues/7904,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2019-12-30T13:38:31Z
issue,30812,Tensorboard graph failing to nest modules,https://github.com/pytorch/pytorch/issues/30812,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: visualization,categorize and decide next maintainer owner/action,2019-12-17T10:04:38Z
issue,18312,[jit] Trace perform weirdly on BatchNorm,https://github.com/pytorch/pytorch/issues/18312,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2019-11-22T03:26:53Z
issue,15116,torch.save does not work if nn.Module has partial JIT.,https://github.com/pytorch/pytorch/issues/15116,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2019-11-21T19:08:31Z
issue,14270,Pytorch C++ API with cuda : Expected object of backend CPU but got backend CUDA for sequence element 1 in sequence argument at position #1 'tensors',https://github.com/pytorch/pytorch/issues/14270,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2019-10-14T07:51:09Z
issue,18975,Tracing and Scripting nn.Conv2d shows different op in the graph,https://github.com/pytorch/pytorch/issues/18975,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2019-07-29T16:50:57Z
issue,9797,Build torch as a submodule with static linking doesn't work (CAFFE2_PERF_WITH_AVX2 is not defined),https://github.com/pytorch/pytorch/issues/9797,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: static linking,categorize and decide next maintainer owner/action,2019-06-13T23:05:26Z
issue,21255,IsType<T>() ASSERT FAILED [Detectron e2e_mask_rcnn_R-50-C4_1x.yaml],https://github.com/pytorch/pytorch/issues/21255,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: assert failure,categorize and decide next maintainer owner/action,2019-06-04T15:43:54Z
issue,16263,[JIT] Support C++ front end module and JIT interop,https://github.com/pytorch/pytorch/issues/16263,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2019-05-29T23:07:18Z
issue,15421,JIT is not compatible with data parallel,https://github.com/pytorch/pytorch/issues/15421,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2019-05-14T00:57:08Z
issue,6898,[Caffe2] cudnn versions compatibility issue.,https://github.com/pytorch/pytorch/issues/6898,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: cudnn,categorize and decide next maintainer owner/action,2019-04-09T16:17:14Z
issue,18376,Cleanup code in register_c10_ops.cpp,https://github.com/pytorch/pytorch/issues/18376,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2019-04-06T16:06:32Z
issue,18380,"C++ inference (CPU-only) stalls in Android, and crashes on Mac/Linux",https://github.com/pytorch/pytorch/issues/18380,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: android,categorize and decide next maintainer owner/action,2019-04-06T16:05:03Z
issue,16293,Unexpected behavior of jit.trace when PYTORCH_JIT=0,https://github.com/pytorch/pytorch/issues/16293,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2019-03-22T16:21:57Z
issue,15043,[JIT] jit.trace fails with custom GRUs and CUDA when the sequence is longer,https://github.com/pytorch/pytorch/issues/15043,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2019-03-11T21:09:23Z
issue,14750,[caffe2] ConvTranspose with group attribute,https://github.com/pytorch/pytorch/issues/14750,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: op-unification,categorize and decide next maintainer owner/action,2019-01-07T07:28:48Z
issue,14484,prim::ConstantChunk derivative formula doesn't handle undefined inputs,https://github.com/pytorch/pytorch/issues/14484,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2018-11-28T18:20:39Z
issue,14332,[jit][script] support slicing with tensor literals,https://github.com/pytorch/pytorch/issues/14332,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2018-11-23T06:27:05Z
issue,13477,[jit] restrict promotion of single-element arguments to lists,https://github.com/pytorch/pytorch/issues/13477,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2018-11-01T20:57:14Z
issue,13170,Allow traced modules to return dictionaries,https://github.com/pytorch/pytorch/issues/13170,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2018-10-29T17:38:28Z
issue,11967,Jit cannot trace autograd for certain operator,https://github.com/pytorch/pytorch/issues/11967,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2018-09-24T17:43:00Z
issue,11532,[JIT][tracer] Slicing shape is specialized to tensor rank,https://github.com/pytorch/pytorch/issues/11532,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2018-09-11T18:24:53Z
issue,8453,"[JIT] Generalize checkTrace() to also compare gradients of parameters, not just inputs",https://github.com/pytorch/pytorch/issues/8453,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2018-07-10T16:24:16Z
issue,8556,[JIT] Interleaved C++-Python execution loses inner Python stacks,https://github.com/pytorch/pytorch/issues/8556,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2018-06-15T15:35:11Z
issue,6465,[caffe2] ChannelShuffle example,https://github.com/pytorch/pytorch/issues/6465,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: docs,categorize and decide next maintainer owner/action,2018-04-10T18:31:10Z
issue,6010,Develop a strategy for writing leak tests for pytorch.,https://github.com/pytorch/pytorch/issues/6010,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2018-03-28T17:50:31Z
issue,6007,Clean up the extra copies that occur in the execution engine and other places,https://github.com/pytorch/pytorch/issues/6007,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2018-03-28T17:50:22Z
pr,187404,[inductor][cutlass] Normalize reshaped external reads in EVT epilogue fusion,https://github.com/pytorch/pytorch/pull/187404,ready for maintainer decision,45,"CI lane labels: ciflow/inductor, ciflow/torchtitan; open non-draft PR has no obvious automated blocker",review diff and decide acceptance path,2026-07-09T01:25:28Z
pr,188029,[dynamo] Add namespace for cell variables that collide with locals,https://github.com/pytorch/pytorch/pull/188029,ready for maintainer decision,45,"CI lane labels: ciflow/inductor, ciflow/torchtitan; open non-draft PR has no obvious automated blocker",review diff and decide acceptance path,2026-07-09T01:03:53Z
pr,189306,"[dynamo, 3.15] Fix test_bytecode_utils",https://github.com/pytorch/pytorch/pull/189306,ready for maintainer decision,45,CI lane labels: ciflow/inductor; open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-07-09T00:20:53Z
pr,189022,[dynamo] Support itertools.count keyword args and repr,https://github.com/pytorch/pytorch/pull/189022,ready for maintainer decision,45,"CI lane labels: ciflow/inductor, ciflow/torchtitan; open non-draft PR has no obvious automated blocker",review diff and decide acceptance path,2026-07-08T23:49:13Z
pr,188176,[dynamo] fix test_perf.py tests under canonicalization,https://github.com/pytorch/pytorch/pull/188176,ready for maintainer decision,45,CI lane labels: ciflow/inductor; open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-07-08T23:22:21Z
pr,189188,flex_gemm: support scalar captured epilogue args,https://github.com/pytorch/pytorch/pull/189188,ready for maintainer decision,45,"CI lane labels: ciflow/inductor, ciflow/torchtitan; open non-draft PR has no obvious automated blocker",review diff and decide acceptance path,2026-07-08T23:12:52Z
pr,188468,flex_gemm: add local reduce analysis helpers,https://github.com/pytorch/pytorch/pull/188468,ready for maintainer decision,45,"CI lane labels: ciflow/inductor, ciflow/torchtitan; open non-draft PR has no obvious automated blocker",review diff and decide acceptance path,2026-07-08T23:12:48Z
pr,180020,Add a debug tool to warn on NULL stream use,https://github.com/pytorch/pytorch/pull/180020,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-07-08T21:58:38Z
pr,189155,Add CUDA Support for Fused `_add_relu` With Half and BFloat16,https://github.com/pytorch/pytorch/pull/189155,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-07-08T21:37:10Z
pr,189274,[dynamo] Route set/dict-view/list dunder methods through tp_slots,https://github.com/pytorch/pytorch/pull/189274,ready for maintainer decision,45,"CI lane labels: ciflow/inductor, ciflow/torchtitan; open non-draft PR has no obvious automated blocker",review diff and decide acceptance path,2026-07-08T21:27:00Z
pr,188802,[MPS] Migrate conv3d to metal kernels,https://github.com/pytorch/pytorch/pull/188802,ready for maintainer decision,45,CI lane labels: ciflow/mps; open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-07-08T21:26:44Z
pr,189199,[DTensor] Fix spurious cross-device FakeTensor mismatch in multi-threaded tests,https://github.com/pytorch/pytorch/pull/189199,ready for maintainer decision,45,"CI lane labels: ciflow/dtensor, ciflow/inductor, ciflow/torchtitan; open non-draft PR has no obvious automated blocker",review diff and decide acceptance path,2026-07-08T21:18:46Z
pr,189303,"Reapply ""Dispatch index_put_ to fill_ for scalar values (#188267 / #189105)""",https://github.com/pytorch/pytorch/pull/189303,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-07-08T21:16:25Z
pr,188991,[NVGEMM] Support `cutlass.operators`,https://github.com/pytorch/pytorch/pull/188991,ready for maintainer decision,45,"CI lane labels: ciflow/inductor, ciflow/torchtitan; open non-draft PR has no obvious automated blocker",review diff and decide acceptance path,2026-07-08T21:16:04Z
pr,189102,[TEST][Inductor][CUDA] Bump tolerances for `test_comprehensive_combinations_cuda_float16`,https://github.com/pytorch/pytorch/pull/189102,ready for maintainer decision,45,CI lane labels: ciflow/inductor; open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-07-08T21:13:45Z
pr,189037,Preserve strides for CuTeDSL grouped MM fake tensors,https://github.com/pytorch/pytorch/pull/189037,ready for maintainer decision,45,"CI lane labels: ciflow/inductor, ciflow/torchtitan; open non-draft PR has no obvious automated blocker",review diff and decide acceptance path,2026-07-08T21:13:02Z
pr,189100,[ROCm] Fix test_is_pinned_no_context for Python 3.14 forkserver default,https://github.com/pytorch/pytorch/pull/189100,ready for maintainer decision,45,CI lane labels: ciflow/trunk; open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-07-08T21:11:41Z
pr,189148,Add maybe_unused label to strides when strides is unused,https://github.com/pytorch/pytorch/pull/189148,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-07-08T21:00:41Z
pr,189182,make the return more specified,https://github.com/pytorch/pytorch/pull/189182,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-07-08T20:55:38Z
pr,189053,[dynamo] Support integer arithmetic on a sourceless id()/hash() FakeIdVariable,https://github.com/pytorch/pytorch/pull/189053,ready for maintainer decision,45,"CI lane labels: ciflow/inductor, ciflow/torchtitan; open non-draft PR has no obvious automated blocker",review diff and decide acceptance path,2026-07-08T20:53:25Z
pr,189052,[dynamo] Model deque iterators with mutation-during-iteration detection,https://github.com/pytorch/pytorch/pull/189052,ready for maintainer decision,45,"CI lane labels: ciflow/inductor, ciflow/torchtitan; open non-draft PR has no obvious automated blocker",review diff and decide acceptance path,2026-07-08T20:53:24Z
pr,188814,"Fix typos in comments and docstrings across ao, distributed, masked, nn, and sparse modules",https://github.com/pytorch/pytorch/pull/188814,ready for maintainer decision,45,"CI lane labels: ciflow/torchtitan, ciflow/trunk; open non-draft PR has no obvious automated blocker",review diff and decide acceptance path,2026-07-08T20:53:13Z
pr,188870,"Fix typos in comments, docstrings, and error messages",https://github.com/pytorch/pytorch/pull/188870,ready for maintainer decision,45,"CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk; open non-draft PR has no obvious automated blocker",review diff and decide acceptance path,2026-07-08T20:52:54Z
pr,180230,[TEST][CUDA][TF32] Update TF32 test decorators to use new fp32 precision path,https://github.com/pytorch/pytorch/pull/180230,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-07-08T20:48:25Z
pr,189160,Use SafePyObject in PyAnomalyMetadata and PySavedVariableHooks,https://github.com/pytorch/pytorch/pull/189160,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-07-08T20:39:09Z
pr,189123,cuda graphs: accept None && float graph-partition inputs,https://github.com/pytorch/pytorch/pull/189123,ready for maintainer decision,45,"CI lane labels: ciflow/inductor, ciflow/torchtitan; open non-draft PR has no obvious automated blocker",review diff and decide acceptance path,2026-07-08T20:36:26Z
pr,189146,[MPS] softmax: non-last-dim and huge-row native Metal kernels (perf specializations),https://github.com/pytorch/pytorch/pull/189146,ready for maintainer decision,45,CI lane labels: ciflow/mps; open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-07-08T20:22:40Z
pr,188973,"Reapply ""Avoid redundant scalar-to-tensor conversion in `real`/`imag`setters on complex tensors (#188603)",https://github.com/pytorch/pytorch/pull/188973,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-07-08T20:20:39Z
pr,189162,[inductor] Use non-NaN-propagating fmax in softmax persistent reduction,https://github.com/pytorch/pytorch/pull/189162,ready for maintainer decision,45,"CI lane labels: ciflow/inductor, ciflow/torchtitan; open non-draft PR has no obvious automated blocker",review diff and decide acceptance path,2026-07-08T20:19:52Z
pr,188681,"[Inductor] Use CUDA toolkit's libdevice by default, fallback to Triton's bundled if not found",https://github.com/pytorch/pytorch/pull/188681,ready for maintainer decision,45,"CI lane labels: ciflow/inductor, ciflow/torchtitan; open non-draft PR has no obvious automated blocker",review diff and decide acceptance path,2026-07-08T20:19:31Z
pr,189251,[Dynamo] Store backend tags in the registry,https://github.com/pytorch/pytorch/pull/189251,ready for maintainer decision,45,"CI lane labels: ciflow/inductor, ciflow/torchtitan; open non-draft PR has no obvious automated blocker",review diff and decide acceptance path,2026-07-08T20:18:59Z
pr,189163,Enable expandable_segments on Windows (port NVML driver API as prerequisite),https://github.com/pytorch/pytorch/pull/189163,ready for maintainer decision,45,CI lane labels: ciflow/trunk; open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-07-08T20:01:21Z
pr,185206,Make minifier_tests more device agnostic,https://github.com/pytorch/pytorch/pull/185206,ready for maintainer decision,45,"CI lane labels: ciflow/inductor, ciflow/torchtitan; open non-draft PR has no obvious automated blocker",review diff and decide acceptance path,2026-07-08T18:09:41Z
pr,189260,PERF: improve cuda/xpu Stream.__hash__,https://github.com/pytorch/pytorch/pull/189260,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-07-08T17:38:31Z
pr,185917,[Test] Refactor test_autocast.py: make behavior/cache tests device-agnostic,https://github.com/pytorch/pytorch/pull/185917,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-07-08T17:33:19Z
pr,188366,[functorch] Inline codegen'd epilogue fns directly into the runtime wrapper,https://github.com/pytorch/pytorch/pull/188366,ready for maintainer decision,45,CI lane labels: ciflow/inductor; open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-07-08T16:57:44Z
pr,188367,[functorch] Extract _compile_and_exec_source into a generic codegen.py,https://github.com/pytorch/pytorch/pull/188367,ready for maintainer decision,45,CI lane labels: ciflow/inductor; open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-07-08T16:57:43Z
pr,188376,[functorch] Add source_emit: reconstruct live values as standalone source,https://github.com/pytorch/pytorch/pull/188376,ready for maintainer decision,45,CI lane labels: ciflow/inductor; open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-07-08T16:57:40Z
pr,182024,[docs] Fix typo in data page,https://github.com/pytorch/pytorch/pull/182024,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-07-08T14:17:29Z
pr,187457,[MPS] Migrate log_softmax to native Metal kernels,https://github.com/pytorch/pytorch/pull/187457,ready for maintainer decision,45,CI lane labels: ciflow/mps; open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-07-08T09:53:32Z
pr,184678,[5/N][Test] Migrate TestMeta device-agnostic skips to `@skipOps`,https://github.com/pytorch/pytorch/pull/184678,ready for maintainer decision,45,CI lane labels: ciflow/trunk; open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-07-08T09:31:55Z
pr,187456,[MPS] Migrate softmax and _softmax_backward to native Metal kernels (core),https://github.com/pytorch/pytorch/pull/187456,ready for maintainer decision,45,CI lane labels: ciflow/mps; open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-07-08T09:10:26Z
pr,186345,[Test] Make test_comparison_utils.py device-generic,https://github.com/pytorch/pytorch/pull/186345,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-07-08T08:40:27Z
pr,182987,[Inductor] Parallel batched precompile in combo sequential autotune,https://github.com/pytorch/pytorch/pull/182987,ready for maintainer decision,45,"CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk; open non-draft PR has no obvious automated blocker",review diff and decide acceptance path,2026-07-08T08:18:47Z
pr,180209,XPU oneDNN: thread-local LRU cache for FP matmul primitives + post_ops deduplication,https://github.com/pytorch/pytorch/pull/180209,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-07-08T08:18:40Z
pr,181109,tools/stats: add script to generate test-times.json from local JUnit XML reports,https://github.com/pytorch/pytorch/pull/181109,ready for maintainer decision,45,"CI lane labels: ciflow/inductor, ciflow/torchtitan; open non-draft PR has no obvious automated blocker",review diff and decide acceptance path,2026-07-08T04:49:48Z
pr,182377,[DTensor] SymmetricMemory-backed DTensors,https://github.com/pytorch/pytorch/pull/182377,ready for maintainer decision,45,"CI lane labels: ciflow/dtensor, ciflow/inductor, ciflow/torchtitan; open non-draft PR has no obvious automated blocker",review diff and decide acceptance path,2026-07-08T04:44:20Z
pr,186273,[TEST][CUDA][TF32] Re-enable TF32 setter check test,https://github.com/pytorch/pytorch/pull/186273,ready for maintainer decision,45,CI lane labels: ciflow/trunk; open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-07-08T03:39:16Z
pr,189011,include activation epilogue in GemmAndBiasParams::Signature(),https://github.com/pytorch/pytorch/pull/189011,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-07-08T02:58:16Z
pr,189184,[inductor][fb] Whole-graph batch_addmm->bmm fusion for independent same-shape addmm (#189184),https://github.com/pytorch/pytorch/pull/189184,ready for maintainer decision,45,"CI lane labels: ciflow/inductor, ciflow/torchtitan; open non-draft PR has no obvious automated blocker",review diff and decide acceptance path,2026-07-07T23:17:51Z
pr,176830,feat: add polyfills for setitem/delitem op,https://github.com/pytorch/pytorch/pull/176830,ready for maintainer decision,45,"CI lane labels: ciflow/inductor, ciflow/torchtitan; open non-draft PR has no obvious automated blocker",review diff and decide acceptance path,2026-07-07T22:49:45Z
pr,189177,[inductor] Drop unused cache_info from CacheCompiledArtifact._validate_and_unpack,https://github.com/pytorch/pytorch/pull/189177,ready for maintainer decision,45,"CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk; open non-draft PR has no obvious automated blocker",review diff and decide acceptance path,2026-07-07T21:08:24Z
pr,188999,Add RISC-V CPU pause hint to cpu_atomic_add_float spin loop,https://github.com/pytorch/pytorch/pull/188999,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-07-07T20:50:48Z
pr,182785,[Inductor] Fix FX wrapper codegen for symbolic floor/ceil offsets in ReinterpretView (#182785),https://github.com/pytorch/pytorch/pull/182785,ready for maintainer decision,45,"CI lane labels: ciflow/inductor, ciflow/torchtitan; open non-draft PR has no obvious automated blocker",review diff and decide acceptance path,2026-07-07T20:12:15Z
pr,188250,Remove dead USE_ROCM block in CUDADeviceAssertionHost,https://github.com/pytorch/pytorch/pull/188250,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-07-07T20:00:42Z
pr,186245,[1/N] Optimize PyTorch CUDA build with LLVM BOLT,https://github.com/pytorch/pytorch/pull/186245,ready for maintainer decision,45,"CI lane labels: ciflow/binaries, ciflow/docker, ciflow/trunk; open non-draft PR has no obvious automated blocker",review diff and decide acceptance path,2026-07-07T19:56:04Z
pr,188946,[Bugfix][FSDP] Capture out-of-place redistribute() return value in optimizer state reconstruction,https://github.com/pytorch/pytorch/pull/188946,ready for maintainer decision,45,"CI lane labels: ciflow/inductor, ciflow/torchtitan; open non-draft PR has no obvious automated blocker",review diff and decide acceptance path,2026-07-07T18:20:48Z
pr,188816,"[Metal][ops] remove nonzero INT_MAX limit, widen to 64-bit (#149325)",https://github.com/pytorch/pytorch/pull/188816,ready for maintainer decision,45,CI lane labels: ciflow/mps; open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-07-07T18:14:26Z
pr,188434,[dcp] Free staged dict on caller thread via weakref.finalize in async_save (#188434),https://github.com/pytorch/pytorch/pull/188434,ready for maintainer decision,45,"CI lane labels: ciflow/torchtitan, ciflow/trunk; open non-draft PR has no obvious automated blocker",review diff and decide acceptance path,2026-07-07T17:56:04Z
pr,187901,[TEST] Prevent PLATFORM_SUPPORTS_* from skipping tests on privateuse1 backends,https://github.com/pytorch/pytorch/pull/187901,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-07-07T17:45:17Z
pr,189089,[fmha_fwd][rocm] Extend CK ck_tile FMHA seqlen-pad guard to the GQA swap path,https://github.com/pytorch/pytorch/pull/189089,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-07-07T17:42:50Z
pr,189108,[Inductor] Bound exact index expressions for int32 indexing eligibility,https://github.com/pytorch/pytorch/pull/189108,ready for maintainer decision,45,"CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk; open non-draft PR has no obvious automated blocker",review diff and decide acceptance path,2026-07-07T17:42:13Z
pr,184718,[Test] Make test_testing.py device-agnostic for out-of-tree accelerators,https://github.com/pytorch/pytorch/pull/184718,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-07-07T17:17:47Z
pr,189028,[HOP] Moved branch clamping to user surface for switch,https://github.com/pytorch/pytorch/pull/189028,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-07-07T16:41:56Z
pr,188951,Add ROCm Docker release images,https://github.com/pytorch/pytorch/pull/188951,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-07-07T13:51:06Z
pr,184742,[Test] Make test_expanded_weights.py device-agnostic for out-of-tree backends,https://github.com/pytorch/pytorch/pull/184742,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-07-07T13:41:45Z
pr,189143,[Test] Make test_native_mha.py testcases device-agnostic,https://github.com/pytorch/pytorch/pull/189143,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-07-07T12:33:39Z
pr,189140,[Test] Update device type tests in TestVarlenAttention for device-generic,https://github.com/pytorch/pytorch/pull/189140,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-07-07T11:39:31Z
pr,188939,[profiler][cupti] Environment + per-kernel cycle GPU counters in the JSON export,https://github.com/pytorch/pytorch/pull/188939,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-07-07T05:51:28Z
pr,187898,"[profiler][cupti] Perfetto-native (.pftrace) export, full chrome-args parity",https://github.com/pytorch/pytorch/pull/187898,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-07-07T05:51:14Z
pr,188867,Add USE_KLEIDIAI_SME option to skip SME/SME2 kernels on CPUs without SME,https://github.com/pytorch/pytorch/pull/188867,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-07-07T05:50:25Z
pr,185486,[OpenReg] Add minimal implementation for inductor integration in openreg,https://github.com/pytorch/pytorch/pull/185486,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-07-07T05:41:40Z
pr,189101,[inductor] Fix data-dependent GuardOnDataDependentSymNode in get_fill_order,https://github.com/pytorch/pytorch/pull/189101,ready for maintainer decision,45,"CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk; open non-draft PR has no obvious automated blocker",review diff and decide acceptance path,2026-07-07T02:59:08Z
pr,188903,[HOP] Adding input mutation support for switch (2/2),https://github.com/pytorch/pytorch/pull/188903,ready for maintainer decision,45,"CI lane labels: ciflow/inductor, ciflow/torchtitan; open non-draft PR has no obvious automated blocker",review diff and decide acceptance path,2026-07-06T23:06:37Z
pr,182687,Support generator arg in dirichlet and beta torch.distributions API,https://github.com/pytorch/pytorch/pull/182687,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-07-06T21:59:31Z
pr,188846,"Fix torch.pow docstring: scalar params are Number, not float",https://github.com/pytorch/pytorch/pull/188846,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-07-06T20:41:24Z
pr,138068,Add TORCH_CHECK_INDEX in convert_indices_from_coo_to_csr_cpu,https://github.com/pytorch/pytorch/pull/138068,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-07-06T20:13:54Z
pr,126348,[AOTAutograd] tweak min-cut partitioner to avoid saving softmax output,https://github.com/pytorch/pytorch/pull/126348,ready for maintainer decision,45,CI lane labels: ciflow/inductor; open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-07-06T20:11:52Z
pr,188976,[HOP] Inductor lowering for switch (1/2),https://github.com/pytorch/pytorch/pull/188976,ready for maintainer decision,45,"CI lane labels: ciflow/inductor, ciflow/torchtitan; open non-draft PR has no obvious automated blocker",review diff and decide acceptance path,2026-07-06T19:21:25Z
pr,188947,[Bugfix][DTensor] Fix wrong shard offsets for uneven DTensor shards in TP-FSDP metadata,https://github.com/pytorch/pytorch/pull/188947,ready for maintainer decision,45,"CI lane labels: ciflow/dtensor, ciflow/inductor, ciflow/torchtitan; open non-draft PR has no obvious automated blocker",review diff and decide acceptance path,2026-07-06T19:01:04Z
pr,188885,[DataLoader] Clean up stale Py2 comments and pickle-error warning,https://github.com/pytorch/pytorch/pull/188885,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-07-06T18:55:40Z
pr,188874,Enable XPUGraph native recording mode,https://github.com/pytorch/pytorch/pull/188874,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-07-06T18:47:12Z
pr,188804,[MPS] Add solo large-tensor allocator + torch.mps.set_allocator_settings(),https://github.com/pytorch/pytorch/pull/188804,ready for maintainer decision,45,CI lane labels: ciflow/mps; open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-07-06T18:45:20Z
pr,188944,[NCCL] try to remove the redundant copy in functional collectives,https://github.com/pytorch/pytorch/pull/188944,ready for maintainer decision,45,CI lane labels: ciflow/torchtitan; open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-07-06T16:59:48Z
pr,188761,MSVC compiler flag change for JIT cpp extensions on Windows,https://github.com/pytorch/pytorch/pull/188761,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-07-06T16:39:28Z
pr,188823,[Inductor][Host-side TMA] Cache host-side TMA descriptors across launches,https://github.com/pytorch/pytorch/pull/188823,ready for maintainer decision,45,"CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk; open non-draft PR has no obvious automated blocker",review diff and decide acceptance path,2026-07-06T15:48:40Z
pr,183755,[MPS] Enable bf16/fp16 ConvTranspose3D by removing the gating,https://github.com/pytorch/pytorch/pull/183755,ready for maintainer decision,45,CI lane labels: ciflow/mps; open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-07-06T14:32:58Z
pr,185881,[Test] Refactor test/test_reductions.py,https://github.com/pytorch/pytorch/pull/185881,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-07-06T13:43:34Z
pr,188156,[MPS] reduce: share the two-pass value_reduction kernel for sum and value,https://github.com/pytorch/pytorch/pull/188156,ready for maintainer decision,45,CI lane labels: ciflow/mps; open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-07-06T13:39:46Z
pr,182747,Resolve PYTHON_SIX_SOURCE_DIR from installed `six` to skip pypi download,https://github.com/pytorch/pytorch/pull/182747,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-07-06T09:49:59Z
pr,187317,[pr-review skill] Add device portability checks for shared accelerator code,https://github.com/pytorch/pytorch/pull/187317,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-07-06T09:39:58Z
pr,188977,Fix typos in docstrings and comments across torch modules,https://github.com/pytorch/pytorch/pull/188977,ready for maintainer decision,45,"CI lane labels: ciflow/torchtitan, ciflow/trunk; open non-draft PR has no obvious automated blocker",review diff and decide acceptance path,2026-07-06T07:50:58Z
pr,182582,Add addmm quantization graph passes for PT2 readiness benchmarks (#182582),https://github.com/pytorch/pytorch/pull/182582,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-07-06T01:09:42Z
pr,182159,Fix List isSubtypeExt,https://github.com/pytorch/pytorch/pull/182159,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-07-05T22:49:44Z
pr,179485,Add torchbind custom class inheritance support (#179485),https://github.com/pytorch/pytorch/pull/179485,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-07-05T20:49:08Z
pr,185788,Support 64-bit indexing in channels_last bilinear upsampling,https://github.com/pytorch/pytorch/pull/185788,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-07-05T18:30:10Z
pr,188500,Allow pre-split pipeline microbatches,https://github.com/pytorch/pytorch/pull/188500,ready for maintainer decision,45,"CI lane labels: ciflow/torchtitan, ciflow/trunk; open non-draft PR has no obvious automated blocker",review diff and decide acceptance path,2026-07-05T17:24:13Z
pr,181622,[Inductor][NVGEMM] Refactor rendering,https://github.com/pytorch/pytorch/pull/181622,ready for maintainer decision,45,"CI lane labels: ciflow/inductor, ciflow/torchtitan; open non-draft PR has no obvious automated blocker",review diff and decide acceptance path,2026-07-04T21:41:47Z
pr,181902,[Inductor][NVGEMM] Remove duplicate nvgemm_max_profiling config,https://github.com/pytorch/pytorch/pull/181902,ready for maintainer decision,45,"CI lane labels: ciflow/inductor, ciflow/torchtitan; open non-draft PR has no obvious automated blocker",review diff and decide acceptance path,2026-07-04T21:37:22Z
pr,181903,[Inductor][NVGEMM] Avoid double iteration over kernel cache during choice enumeration,https://github.com/pytorch/pytorch/pull/181903,ready for maintainer decision,45,"CI lane labels: ciflow/inductor, ciflow/torchtitan; open non-draft PR has no obvious automated blocker",review diff and decide acceptance path,2026-07-04T21:37:06Z
pr,181910,[Inductor][NVGEMM] Link tracking issue in supplement configs comment,https://github.com/pytorch/pytorch/pull/181910,ready for maintainer decision,45,"CI lane labels: ciflow/inductor, ciflow/torchtitan; open non-draft PR has no obvious automated blocker",review diff and decide acceptance path,2026-07-04T21:32:46Z
pr,182061,Document low-memory state_dict loading with meta device,https://github.com/pytorch/pytorch/pull/182061,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-07-04T18:50:04Z
pr,182053,Partition load_state_dict child keys once,https://github.com/pytorch/pytorch/pull/182053,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-07-04T18:50:02Z
pr,181862,collect_env: add uv support,https://github.com/pytorch/pytorch/pull/181862,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-07-04T12:16:34Z
pr,175083,[pytree] add another simplified pytree module `torch.pytree`,https://github.com/pytorch/pytorch/pull/175083,ready for maintainer decision,45,CI lane labels: ciflow/trunk; open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-07-04T12:10:37Z
pr,182340,Bump Windows CPU trunk test shards from 4 to 5,https://github.com/pytorch/pytorch/pull/182340,ready for maintainer decision,45,CI lane labels: ciflow/trunk; open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-07-04T01:04:22Z
pr,187556,[MPS] Migrate smooth_l1_loss and huber_loss to native Metal kernels,https://github.com/pytorch/pytorch/pull/187556,ready for maintainer decision,45,CI lane labels: ciflow/mps; open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-07-03T22:25:26Z
pr,182426,Enable FP32 emulation via BF16 in cuSOLVER for qr and lu,https://github.com/pytorch/pytorch/pull/182426,ready for maintainer decision,45,CI lane labels: ciflow/b200; open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-07-03T13:51:18Z
pr,179212,feat(pipelining): add configurable auto-partitioning algorithm for pipeline parallelism,https://github.com/pytorch/pytorch/pull/179212,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-07-03T13:20:31Z
pr,188660,[Inductor] Enable mix-order reduction for wide reductions + tests,https://github.com/pytorch/pytorch/pull/188660,ready for maintainer decision,45,"CI lane labels: ciflow/inductor, ciflow/torchtitan; open non-draft PR has no obvious automated blocker",review diff and decide acceptance path,2026-07-03T11:52:52Z
pr,187557,[MPS] Migrate binary_cross_entropy to native Metal kernels,https://github.com/pytorch/pytorch/pull/187557,ready for maintainer decision,45,CI lane labels: ciflow/mps; open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-07-03T10:54:23Z
pr,188760,Guard user stream compile tests with xfailIfNoAcceleratorTriton,https://github.com/pytorch/pytorch/pull/188760,ready for maintainer decision,45,CI lane labels: ciflow/inductor; open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-07-03T10:11:34Z
pr,181849,Enable FP32 emulation via BF16 in cuSOLVER,https://github.com/pytorch/pytorch/pull/181849,ready for maintainer decision,45,CI lane labels: ciflow/b200; open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-07-03T07:39:16Z
pr,188379,[HOP] None input handling for scan,https://github.com/pytorch/pytorch/pull/188379,ready for maintainer decision,45,"CI lane labels: ciflow/inductor, ciflow/torchtitan; open non-draft PR has no obvious automated blocker",review diff and decide acceptance path,2026-07-03T07:14:35Z
pr,182919,Add forward-only cuDNN FP8 per-tensor attention,https://github.com/pytorch/pytorch/pull/182919,ready for maintainer decision,45,CI lane labels: ciflow/inductor; open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-07-03T01:01:36Z
pr,188126,[inductor] Enable cudagraph partition inside invoke_subgraph subgraphs,https://github.com/pytorch/pytorch/pull/188126,ready for maintainer decision,45,"CI lane labels: ciflow/inductor, ciflow/torchtitan; open non-draft PR has no obvious automated blocker",review diff and decide acceptance path,2026-07-02T22:20:37Z
pr,188613,[MPS] copy for contiguous same-dtype using a compute kernel,https://github.com/pytorch/pytorch/pull/188613,ready for maintainer decision,45,CI lane labels: ciflow/mps; open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-07-02T16:07:21Z
pr,188315,[HOP] Autograd implementation for switch,https://github.com/pytorch/pytorch/pull/188315,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-07-02T10:36:24Z
pr,188348,[HOP] Fixes for zero-length scan dim,https://github.com/pytorch/pytorch/pull/188348,ready for maintainer decision,45,"CI lane labels: ciflow/inductor, ciflow/torchtitan; open non-draft PR has no obvious automated blocker",review diff and decide acceptance path,2026-07-02T07:59:29Z
pr,188555,Fix rocm lazy context initialization in test_set_device_0,https://github.com/pytorch/pytorch/pull/188555,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-07-02T00:57:00Z
pr,188606,Relax complex tensor consistency tolerances for CUDA binary nonlinear ops,https://github.com/pytorch/pytorch/pull/188606,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-07-02T00:56:01Z
pr,183716,Validate dim in cummax/cummin meta for empty tensors,https://github.com/pytorch/pytorch/pull/183716,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-07-01T17:12:31Z
pr,188333,[MPS] prod: migrate to the shared value_reduction<ProdOp> kernel + shape specializations,https://github.com/pytorch/pytorch/pull/188333,ready for maintainer decision,45,CI lane labels: ciflow/mps; open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-07-01T16:44:40Z
pr,151218,Implement MKLGenerator,https://github.com/pytorch/pytorch/pull/151218,ready for maintainer decision,45,CI lane labels: ciflow/trunk; open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-07-01T12:19:18Z
pr,188329,Optimize all-gather copy-in with a flat parameter buffer,https://github.com/pytorch/pytorch/pull/188329,ready for maintainer decision,45,"CI lane labels: ciflow/inductor, ciflow/torchtitan; open non-draft PR has no obvious automated blocker",review diff and decide acceptance path,2026-07-01T09:36:32Z
pr,186603,Exposing MiMalloc options and stats,https://github.com/pytorch/pytorch/pull/186603,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-07-01T08:34:53Z
pr,188626,build-almalinux-images: migrate to OSDC via remote BuildKit,https://github.com/pytorch/pytorch/pull/188626,ready for maintainer decision,45,CI lane labels: ciflow/docker; open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-07-01T03:44:03Z
pr,188625,binary-docker-build: add opt-in remote BuildKit (OSDC) mode,https://github.com/pytorch/pytorch/pull/188625,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-07-01T02:22:23Z
pr,188530,optim: remove redundant Adafactor assert,https://github.com/pytorch/pytorch/pull/188530,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-30T22:49:40Z
pr,188402,Fix matrix_exp crash on non-contiguous (channels-last) inputs,https://github.com/pytorch/pytorch/pull/188402,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-30T18:30:39Z
pr,186617,Added profiler record for THPVariable_dealloc,https://github.com/pytorch/pytorch/pull/186617,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-30T13:54:47Z
pr,188407,[Windows] Add triton_windows commit pin for Windows nightly builds,https://github.com/pytorch/pytorch/pull/188407,ready for maintainer decision,45,"CI lane labels: ciflow/docker, ciflow/inductor; open non-draft PR has no obvious automated blocker",review diff and decide acceptance path,2026-06-29T22:39:43Z
pr,188364,Allow from_blob / TensorMaker to attach BackendMeta at construction,https://github.com/pytorch/pytorch/pull/188364,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-29T21:14:50Z
pr,179020,Update third-party/rocm_composable_kernel to f1746955,https://github.com/pytorch/pytorch/pull/179020,ready for maintainer decision,45,CI lane labels: ciflow/rocm-mi300; open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-29T20:13:31Z
pr,188164,Fix THPDTypeInfo_compare reading off the end of non-finfo/iinfo objects,https://github.com/pytorch/pytorch/pull/188164,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-29T19:12:40Z
pr,188433,linear_cross_entropy: test and document bias with probability targets,https://github.com/pytorch/pytorch/pull/188433,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-29T18:51:42Z
pr,182030,[ROCm] Make AOTriton bundling optional via `BUILD_AOTRITON_INTO_WHEEL` flag,https://github.com/pytorch/pytorch/pull/182030,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-29T17:34:08Z
pr,182040,Download magma & sccache install scripts over https instead of aws s3,https://github.com/pytorch/pytorch/pull/182040,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-29T17:23:04Z
pr,188293,Allow tuple subclasses for run_backward inputs,https://github.com/pytorch/pytorch/pull/188293,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-29T16:27:54Z
pr,181996,[torch.package] Register package modules in sys.modules for Python 3.12 dataclass compat,https://github.com/pytorch/pytorch/pull/181996,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-29T10:04:28Z
pr,187999,Fix uninitialized memory warnings in CPU linear algebra kernels,https://github.com/pytorch/pytorch/pull/187999,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-28T20:29:44Z
pr,188031,Fix 64-bit indexing in spatial softmax backward,https://github.com/pytorch/pytorch/pull/188031,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-28T20:23:09Z
pr,181330,[Profiler] Add a skip_build_tree metadata field to Kineto events to m…,https://github.com/pytorch/pytorch/pull/181330,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-27T22:50:18Z
pr,176584,[RNG] Decouple Philox subsequence from hardware topology for cross-platform consistency,https://github.com/pytorch/pytorch/pull/176584,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-27T17:50:48Z
pr,188191,Add sm_121a (NVIDIA GB10/DGX Spark) to default CUDA arch list,https://github.com/pytorch/pytorch/pull/188191,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-27T17:12:14Z
pr,186881,nn/utils: skip no-op .to(first_device) in _get_total_norm,https://github.com/pytorch/pytorch/pull/186881,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-27T10:32:51Z
pr,188252,Fix ROCm round-robin counter reset in initSingleStream,https://github.com/pytorch/pytorch/pull/188252,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-27T01:02:30Z
pr,181030,[MPS] Migrate SDPA from MPSGraph to Metal,https://github.com/pytorch/pytorch/pull/181030,ready for maintainer decision,45,CI lane labels: ciflow/mps; open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-26T21:53:45Z
pr,176317,[MPS] Leveraging unified memory memcpy instead of blit encoding for tensor device transfers,https://github.com/pytorch/pytorch/pull/176317,ready for maintainer decision,45,CI lane labels: ciflow/mps; open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-26T20:09:48Z
pr,188263,(fix) NNPACK PeachPy air gapped build,https://github.com/pytorch/pytorch/pull/188263,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-26T19:26:34Z
pr,188125,[AI Codemod][PerfAICT-General] perf: Special case non-symbolic set_sizes_contiguous,https://github.com/pytorch/pytorch/pull/188125,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-26T16:45:01Z
pr,181285,"[TEST][ARM] Remove Pointer Authentication Code from address, fall back to `dladdr_lookup` where possible",https://github.com/pytorch/pytorch/pull/181285,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-26T16:17:38Z
pr,188045,[MPS] better memory allocator,https://github.com/pytorch/pytorch/pull/188045,ready for maintainer decision,45,CI lane labels: ciflow/mps; open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-26T12:09:36Z
pr,180758,Remove dead TORCH_HIP_VERSION guards,https://github.com/pytorch/pytorch/pull/180758,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-26T06:09:24Z
pr,186145,"Spelling fix: Change ""structeur"" to ""structure"" in Lumen CLI VLLM test",https://github.com/pytorch/pytorch/pull/186145,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-25T22:07:38Z
pr,178037,[MPS] Raise clear error when MPS is used in forked subprocess,https://github.com/pytorch/pytorch/pull/178037,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-25T16:24:00Z
pr,187813,Generalize memory viz and cycle detection,https://github.com/pytorch/pytorch/pull/187813,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-24T19:32:34Z
pr,188089,[forward AD] Remove obsolete __debug__ guard when loading JVP decompositions,https://github.com/pytorch/pytorch/pull/188089,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-24T18:51:53Z
pr,186624,Fix CLANGTIDY errors in c10/cuda,https://github.com/pytorch/pytorch/pull/186624,ready for maintainer decision,45,CI lane labels: ciflow/trunk; open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-24T18:46:28Z
pr,187713,Fix B008 flake8-bugbear violations (function calls in default arguments),https://github.com/pytorch/pytorch/pull/187713,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-24T18:32:01Z
pr,187687,fix: aten embed bwd thread sync before proceeding to next batch,https://github.com/pytorch/pytorch/pull/187687,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-24T03:00:15Z
pr,180584,Test,https://github.com/pytorch/pytorch/pull/180584,ready for maintainer decision,45,"CI lane labels: ciflow/b200, ciflow/h100, ciflow/inductor, ciflow/rocm-mi300; open non-draft PR has no obvious automated blocker",review diff and decide acceptance path,2026-06-23T22:02:02Z
pr,178069,[docs] Add examples to reduce_scatter documentation,https://github.com/pytorch/pytorch/pull/178069,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-23T15:36:42Z
pr,178068,[docs] Add comprehensive introduction to PyTorch Elastic documentation,https://github.com/pytorch/pytorch/pull/178068,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-23T15:36:40Z
pr,187881,Fix margin docstring rendering in HingeEmbeddingLoss,https://github.com/pytorch/pytorch/pull/187881,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-23T14:26:10Z
pr,187816,Clean up mentions of deprecated nvidia-docker in README files,https://github.com/pytorch/pytorch/pull/187816,ready for maintainer decision,45,CI lane labels: ciflow/docker; open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-23T14:23:49Z
pr,187847,fix(nn.utils.prune): don't crash adding a method to an empty PruningContainer,https://github.com/pytorch/pytorch/pull/187847,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-23T14:22:49Z
pr,187796,Add native MPS SVD path,https://github.com/pytorch/pytorch/pull/187796,ready for maintainer decision,45,CI lane labels: ciflow/mps; open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-23T14:21:30Z
pr,187800,Verify Magma archive before extraction,https://github.com/pytorch/pytorch/pull/187800,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-23T14:21:22Z
pr,187742,Fix calculate_out_dimension using stride[0] for the width dimension,https://github.com/pytorch/pytorch/pull/187742,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-23T14:21:17Z
pr,187799,Fix ROCm libdrm readlink buffer bounds,https://github.com/pytorch/pytorch/pull/187799,ready for maintainer decision,45,CI lane labels: ciflow/docker; open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-23T14:20:51Z
pr,187769,[docs] Clarify Softplus overflow behavior for reduced-precision dtypes,https://github.com/pytorch/pytorch/pull/187769,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-23T14:00:44Z
pr,187717,[CUDA] solve_triangular: fix silently wrong batched results and slowness in large-RHS workaround,https://github.com/pytorch/pytorch/pull/187717,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-23T13:58:43Z
pr,187921,Support freeing blocks with stream uses during graph capture,https://github.com/pytorch/pytorch/pull/187921,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-23T13:35:14Z
pr,187694,Unpack Python autograd saved tensors in reverse order,https://github.com/pytorch/pytorch/pull/187694,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-23T13:12:02Z
pr,184398,Enable 3.15 in CI,https://github.com/pytorch/pytorch/pull/184398,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-22T23:27:15Z
pr,185446,Adding a test case using fx.wrap to test whether TLX user-defined kernel is visible in the inductor output (#185301),https://github.com/pytorch/pytorch/pull/185446,ready for maintainer decision,45,"CI lane labels: ciflow/inductor, ciflow/torchtitan; open non-draft PR has no obvious automated blocker",review diff and decide acceptance path,2026-06-22T20:20:57Z
pr,178373,docs(nn): clarify return type for MaxPool1d/2d/3d when return_indices=True,https://github.com/pytorch/pytorch/pull/178373,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-22T15:24:46Z
pr,178568,Add support for activation checkpointing with nested data structures,https://github.com/pytorch/pytorch/pull/178568,ready for maintainer decision,45,CI lane labels: ciflow/torchtitan; open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-22T11:27:58Z
pr,175512,[kineto] Generate unique group_trace_id instead of reusing trace_id (#175512),https://github.com/pytorch/pytorch/pull/175512,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-22T08:34:02Z
pr,162454,[xpu][feature] Integrate OneDNN SDPA training forward/backward into XPU OVERRIDEABLE Backend,https://github.com/pytorch/pytorch/pull/162454,ready for maintainer decision,45,CI lane labels: ciflow/binaries_wheel; open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-22T06:57:10Z
pr,180171,Fix empty changed-files handling in Filepath heuristic,https://github.com/pytorch/pytorch/pull/180171,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-21T21:58:34Z
pr,175425,Improve error message for CPU-CUDA copy during graph capture,https://github.com/pytorch/pytorch/pull/175425,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-21T15:14:30Z
pr,180788,"Use C++20 concepts/requires in Tensor, TensorBase, and IValue",https://github.com/pytorch/pytorch/pull/180788,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-21T06:27:52Z
pr,180988,[fbandroid][buckconfig removal] removal of last files referencing tools/build_defs/android/build_mode_defs.bzl,https://github.com/pytorch/pytorch/pull/180988,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-20T17:57:38Z
pr,187333,[CUDA] linalg: overlap looped cuSOLVER LU (getrf) across auxiliary streams,https://github.com/pytorch/pytorch/pull/187333,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-20T15:58:08Z
pr,183811,[docs] Document 10 public APIs in torch.nn.parallel submodules,https://github.com/pytorch/pytorch/pull/183811,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-20T06:31:10Z
pr,180686,test: relax error tolerances for complex64 tests with tan and tanh,https://github.com/pytorch/pytorch/pull/180686,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-19T20:55:26Z
pr,180187,add OneCycleLR degenerate phase cases,https://github.com/pytorch/pytorch/pull/180187,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-19T17:18:03Z
pr,175706,add C++20 bit operations and contains to C10,https://github.com/pytorch/pytorch/pull/175706,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-18T15:46:18Z
pr,185930,[Elastic][Distributed] Flush dirty rendezvous state immediately,https://github.com/pytorch/pytorch/pull/185930,ready for maintainer decision,45,CI lane labels: ciflow/torchtitan; open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-18T14:33:39Z
pr,178088,Use more structure binding,https://github.com/pytorch/pytorch/pull/178088,ready for maintainer decision,45,CI lane labels: ciflow/inductor; open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-18T08:15:24Z
pr,180527,Add thread_time_ns and idle_pct to dynamo_timed chromium events (#180527),https://github.com/pytorch/pytorch/pull/180527,ready for maintainer decision,45,"CI lane labels: ciflow/inductor, ciflow/torchtitan; open non-draft PR has no obvious automated blocker",review diff and decide acceptance path,2026-06-17T21:25:15Z
pr,187566,[docs] Improve autograd anomaly mode documentation,https://github.com/pytorch/pytorch/pull/187566,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-17T17:17:48Z
pr,179747,Add rotary positional embeddings support to nn.MultiheadAttention,https://github.com/pytorch/pytorch/pull/179747,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-17T03:48:39Z
pr,187410,[distributed] Use node.target instead of node.name in apply_manual_re…,https://github.com/pytorch/pytorch/pull/187410,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-16T21:31:07Z
pr,187245,update backend parametrization for varlen attn,https://github.com/pytorch/pytorch/pull/187245,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-16T21:26:56Z
pr,179700,[torch][inductor] Add RAIIAtenRecordFunctionHandle profiling to remaining uncovered kernel paths (#179700),https://github.com/pytorch/pytorch/pull/179700,ready for maintainer decision,45,"CI lane labels: ciflow/inductor, ciflow/torchtitan; open non-draft PR has no obvious automated blocker",review diff and decide acceptance path,2026-06-16T20:31:36Z
pr,185790,Fix torch.nn.functional.gelu returning NaN for infinite inputs,https://github.com/pytorch/pytorch/pull/185790,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-16T09:23:20Z
pr,179698,[torch][inductor] Fix numel variable scoping with enable_kernel_profile (#179698),https://github.com/pytorch/pytorch/pull/179698,ready for maintainer decision,45,"CI lane labels: ciflow/inductor, ciflow/torchtitan; open non-draft PR has no obvious automated blocker",review diff and decide acceptance path,2026-06-16T03:48:36Z
pr,187192,Protect GraphPassRegistry map with shared_mutex against concurrent access (#187192),https://github.com/pytorch/pytorch/pull/187192,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-15T16:49:09Z
pr,175476,[inductor] Hoist output buffer allocations into prior CUDAGraph partition,https://github.com/pytorch/pytorch/pull/175476,ready for maintainer decision,45,"CI lane labels: ciflow/h100-symm-mem, ciflow/inductor, ciflow/torchtitan; open non-draft PR has no obvious automated blocker",review diff and decide acceptance path,2026-06-15T14:56:19Z
pr,173182,Updated foreach_norm to accept dim and keepdim arguments.,https://github.com/pytorch/pytorch/pull/173182,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-15T12:32:54Z
pr,177147,[FSDP2] Remove redundant event wait for all_gather_stream,https://github.com/pytorch/pytorch/pull/177147,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-15T03:33:33Z
pr,176936,Add driver api guard to cuda utils,https://github.com/pytorch/pytorch/pull/176936,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-14T20:03:45Z
pr,174605,make Store timeout_ atomic for thread safety.,https://github.com/pytorch/pytorch/pull/174605,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-14T20:03:40Z
pr,179594,[CUDA] Use #pragma unroll in Normaization.cuh,https://github.com/pytorch/pytorch/pull/179594,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-14T16:59:33Z
pr,179399,Improve GELU documentation formula,https://github.com/pytorch/pytorch/pull/179399,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-14T16:59:33Z
pr,179000,Add type conversion test for IValue: implements TODO(gmagogsfm),https://github.com/pytorch/pytorch/pull/179000,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-14T16:59:26Z
pr,179372,[Inductor] Fix view_as_complex stride requirement in Inductor backward by forcing contiguous layout,https://github.com/pytorch/pytorch/pull/179372,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-14T16:03:23Z
pr,187126,[c10] Remove deprecated CUDAAllocatorConfig forwarders,https://github.com/pytorch/pytorch/pull/187126,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-13T02:18:03Z
pr,187097,[BE] Fix B007 unused loop variables,https://github.com/pytorch/pytorch/pull/187097,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-12T17:12:17Z
pr,187146,test all named dims test with dynamic sepcs using automatic conversion,https://github.com/pytorch/pytorch/pull/187146,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-12T07:28:34Z
pr,187135,[xpu] Remove deprecated at::xpu CachingHostAllocator forwarders,https://github.com/pytorch/pytorch/pull/187135,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-12T05:51:32Z
pr,186046,[MPS] Remove dead xfail/tolerance keys from MPS test infra,https://github.com/pytorch/pytorch/pull/186046,ready for maintainer decision,45,CI lane labels: ciflow/mps; open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-12T05:51:16Z
pr,187006,Avoid duplicate sm_103 family-specific CUDA codegen,https://github.com/pytorch/pytorch/pull/187006,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-11T17:40:48Z
pr,181292,[inductor] Autotune cache stability and perf improvements (#181292),https://github.com/pytorch/pytorch/pull/181292,ready for maintainer decision,45,"CI lane labels: ciflow/inductor, ciflow/torchtitan; open non-draft PR has no obvious automated blocker",review diff and decide acceptance path,2026-06-11T16:46:20Z
pr,178924,[inductor] Fix FP8 dtype detection in _get_estimated_runtime,https://github.com/pytorch/pytorch/pull/178924,ready for maintainer decision,45,"CI lane labels: ciflow/inductor, ciflow/torchtitan; open non-draft PR has no obvious automated blocker",review diff and decide acceptance path,2026-06-11T16:34:57Z
pr,186924,Also search $MKLROOT/lib in Linux for mkl wheel and install_mkl.sh,https://github.com/pytorch/pytorch/pull/186924,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-11T10:08:12Z
pr,186232,Add opt-in cross-stream segment reclaim to CUDA allocator,https://github.com/pytorch/pytorch/pull/186232,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-11T08:32:39Z
pr,184190,[fx]: Add SymNode helper to get raw expr,https://github.com/pytorch/pytorch/pull/184190,ready for maintainer decision,45,CI lane labels: ciflow/inductor; open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-11T08:23:21Z
pr,183343,dynamo: trace torch.cuda.Event.record / Event.wait under fx proxy mode,https://github.com/pytorch/pytorch/pull/183343,ready for maintainer decision,45,"CI lane labels: ciflow/inductor, ciflow/torchtitan; open non-draft PR has no obvious automated blocker",review diff and decide acceptance path,2026-06-10T19:15:57Z
pr,186887,Update test_mm_sparse_first_NT/NN input generation,https://github.com/pytorch/pytorch/pull/186887,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-10T18:05:18Z
pr,178545,[MPS] Add native ops fast path for SDPA on long sequences (~2x speedup),https://github.com/pytorch/pytorch/pull/178545,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-10T11:45:28Z
pr,183852,ci: declare workflow-level `contents: read` on 8 workflows,https://github.com/pytorch/pytorch/pull/183852,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-10T08:09:37Z
pr,178960,profiler: fix USE_KINETO=OFF build failure (unconditional ActivityType.h include),https://github.com/pytorch/pytorch/pull/178960,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-10T05:03:49Z
pr,185507,Pre-compute igates for XPU LSTM to reduce per-step kernel launches,https://github.com/pytorch/pytorch/pull/185507,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-09T14:02:36Z
pr,184585,[MPS] Wire SDPBackend dispatch for scaled_dot_product_attention,https://github.com/pytorch/pytorch/pull/184585,ready for maintainer decision,45,"CI lane labels: ciflow/inductor, ciflow/mps; open non-draft PR has no obvious automated blocker",review diff and decide acceptance path,2026-06-09T01:51:15Z
pr,184559,[Test]Refactor test/test_reductions.py to be device generic,https://github.com/pytorch/pytorch/pull/184559,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-08T17:31:19Z
pr,186568,Add expected-failure test for ConstantLR factor=0,https://github.com/pytorch/pytorch/pull/186568,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-08T14:08:52Z
pr,186307,Add contains() to ska::flat_hash_map/set,https://github.com/pytorch/pytorch/pull/186307,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-08T14:06:30Z
pr,186565,Support parametrized modules in functional_call,https://github.com/pytorch/pytorch/pull/186565,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-08T13:51:20Z
pr,186559,Remove deprecated pytree_to_str and str_to_pytree functions,https://github.com/pytorch/pytorch/pull/186559,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-08T13:41:45Z
pr,186473,[accelerator] Support multiple private backends,https://github.com/pytorch/pytorch/pull/186473,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-08T13:36:52Z
pr,186239,[ROCm][CI] Add linux.rocm.mi350.docker-cache to docker-cache-rocm,https://github.com/pytorch/pytorch/pull/186239,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-08T13:34:18Z
pr,185901,[Test] Refactor test_schema_check.py to be device-agnostic,https://github.com/pytorch/pytorch/pull/185901,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-05T12:51:21Z
pr,185648,[Test] Make sigmoid_add extension test device-agnostic in test_cpp_extensions_aot.py,https://github.com/pytorch/pytorch/pull/185648,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-05T07:47:34Z
pr,185184,[Test][distributed] Refactor some distributed test helpers to be device-agnostic,https://github.com/pytorch/pytorch/pull/185184,ready for maintainer decision,45,CI lane labels: ciflow/trunk; open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-04T06:55:13Z
pr,185883,Simplify ostream formatting with fmt::join and fmt::print,https://github.com/pytorch/pytorch/pull/185883,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-03T22:45:38Z
pr,185993,Fix SM121 compute capability corruption in CMake GPU arch detection,https://github.com/pytorch/pytorch/pull/185993,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-03T22:38:53Z
pr,185906,add type hint to seed argument in torch.random.manual_seed,https://github.com/pytorch/pytorch/pull/185906,ready for maintainer decision,45,CI lane labels: ciflow/torchtitan; open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-03T22:18:32Z
pr,186045,Merge identical elif branches and inline single-use locals in torchgen,https://github.com/pytorch/pytorch/pull/186045,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-03T22:01:20Z
pr,186110,Remove test_transformers from windows blocklist,https://github.com/pytorch/pytorch/pull/186110,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-03T22:01:09Z
pr,184548,[Test]Refactor test/test_sort_and_select.py to be device generic,https://github.com/pytorch/pytorch/pull/184548,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-03T09:35:44Z
pr,183855,[c10d] Surface operational metadata in WorkNCCL operator<<,https://github.com/pytorch/pytorch/pull/183855,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-03T04:34:32Z
pr,178590,Correct type hints for ConvNd subclasses,https://github.com/pytorch/pytorch/pull/178590,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-02T09:19:15Z
pr,185758,Tidy MPSGuardImpl: name capability bitmask and fix clang-tidy nits,https://github.com/pytorch/pytorch/pull/185758,ready for maintainer decision,45,CI lane labels: ciflow/mps; open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-02T03:10:57Z
pr,171313,Support torch.accelerator.Graph on CUDA,https://github.com/pytorch/pytorch/pull/171313,ready for maintainer decision,45,CI lane labels: ciflow/trunk; open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-02T02:02:32Z
pr,181833,Add boolean evaluation for LazyVal wrapping str,https://github.com/pytorch/pytorch/pull/181833,ready for maintainer decision,45,CI lane labels: ciflow/trunk; open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-01T17:59:36Z
pr,180954,Add value property to LazyVal,https://github.com/pytorch/pytorch/pull/180954,ready for maintainer decision,45,CI lane labels: ciflow/trunk; open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-01T17:59:35Z
pr,185743,Mark CUDA kernel helpers static,https://github.com/pytorch/pytorch/pull/185743,ready for maintainer decision,45,CI lane labels: ciflow/trunk; open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-01T16:44:09Z
pr,184859,Deprecate c10::ssize as alias for std::ssize,https://github.com/pytorch/pytorch/pull/184859,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-05-30T03:13:49Z
pr,185125,Remove unused caffe2/utils/string_utils,https://github.com/pytorch/pytorch/pull/185125,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-05-30T01:29:45Z
pr,184142,Fix LPPool signed fractional inputs,https://github.com/pytorch/pytorch/pull/184142,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-05-29T17:19:31Z
pr,182289,[MPS] Pinned memory - Add pinned memory allocator,https://github.com/pytorch/pytorch/pull/182289,ready for maintainer decision,45,CI lane labels: ciflow/mps; open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-05-29T12:36:07Z
pr,183870,[stats] Handle alternate test-reports path in backfill_test_jsons_while_running,https://github.com/pytorch/pytorch/pull/183870,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-05-29T05:58:43Z
pr,184426,Support getting CUDA patch version,https://github.com/pytorch/pytorch/pull/184426,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-05-28T15:24:58Z
pr,152159,Add dynamo config to HOP-ify context managers,https://github.com/pytorch/pytorch/pull/152159,ready for maintainer decision,45,CI lane labels: ciflow/inductor; open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-05-28T12:46:52Z
pr,183738,Multi-warp TMA scatter_add via PerWarpTmaPipeline,https://github.com/pytorch/pytorch/pull/183738,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-05-27T12:57:44Z
pr,183501,[PyTorch][Pinned Memory] Fix stuck event blocking in CachingHostAllocator,https://github.com/pytorch/pytorch/pull/183501,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-05-27T06:42:14Z
pr,183739,TMA scatter_add: 2D-tile multi-row path for small-N shapes,https://github.com/pytorch/pytorch/pull/183739,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-05-26T16:55:12Z
pr,185137,[fx][_library] Honor wrapped op's schema.is_mutable in is_impure's auto_functionalized HOP branch (#185137),https://github.com/pytorch/pytorch/pull/185137,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-05-26T15:09:04Z
pr,184866,[distributed] Handle empty inputs and kwargs in _to_kwargs,https://github.com/pytorch/pytorch/pull/184866,ready for maintainer decision,45,CI lane labels: ciflow/torchtitan; open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-05-26T13:37:56Z
pr,183752,[BE]: Revert back to upstream uv action to solve rate-limiting,https://github.com/pytorch/pytorch/pull/183752,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-05-26T04:42:21Z
pr,178082,Fix ninja workdir collisions in BuildExtension,https://github.com/pytorch/pytorch/pull/178082,waiting on contributor,45,review decision requests changes; PR closes or references open issue numbers; PR has not updated for at least 45 days,wait for contributor update,2026-05-21T03:55:36Z
pr,179719,Skip div-to-mul-reciprocal when division_rounding is enabled,https://github.com/pytorch/pytorch/pull/179719,waiting on contributor,45,"review decision requests changes; PR closes or references open issue numbers; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk",wait for contributor update,2026-05-15T13:39:14Z
pr,172003,Add __getitems__ to TensorDataset for faster DataLoader iteration,https://github.com/pytorch/pytorch/pull/172003,waiting on contributor,45,review decision requests changes; PR closes or references open issue numbers; PR has not updated for at least 45 days,wait for contributor update,2026-05-14T21:34:37Z
pr,173072,s390x: rework ZVECTOR data into 16-byte-sized vectors,https://github.com/pytorch/pytorch/pull/173072,waiting on contributor,45,"review decision requests changes; PR closes or references open issue numbers; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/s390, ciflow/torchtitan",wait for contributor update,2026-05-11T23:32:03Z
pr,181096,inductor/flex_attention: fix fake block-mask generator for autotuner,https://github.com/pytorch/pytorch/pull/181096,waiting on contributor,45,"review decision requests changes; PR closes or references open issue numbers; PR has not updated for at least 45 days; CI lane labels: ciflow/h100-symm-mem, ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-05-11T21:25:32Z
pr,172991,Fix torch.compile for jagged nested tensors created inside compiled functions,https://github.com/pytorch/pytorch/pull/172991,waiting on contributor,45,review decision requests changes; PR closes or references open issue numbers; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor,wait for contributor update,2026-04-16T13:38:42Z
issue,150199,`asarray`: device does not propagate from input to output after `set_default_device`,https://github.com/pytorch/pytorch/issues/150199,has linked PR,44,"open PR references this issue; component/oncall labels: module: python array api, module: python frontend",review linked PR state before touching issue,2026-07-08T21:44:27Z
issue,141287,MPS operator coverage tracking issue (2.6+ version),https://github.com/pytorch/pytorch/issues/141287,has linked PR,44,open PR references this issue; component/oncall labels: module: mps,review linked PR state before touching issue,2026-07-08T20:26:39Z
issue,189271,[inductor] KeyError in Scheduler.compute_ancestors: self-edge from realizing a reshaped in-place-mutated buffer,https://github.com/pytorch/pytorch/issues/189271,has linked PR,44,"open PR references this issue; component/oncall labels: module: custom-operators, module: dynamic shapes, module: inductor, module: pt2-dispatcher",review linked PR state before touching issue,2026-07-08T15:25:09Z
issue,129131,"Torch compile initialises CUDA context, even compiling CPU functions",https://github.com/pytorch/pytorch/issues/129131,has linked PR,44,"open PR references this issue; component/oncall labels: module: dynamo, module: pt2-dispatcher, oncall: pt2",review linked PR state before touching issue,2026-07-08T13:42:20Z
issue,187811,DISABLED test_combo_kernel_dynamic_scale_rblock (__main__.ComboKernelTests),https://github.com/pytorch/pytorch/issues/187811,has linked PR,44,open PR references this issue; component/oncall labels: module: xpu,review linked PR state before touching issue,2026-07-08T02:07:23Z
issue,187810,DISABLED test_combo_kernel_dynamic_scale_rblock (__main__.ComboKernelTestsPerSubkernelBlocks),https://github.com/pytorch/pytorch/issues/187810,has linked PR,44,open PR references this issue; component/oncall labels: module: xpu,review linked PR state before touching issue,2026-07-08T02:07:08Z
issue,186796,weakrefs don't when tensor swap is used,https://github.com/pytorch/pytorch/issues/186796,has linked PR,44,"open PR references this issue; component/oncall labels: module: dynamo, module: guards, oncall: pt2",review linked PR state before touching issue,2026-07-07T20:58:51Z
issue,167725,`torch._inductor.utils.fresh_cache` fails due to persistant process pool,https://github.com/pytorch/pytorch/issues/167725,has linked PR,44,"open PR references this issue; component/oncall labels: module: inductor, oncall: pt2",review linked PR state before touching issue,2026-07-07T18:19:40Z
issue,189157,GHA stats artifact downloader does not skip mismatched workflow run attempts,https://github.com/pytorch/pytorch/issues/189157,has linked PR,44,open PR references this issue; component/oncall labels: module: ci,review linked PR state before touching issue,2026-07-07T14:58:03Z
issue,189133,DISABLED test_vmap (__main__.FuncTorchHigherOrderOpTestsWithCompiledAutograd),https://github.com/pytorch/pytorch/issues/189133,has linked PR,44,"open PR references this issue; component/oncall labels: module: flaky-tests, module: inductor, oncall: pt2",review linked PR state before touching issue,2026-07-07T13:25:26Z
issue,186867,DISABLED test_group_norm_backward_large_input_float16_trigger_32bit_overflow_True_mps_float16 (__main__.TestConsistencyMPS),https://github.com/pytorch/pytorch/issues/186867,has linked PR,44,"open PR references this issue; component/oncall labels: module: flaky-tests, module: macos, module: mps",review linked PR state before touching issue,2026-07-07T07:52:44Z
issue,187093,[inductor] baddbmm with broadcast bias: autotune never considers the bmm + pointwise-bias decomposition (1.3-1.45x left on the table at MoE expert-GEMM shapes),https://github.com/pytorch/pytorch/issues/187093,has linked PR,44,"open PR references this issue; component/oncall labels: module: inductor, module: performance, oncall: pt2",review linked PR state before touching issue,2026-07-07T05:17:10Z
issue,189118,tools/testing: upload test-jsons artifact from the generated JSON zip,https://github.com/pytorch/pytorch/issues/189118,has linked PR,44,"open PR references this issue; component/oncall labels: module: ci, module: tests",review linked PR state before touching issue,2026-07-07T02:18:00Z
issue,186605,DISABLED test_group_norm_backward_large_input_float16_trigger_32bit_overflow_False_mps_float16 (__main__.TestConsistencyMPS),https://github.com/pytorch/pytorch/issues/186605,has linked PR,44,"open PR references this issue; component/oncall labels: module: flaky-tests, module: macos, module: mps",review linked PR state before touching issue,2026-07-07T01:15:52Z
issue,188773,DISABLED test_cow_inputs_accepted_by_override (__main__.TestBmmOuterProduct),https://github.com/pytorch/pytorch/issues/188773,has linked PR,44,open PR references this issue; component/oncall labels: module: xpu,review linked PR state before touching issue,2026-07-07T01:04:53Z
issue,189106,`get_signature_for_torch_op` reports incorrect `return_annotation` for `tuple`s,https://github.com/pytorch/pytorch/issues/189106,has linked PR,44,open PR references this issue; component/oncall labels: module: fx,review linked PR state before touching issue,2026-07-06T22:42:29Z
issue,188680,[test] OpInfo gap: allclose-based compile-vs-eager checks are blind to signed-zero and signed-infinity divergences,https://github.com/pytorch/pytorch/issues/188680,has linked PR,44,"open PR references this issue; component/oncall labels: module: tests, oncall: pt2",review linked PR state before touching issue,2026-07-06T16:41:39Z
issue,160744,MPS copy_ produces incorrect output when input has more than one element,https://github.com/pytorch/pytorch/issues/160744,has linked PR,44,"open PR references this issue; component/oncall labels: module: correctness (silent), module: edge cases, module: mps",review linked PR state before touching issue,2026-07-06T16:27:22Z
issue,188422,Some test_sdpa_can_compile_dynamic in transformers fail (with torch.compile(dynamic=True)) with torch 2.13 RC but pass with torch 2.12,https://github.com/pytorch/pytorch/issues/188422,has linked PR,44,"open PR references this issue; component/oncall labels: module: dynamic shapes, oncall: pt2",review linked PR state before touching issue,2026-07-06T12:40:03Z
issue,186348,torch.mm slower than pointwise mul+sum when K and N are small (under 8),https://github.com/pytorch/pytorch/issues/186348,has linked PR,44,"open PR references this issue; component/oncall labels: module: decompositions, module: inductor, module: linear algebra, module: performance",review linked PR state before touching issue,2026-07-06T05:51:22Z
issue,188718,torch.compile (Inductor) fails with KeyError: 'complex64' on view_as_complex of a valid non-contiguous CUDA tensor,https://github.com/pytorch/pytorch/issues/188718,has linked PR,44,"open PR references this issue; component/oncall labels: module: complex, module: cuda, module: inductor, oncall: pt2",review linked PR state before touching issue,2026-07-06T05:07:14Z
issue,188989,[FSDP] summon_full_params(offload_to_cpu=True) fails with CPU-offloaded FSDP model,https://github.com/pytorch/pytorch/issues/188989,has linked PR,44,"open PR references this issue; component/oncall labels: module: fsdp, oncall: distributed, oncall: distributed parallelisms",review linked PR state before touching issue,2026-07-06T02:43:21Z
issue,161132,The Inductor test suite test_torchinductor_opinfo was incorrectly disabled on all devices except CUDA.,https://github.com/pytorch/pytorch/issues/161132,has linked PR,44,"open PR references this issue; component/oncall labels: module: inductor, module: tests, oncall: cpu inductor, oncall: pt2",review linked PR state before touching issue,2026-07-05T20:41:16Z
issue,182277,[CUDA][CUDACachingAllocator] size overflow in caching_allocator_alloc raises SystemError for size >= 2**63,https://github.com/pytorch/pytorch/issues/182277,has linked PR,44,"open PR references this issue; component/oncall labels: module: cuda, module: edge cases, module: error checking",review linked PR state before touching issue,2026-07-05T08:31:09Z
issue,173232,error C2872: 'std': ambiguous symbol when compiling `csrc/dynamo/compiled_autograd.h`,https://github.com/pytorch/pytorch/issues/173232,has linked PR,44,"open PR references this issue; component/oncall labels: module: compiled autograd, module: windows, oncall: pt2",review linked PR state before touching issue,2026-07-05T03:28:25Z
issue,184575,"`F.normalize` returns ~1e12 gradient at zero-vector input instead of NaN, while the forward pass already returns NaN",https://github.com/pytorch/pytorch/issues/184575,has linked PR,44,"open PR references this issue; component/oncall labels: module: autograd, module: edge cases, module: norms and normalization",review linked PR state before touching issue,2026-07-04T08:42:22Z
issue,187759,"torch.linalg.svdvals swallows NaN (returns finite values), disagreeing with torch.linalg.svd; matrix_rank then reports full rank for a NaN matrix",https://github.com/pytorch/pytorch/issues/187759,has linked PR,44,"open PR references this issue; component/oncall labels: module: correctness (silent), module: edge cases, module: linear algebra, module: nans and infs",review linked PR state before touching issue,2026-07-04T08:42:06Z
issue,188891,"F.linear with a 1-D weight and bias is inconsistent: eager fails for 2-D input, torch.func transforms (grad/jvp/vmap) fail for all input ranks",https://github.com/pytorch/pytorch/issues/188891,has linked PR,44,"open PR references this issue; component/oncall labels: module: functorch, module: nn",review linked PR state before touching issue,2026-07-03T19:01:49Z
issue,188723,ShapeEnv lacks a metadata-tree transfer API that preserves foreign unbacked SymInt relationships,https://github.com/pytorch/pytorch/issues/188723,has linked PR,44,"open PR references this issue; component/oncall labels: module: dynamic shapes, module: faketensor, module: fx, module: pt2-dispatcher",review linked PR state before touching issue,2026-07-03T16:45:20Z
issue,188882,Empty autograd section in backend YAML is rejected,https://github.com/pytorch/pytorch/issues/188882,has linked PR,44,"open PR references this issue; component/oncall labels: module: backend, module: codegen",review linked PR state before touching issue,2026-07-03T11:20:21Z
issue,188536,Graph-break reconstruction loses identity of graph-created torch.Event subclasses; record() and wait() operate on different event objects,https://github.com/pytorch/pytorch/issues/188536,has linked PR,44,"open PR references this issue; component/oncall labels: module: dynamo, oncall: pt2",review linked PR state before touching issue,2026-07-03T09:04:51Z
issue,167223,[feature request] Re-enable `__torch_function__` dispatch when invoking autograd hooks for Tensor subclasses,https://github.com/pytorch/pytorch/issues/167223,has linked PR,44,"open PR references this issue; component/oncall labels: module: __torch_function__, module: autograd, module: docs",review linked PR state before touching issue,2026-07-03T07:36:06Z
issue,188866,[dynamo] TVM backend fails to import with all pip-installable TVM versions (relay removed in TVM 0.20),https://github.com/pytorch/pytorch/issues/188866,has linked PR,44,"open PR references this issue; component/oncall labels: module: backend, module: dependency bug, module: dynamo, module: third_party",review linked PR state before touching issue,2026-07-03T03:44:18Z
issue,171905,Use typing_extensions.TypeAliasType for better reexport of `__module__`,https://github.com/pytorch/pytorch/issues/171905,has linked PR,44,"open PR references this issue; component/oncall labels: module: lint, module: typing",review linked PR state before touching issue,2026-07-02T18:55:05Z
issue,188796,[opaque object] Callable value Opaque Object on a tensor subclass is not accessible for method calls in torch.compile and leads to graph-break,https://github.com/pytorch/pytorch/issues/188796,has linked PR,44,"open PR references this issue; component/oncall labels: module: dynamo, module: graph breaks, module: library, module: pt2-dispatcher",review linked PR state before touching issue,2026-07-02T12:39:18Z
issue,188757,Windows CI: test_overrides.py class TestTorchFunctionRedispatchOpsCUDA fails consistently (same as #182869 on Linux),https://github.com/pytorch/pytorch/issues/188757,has linked PR,44,"open PR references this issue; component/oncall labels: module: __torch_function__, module: ci, module: cuda, module: tests",review linked PR state before touching issue,2026-07-02T10:52:33Z
issue,188790,Inconsistent device check in test_staging.py: guard uses torch.cuda.is_available() while block body uses torch.accelerator.is_available(),https://github.com/pytorch/pytorch/issues/188790,has linked PR,44,"open PR references this issue; component/oncall labels: oncall: distributed, oncall: distributed checkpointing",review linked PR state before touching issue,2026-07-02T09:20:09Z
issue,187806,"F.gelu(approximate='none'): 1+erf catastrophic cancellation — up to 57% relative error in fp32 around x≈−5, zeros the negative tail",https://github.com/pytorch/pytorch/issues/187806,has linked PR,44,"open PR references this issue; component/oncall labels: module: correctness (silent), module: nn, module: numerical-stability",review linked PR state before touching issue,2026-07-02T03:17:16Z
issue,188538,torch.fx _format_target emits invalid or semantically incorrect Python for valid module names,https://github.com/pytorch/pytorch/issues/188538,has linked PR,44,"open PR references this issue; component/oncall labels: module: fx, oncall: pt2",review linked PR state before touching issue,2026-07-02T00:57:59Z
issue,188711,DISABLED test_main_loop_scaling_shape0_use_fast_accum_True_scaling_block_sizes2_cuda (__main__.TestFP8LoweringCUDA),https://github.com/pytorch/pytorch/issues/188711,has linked PR,44,open PR references this issue; component/oncall labels: module: cuda,review linked PR state before touching issue,2026-07-02T00:34:17Z
issue,188728,[MPS] F.linear does not respect the PYTORCH_MPS_PREFER_METAL flag,https://github.com/pytorch/pytorch/issues/188728,has linked PR,44,open PR references this issue; component/oncall labels: module: mps,review linked PR state before touching issue,2026-07-01T20:25:44Z
issue,184002,[DTensor][Compile] Compile fails when DTensor function returns an aliased output,https://github.com/pytorch/pytorch/issues/184002,has linked PR,44,"open PR references this issue; component/oncall labels: module: aotdispatch, module: dtensor, module: pt2-dispatcher, oncall: distributed",review linked PR state before touching issue,2026-07-01T19:31:16Z
issue,188635,torch.compile misses fusion for producer -> view/transpose/contiguous layout clone in attention-style paths,https://github.com/pytorch/pytorch/issues/188635,has linked PR,44,"open PR references this issue; component/oncall labels: module: inductor, module: performance, oncall: pt2",review linked PR state before touching issue,2026-07-01T02:52:01Z
issue,188477,[CI][B200] smoke_b200 test_nv_universal_gemm fails: cutlass.cute.arch has no attribute 'ProxyKind' (unpinned cutlass_api vs pinned cutlass-dsl),https://github.com/pytorch/pytorch/issues/188477,has linked PR,44,open PR references this issue; component/oncall labels: module: ci,review linked PR state before touching issue,2026-07-01T01:55:25Z
issue,188544,[opaque object] Value Opaque Object errors with instance static method.,https://github.com/pytorch/pytorch/issues/188544,has linked PR,44,"open PR references this issue; component/oncall labels: module: dynamo, module: guards, oncall: pt2",review linked PR state before touching issue,2026-06-30T12:00:58Z
issue,188398,[Docs] torch.utils.mobile_optimizer link broken in 2.7 API documentation,https://github.com/pytorch/pytorch/issues/188398,has linked PR,44,open PR references this issue; component/oncall labels: module: docs,review linked PR state before touching issue,2026-06-30T11:43:01Z
issue,188325,torch._dynamo: EventVariable.python_type() returns torch.Event instead of type(self.value),https://github.com/pytorch/pytorch/issues/188325,has linked PR,44,"open PR references this issue; component/oncall labels: module: cuda, module: dynamo, oncall: pt2",review linked PR state before touching issue,2026-06-30T11:30:36Z
issue,187107,"[MPS] linalg.lu_factor not implemented, causing RunTimeErrors on torch.linalg.det and torch.linalg.solve",https://github.com/pytorch/pytorch/issues/187107,has linked PR,44,"open PR references this issue; component/oncall labels: module: complex, module: linear algebra, module: mps",review linked PR state before touching issue,2026-06-29T19:49:18Z
issue,188405,"torch.compile reconstructs a graph-created torch.Stream/torch.Event subclass as the base class, losing the subclass type across a graph break",https://github.com/pytorch/pytorch/issues/188405,has linked PR,44,"open PR references this issue; component/oncall labels: module: dynamo, oncall: pt2",review linked PR state before touching issue,2026-06-29T19:22:50Z
issue,188226,[inductor] `ValueRangeError: Invalid ranges [0:-61]` during index propagation,https://github.com/pytorch/pytorch/issues/188226,has linked PR,44,"open PR references this issue; component/oncall labels: module: inductor, module: pooling, oncall: pt2",review linked PR state before touching issue,2026-06-29T18:20:26Z
issue,188166,load_state_dict(assign=True) mutates the caller's state_dict._metadata,https://github.com/pytorch/pytorch/issues/188166,has linked PR,44,open PR references this issue; component/oncall labels: module: nn,review linked PR state before touching issue,2026-06-29T18:19:45Z
issue,188213,[inductor] `LoweringException: AssertionError` in `scatter_reduce_` lowering,https://github.com/pytorch/pytorch/issues/188213,has linked PR,44,"open PR references this issue; component/oncall labels: module: inductor, module: scatter & gather ops, oncall: pt2",review linked PR state before touching issue,2026-06-29T17:38:24Z
issue,188048,"[Inductor][CUDA] torch.div(rounding_mode=""floor"") mismatches eager for infinite inputs",https://github.com/pytorch/pytorch/issues/188048,has linked PR,44,"open PR references this issue; component/oncall labels: module: correctness (silent), module: cuda, module: edge cases, module: inductor",review linked PR state before touching issue,2026-06-29T17:10:11Z
issue,188128,`asarray` does not work for boolean dtype when using buffer protocol,https://github.com/pytorch/pytorch/issues/188128,has linked PR,44,"open PR references this issue; component/oncall labels: module: correctness (silent), module: python array api, module: python frontend, module: tensor creation",review linked PR state before touching issue,2026-06-29T17:07:00Z
issue,188423,test_sdpa_can_compile_dynamic in transformers fails (with torch.compile(dynamic=True)) with torch 2.13 RC but pass with torch 2.12,https://github.com/pytorch/pytorch/issues/188423,has linked PR,44,"open PR references this issue; component/oncall labels: module: dynamic shapes, module: inductor, module: sdpa, oncall: pt2",review linked PR state before touching issue,2026-06-29T14:48:32Z
issue,188131,DISABLED test_fallback_kernel_with_symexpr_output_xpu (__main__.AOTInductorTestABICompatibleGpu),https://github.com/pytorch/pytorch/issues/188131,has linked PR,44,"open PR references this issue; component/oncall labels: module: inductor, module: xpu, oncall: pt2",review linked PR state before touching issue,2026-06-29T08:54:01Z
issue,188390,"[DTensor][Compile][AOTAutograd cache] DTensor stable hashing generates identical AOTAutograd hash keys across ranks, causing false cache hits.",https://github.com/pytorch/pytorch/issues/188390,has linked PR,44,"open PR references this issue; component/oncall labels: module: aotdispatch, module: dtensor, module: pt2-dispatcher, oncall: distributed",review linked PR state before touching issue,2026-06-29T06:19:29Z
issue,186107,torch.scan gaps,https://github.com/pytorch/pytorch/issues/186107,has linked PR,44,"open PR references this issue; component/oncall labels: module: higher order operators, module: pt2-dispatcher, oncall: pt2",review linked PR state before touching issue,2026-06-28T22:53:48Z
issue,188370,[Meta][PrivateUse1]: Data-dependent ATen ops need eager shape-oracle support for PU1 backends.,https://github.com/pytorch/pytorch/issues/188370,has linked PR,44,"open PR references this issue; component/oncall labels: module: backend, module: meta tensors, module: privateuse1",review linked PR state before touching issue,2026-06-28T17:14:35Z
issue,188338,[ROCm]: CUDAGraph safe Convolutional Kernel e.g. `F.conv_transpose1d`,https://github.com/pytorch/pytorch/issues/188338,has linked PR,44,"open PR references this issue; component/oncall labels: module: convolution, module: cuda graphs, module: rocm",review linked PR state before touching issue,2026-06-27T15:36:44Z
issue,188245,Broken tutorial URL,https://github.com/pytorch/pytorch/issues/188245,has linked PR,44,open PR references this issue; component/oncall labels: module: docs,review linked PR state before touching issue,2026-06-26T15:58:35Z
issue,187953,"torch.compile: out= variants of CompositeImplicitAutograd linalg ops (linalg.norm, cholesky, ...) recompile on every input shape",https://github.com/pytorch/pytorch/issues/187953,has linked PR,44,"open PR references this issue; component/oncall labels: module: compile-time, module: dynamic shapes, module: faketensor, module: guards",review linked PR state before touching issue,2026-06-26T14:00:33Z
issue,144580,`torch._foreach_mul` does not support autograd,https://github.com/pytorch/pytorch/issues/144580,has linked PR,44,"open PR references this issue; component/oncall labels: module: autograd, module: mta",review linked PR state before touching issue,2026-06-26T06:37:37Z
issue,188230,[inductor] `index_propagation` `ValueError: The argument 'False' is not comparable`,https://github.com/pytorch/pytorch/issues/188230,has linked PR,44,"open PR references this issue; component/oncall labels: module: inductor, oncall: pt2",review linked PR state before touching issue,2026-06-26T01:25:42Z
issue,188225,[inductor] `index_propagation` `ValueError: The argument 'nan' is not comparable`,https://github.com/pytorch/pytorch/issues/188225,has linked PR,44,"open PR references this issue; component/oncall labels: module: inductor, oncall: pt2",review linked PR state before touching issue,2026-06-26T01:21:30Z
issue,188217,[inductor] `TypeError: reduce() of empty iterable with no initial value` during lowering,https://github.com/pytorch/pytorch/issues/188217,has linked PR,44,"open PR references this issue; component/oncall labels: module: inductor, oncall: pt2",review linked PR state before touching issue,2026-06-26T00:51:46Z
issue,188182,torch.fft has no sharding strategy,https://github.com/pytorch/pytorch/issues/188182,has linked PR,44,"open PR references this issue; component/oncall labels: module: dtensor, oncall: distributed, oncall: distributed parallelisms",review linked PR state before touching issue,2026-06-25T23:20:26Z
issue,188150,torch.compile reduce-overhead: CUDAGraphs recompiles on every batch with dynamic padding (HF training loop),https://github.com/pytorch/pytorch/issues/188150,has linked PR,44,"open PR references this issue; component/oncall labels: module: compile ux, module: cuda graphs, module: dynamic shapes, module: performance",review linked PR state before touching issue,2026-06-25T13:00:29Z
issue,145208,Some FlexAttention learned bias bugs/limitations,https://github.com/pytorch/pytorch/issues/145208,has linked PR,44,"open PR references this issue; component/oncall labels: module: flex attention, module: higher order operators, module: pt2-dispatcher, oncall: pt2",review linked PR state before touching issue,2026-06-25T09:01:08Z
issue,188088,torch.clamp rejects mixed Tensor and Number min/max bounds although Tensor+Tensor and Number+Number both work,https://github.com/pytorch/pytorch/issues/188088,has linked PR,44,open PR references this issue; component/oncall labels: module: python frontend,review linked PR state before touching issue,2026-06-25T07:11:36Z
issue,186544,[Windows] Debug build fails: LNK1104 cannot open file 'python310_d.lib',https://github.com/pytorch/pytorch/issues/186544,has linked PR,44,"open PR references this issue; component/oncall labels: module: build, module: debug-build, module: pybind, module: windows",review linked PR state before touching issue,2026-06-25T02:29:40Z
issue,137096,Expressions involving Min/Max don't simplify with value ranges information,https://github.com/pytorch/pytorch/issues/137096,has linked PR,44,"open PR references this issue; component/oncall labels: module: dynamic shapes, oncall: pt2",review linked PR state before touching issue,2026-06-24T21:53:29Z
issue,188052,Allow PrivateUse1 backends to opt out of CompositeExplicitAutograd decomposition,https://github.com/pytorch/pytorch/issues/188052,has linked PR,44,"open PR references this issue; component/oncall labels: module: dispatch, module: privateuse1",review linked PR state before touching issue,2026-06-24T18:10:51Z
issue,185814,"[Inductor] Mix-Order Reduction: XBLOCK Derivation Ignores N in RMSNorm Backward, Causing Bandwidth Cliff",https://github.com/pytorch/pytorch/issues/185814,has linked PR,44,"open PR references this issue; component/oncall labels: module: inductor, oncall: pt2",review linked PR state before touching issue,2026-06-23T17:31:02Z
issue,185589,"[inductor] Severe numerical inconsistency: diag_embed + erf + normalize(p=-1) under torch.compile(dynamic=True) — max diff up to 565,248",https://github.com/pytorch/pytorch/issues/185589,has linked PR,44,"open PR references this issue; component/oncall labels: module: correctness (silent), module: dynamic shapes, module: inductor, module: numerical-stability",review linked PR state before touching issue,2026-06-23T17:30:41Z
issue,185581,`torch.compile` fails at FakeTensor validation for ConvTranspose2d,https://github.com/pytorch/pytorch/issues/185581,has linked PR,44,open PR references this issue,review linked PR state before touching issue,2026-06-23T17:30:15Z
issue,185481,[inductor] argmax over comparison mask is accepted and used for indexing while eager raises,https://github.com/pytorch/pytorch/issues/185481,has linked PR,44,"open PR references this issue; component/oncall labels: module: error checking, module: inductor, oncall: pt2",review linked PR state before touching issue,2026-06-23T17:29:04Z
issue,185382,`torch.compile`: Inductor silently produces wrong results for `addcdiv_`/`addcmul_` when `.item()` graph break occurs in a loop,https://github.com/pytorch/pytorch/issues/185382,has linked PR,44,"open PR references this issue; component/oncall labels: module: correctness (silent), module: inductor, oncall: pt2",review linked PR state before touching issue,2026-06-23T17:28:05Z
issue,187945,"`replace_pattern` and `replace_pattern_with_filters` do not propagate `meta[""val""]` to replacement nodes, causing downstream filters to silently reject matches",https://github.com/pytorch/pytorch/issues/187945,has linked PR,44,"open PR references this issue; component/oncall labels: module: fx, oncall: pt2",review linked PR state before touching issue,2026-06-23T14:00:38Z
issue,116465,torch.cartesian_prod inconsitent return shape for one input tensor,https://github.com/pytorch/pytorch/issues/116465,has linked PR,44,open PR references this issue; component/oncall labels: module: numpy,review linked PR state before touching issue,2026-06-23T09:36:28Z
issue,185248,`torch.compile` silently accepts negative standard deviations in `torch.normal` that eager mode rejects,https://github.com/pytorch/pytorch/issues/185248,has linked PR,44,"open PR references this issue; component/oncall labels: module: correctness (silent), module: dynamo, module: random, oncall: pt2",review linked PR state before touching issue,2026-06-23T08:11:05Z
issue,185246,torch.compile silently accepts invalid Bernoulli probabilities that eager mode rejects,https://github.com/pytorch/pytorch/issues/185246,has linked PR,44,"open PR references this issue; component/oncall labels: module: correctness (silent), module: dynamo, module: error checking, module: random",review linked PR state before touching issue,2026-06-23T08:07:37Z
issue,187332,[PT2] Output mismatch between eager mode and torch.compile for torch.special.i0 with inf inputs on CUDA,https://github.com/pytorch/pytorch/issues/187332,has linked PR,44,"open PR references this issue; component/oncall labels: module: correctness (silent), module: cuda, module: decompositions, module: edge cases",review linked PR state before touching issue,2026-06-23T08:03:36Z
issue,185696,[Inductor] torch.atan2 yields incorrect result for x=0.0 with float16 on CUDA (loses signed zero),https://github.com/pytorch/pytorch/issues/185696,has linked PR,44,"open PR references this issue; component/oncall labels: module: correctness (silent), module: cuda, module: edge cases, module: half",review linked PR state before touching issue,2026-06-23T08:03:35Z
issue,146938,"ModuleDict has an incomplete (or wrong) typing, producing type errors when accessed with attributes",https://github.com/pytorch/pytorch/issues/146938,has linked PR,44,"open PR references this issue; component/oncall labels: module: nn, module: typing",review linked PR state before touching issue,2026-06-20T02:07:56Z
issue,187178,torch.nn.functional.triplet_margin_loss returns NaN for p < 1 (p-norm overflow),https://github.com/pytorch/pytorch/issues/187178,has linked PR,44,"open PR references this issue; component/oncall labels: module: edge cases, module: nn",review linked PR state before touching issue,2026-06-19T06:35:32Z
issue,187177,"torch.nn.LocalResponseNorm returns +/-Inf when k=0 and alpha=0 (zero denominator, no validation)",https://github.com/pytorch/pytorch/issues/187177,has linked PR,44,"open PR references this issue; component/oncall labels: module: error checking, module: nn, module: norms and normalization",review linked PR state before touching issue,2026-06-19T06:35:23Z
issue,180807,[inductor][static launcher] Missing runtime dtype validation for pointer-typed Triton args can silently produce incorrect results,https://github.com/pytorch/pytorch/issues/180807,has linked PR,44,"open PR references this issue; component/oncall labels: module: correctness (silent), module: inductor, oncall: pt2",review linked PR state before touching issue,2026-06-18T14:57:42Z
issue,141595,FX graph cache does not do provenance tracking for inductor guards,https://github.com/pytorch/pytorch/issues/141595,has linked PR,44,"open PR references this issue; component/oncall labels: module: inductor, oncall: pt2",review linked PR state before touching issue,2026-06-18T04:32:22Z
issue,136264,Compatibility with numpy 2 __array__ interface,https://github.com/pytorch/pytorch/issues/136264,has linked PR,44,open PR references this issue; component/oncall labels: module: numpy,review linked PR state before touching issue,2026-06-17T19:56:18Z
issue,187593,[Dynamo] Avoid guarding on Unused variables,https://github.com/pytorch/pytorch/issues/187593,has linked PR,44,"open PR references this issue; component/oncall labels: module: dynamo, module: guards, oncall: pt2",review linked PR state before touching issue,2026-06-17T18:55:37Z
issue,64947,Quantile is limited to 16 million elements and have poor performance.,https://github.com/pytorch/pytorch/issues/64947,has linked PR,44,"open PR references this issue; component/oncall labels: module: numpy, module: performance, module: sorting and selection",review linked PR state before touching issue,2026-06-17T13:36:29Z
issue,187296,torch.heaviside returns 0 for NaN input instead of propagating NaN,https://github.com/pytorch/pytorch/issues/187296,has linked PR,44,"open PR references this issue; component/oncall labels: module: correctness (silent), module: edge cases, module: nans and infs, module: numpy",review linked PR state before touching issue,2026-06-17T11:54:29Z
issue,187547,Dynamo graph-breaks on raw unbacked SymInt BlockMask lengths in FlexAttention,https://github.com/pytorch/pytorch/issues/187547,has linked PR,44,"open PR references this issue; component/oncall labels: module: dynamic shapes, module: dynamo, oncall: pt2",review linked PR state before touching issue,2026-06-17T04:14:29Z
issue,187336,"[Inductor] torch.compile loses signbit in torch.special.log_ndtr, causing discrepancies in sign-sensitive ops (e.g., copysign, atan2)",https://github.com/pytorch/pytorch/issues/187336,has linked PR,44,"open PR references this issue; component/oncall labels: module: correctness (silent), module: edge cases, module: inductor, module: special",review linked PR state before touching issue,2026-06-17T02:34:08Z
issue,187451,RFC: Polyhedral Optimization Pass for PyTorch Inductor,https://github.com/pytorch/pytorch/issues/187451,has linked PR,44,"open PR references this issue; component/oncall labels: module: inductor, module: performance, oncall: pt2",review linked PR state before touching issue,2026-06-16T17:19:59Z
issue,187338,torch.gcd returns an incorrect negative result when one input contains the minimum value of torch.int8 (-128),https://github.com/pytorch/pytorch/issues/187338,has linked PR,44,"open PR references this issue; component/oncall labels: module: correctness (silent), module: edge cases",review linked PR state before touching issue,2026-06-16T04:39:07Z
issue,187389,Missing reference CPU implementation of sparse `triangular_solve` (Mac OS / ARM and other non-MKL supported targets),https://github.com/pytorch/pytorch/issues/187389,has linked PR,44,"open PR references this issue; component/oncall labels: module: arm, module: cpu, module: linear algebra, module: macos",review linked PR state before touching issue,2026-06-15T22:22:34Z
issue,187295,torch.sign returns 0 for NaN inputs instead of propagating NaN,https://github.com/pytorch/pytorch/issues/187295,has linked PR,44,"open PR references this issue; component/oncall labels: module: correctness (silent), module: edge cases, module: nans and infs, module: testing",review linked PR state before touching issue,2026-06-15T17:23:10Z
issue,186824,torch.distributions.Cauchy.icdf(0) returns large finite value with wrong sign in float32; Gumbel.icdf(0/1) also finite instead of ±inf,https://github.com/pytorch/pytorch/issues/186824,has linked PR,44,"open PR references this issue; component/oncall labels: module: correctness (silent), module: distributions, module: edge cases",review linked PR state before touching issue,2026-06-15T17:02:27Z
issue,187299,torch.nn.functional.unfold — RuntimeError instead of ValueError for zero stride,https://github.com/pytorch/pytorch/issues/187299,has linked PR,44,"open PR references this issue; component/oncall labels: module: edge cases, module: error checking, module: nn",review linked PR state before touching issue,2026-06-15T07:00:17Z
issue,187233,torch.nn.functional.max_unpool3d — internal MPS RuntimeError for complex64 input,https://github.com/pytorch/pytorch/issues/187233,has linked PR,44,"open PR references this issue; component/oncall labels: module: complex, module: error checking, module: mps, module: pooling",review linked PR state before touching issue,2026-06-15T05:14:04Z
issue,187298,convert_sync_batchnorm silently drops the activation / extra behavior of `_BatchNorm` subclasses (e.g. timm `BatchNormAct2d`) - please emit a warning,https://github.com/pytorch/pytorch/issues/187298,has linked PR,44,"open PR references this issue; component/oncall labels: module: ddp, oncall: distributed, oncall: distributed parallelisms",review linked PR state before touching issue,2026-06-15T03:07:13Z
issue,187294,torch.special.zeta returns NaN for q=+Inf instead of 0,https://github.com/pytorch/pytorch/issues/187294,has linked PR,44,"open PR references this issue; component/oncall labels: module: edge cases, module: nans and infs, module: special",review linked PR state before touching issue,2026-06-15T02:43:51Z
issue,176298,"[CPU] Binary ops (add, sub, mul) not implemented for uint16/uint32/uint64",https://github.com/pytorch/pytorch/issues/176298,has linked PR,44,"open PR references this issue; component/oncall labels: module: cpu, module: unsigned int",review linked PR state before touching issue,2026-06-14T20:24:59Z
issue,170367,DTensor redistributes on foreach_X operators are inefficient,https://github.com/pytorch/pytorch/issues/170367,has linked PR,44,"open PR references this issue; component/oncall labels: module: dtensor, module: performance, oncall: distributed, oncall: distributed parallelisms",review linked PR state before touching issue,2026-06-13T16:37:25Z
issue,170648,fsdp2 support grad_dtype,https://github.com/pytorch/pytorch/issues/170648,has linked PR,44,"open PR references this issue; component/oncall labels: module: fsdp, oncall: distributed, oncall: distributed parallelisms",review linked PR state before touching issue,2026-06-12T17:09:39Z
issue,187170,[inductor] test_comprehensive_addbmm_cpu_float16 fails due to overly strict float32 reference comparison,https://github.com/pytorch/pytorch/issues/187170,has linked PR,44,"open PR references this issue; component/oncall labels: module: half, module: inductor, module: tests, oncall: pt2",review linked PR state before touching issue,2026-06-12T11:36:45Z
issue,187155,[FSDP] Cache post-forward DeviceMesh to deduplicate NCCL communicators,https://github.com/pytorch/pytorch/issues/187155,has linked PR,44,"open PR references this issue; component/oncall labels: module: fsdp, module: nccl, oncall: distributed, oncall: distributed parallelisms",review linked PR state before touching issue,2026-06-12T07:43:22Z
issue,187008,"nn.Module.to() fails on a tensor-subclass parameter held by an EMA / external weakref (forced swap_tensors path, no compile)",https://github.com/pytorch/pytorch/issues/187008,has linked PR,44,open PR references this issue; component/oncall labels: module: nn,review linked PR state before touching issue,2026-06-11T22:15:48Z
issue,187089,Support BFloat16 in ATen CPU tensor creation (tensor_cpu),https://github.com/pytorch/pytorch/issues/187089,has linked PR,44,"open PR references this issue; component/oncall labels: module: bfloat16, module: cpu, module: tensor creation",review linked PR state before touching issue,2026-06-11T22:05:58Z
issue,187041,[opaque object] Bug with storing in global dict and consuming in TensorSubclass,https://github.com/pytorch/pytorch/issues/187041,has linked PR,44,"open PR references this issue; component/oncall labels: module: dynamo, module: library, module: pt2-dispatcher, oncall: pt2",review linked PR state before touching issue,2026-06-11T13:22:00Z
issue,73009,Improve loading for nn.modules.lazy.LazyModuleMixin,https://github.com/pytorch/pytorch/issues/73009,has linked PR,44,open PR references this issue; component/oncall labels: module: nn,review linked PR state before touching issue,2026-06-11T10:02:00Z
issue,186148,"`torch.matmul` misses `mm` fold for `[S, 1, H] @ [H, K]` when the size-1 batch dim has a non-contiguous stride",https://github.com/pytorch/pytorch/issues/186148,has linked PR,44,"open PR references this issue; component/oncall labels: module: linear algebra, module: performance",review linked PR state before touching issue,2026-06-11T08:43:58Z
issue,185396,[Inductor][CUDA] Internal KeyError: 'complex64' during Triton signature generation for graph with complex intermediates,https://github.com/pytorch/pytorch/issues/185396,has linked PR,44,"open PR references this issue; component/oncall labels: module: compile ux, module: complex, module: cuda, module: inductor",review linked PR state before touching issue,2026-06-10T21:30:53Z
issue,153485,Double backward error in Pytorch PP,https://github.com/pytorch/pytorch/issues/153485,has linked PR,44,"open PR references this issue; component/oncall labels: module: pipelining, oncall: distributed, oncall: distributed parallelisms",review linked PR state before touching issue,2026-06-10T17:13:54Z
issue,184195,aoti_torch_cuda__scaled_dot_product_fused_attention_overrideable not implemented,https://github.com/pytorch/pytorch/issues/184195,has linked PR,44,"open PR references this issue; component/oncall labels: module: aotinductor, module: privateuse1, module: sdpa, oncall: export",review linked PR state before touching issue,2026-06-10T17:02:07Z
issue,186875,[Inductor] fmod/remainder with Python float scalar diverge from eager for bf16/fp16 on CPU (same class as #185517),https://github.com/pytorch/pytorch/issues/186875,has linked PR,44,"open PR references this issue; component/oncall labels: oncall: cpu inductor, oncall: pt2",review linked PR state before touching issue,2026-06-10T14:51:50Z
issue,186799,"`torch.compile` silently produces wrong gradient for `atan2(x, y).amin(dim)` on float32",https://github.com/pytorch/pytorch/issues/186799,has linked PR,44,"open PR references this issue; component/oncall labels: module: autograd, oncall: cpu inductor, oncall: pt2",review linked PR state before touching issue,2026-06-10T14:14:32Z
issue,186901,Inconsistent behavior of torch.sign on NaN input values,https://github.com/pytorch/pytorch/issues/186901,has linked PR,44,"open PR references this issue; component/oncall labels: module: edge cases, module: nans and infs",review linked PR state before touching issue,2026-06-10T12:21:47Z
issue,146306,torch.compile failed on single-node multigpu setting.,https://github.com/pytorch/pytorch/issues/146306,has linked PR,44,"open PR references this issue; component/oncall labels: module: inductor, oncall: pt2",review linked PR state before touching issue,2026-06-09T17:12:59Z
issue,147075,[inductor] Performance Degradation and Hang in `torch.diff`,https://github.com/pytorch/pytorch/issues/147075,has linked PR,44,"open PR references this issue; component/oncall labels: module: inductor, module: performance, oncall: pt2",review linked PR state before touching issue,2026-06-09T17:12:58Z
issue,151027,Add size/strides/alignment assertions for cpp_wrapper,https://github.com/pytorch/pytorch/issues/151027,has linked PR,44,"open PR references this issue; component/oncall labels: module: aotinductor, module: inductor, oncall: export, oncall: pt2",review linked PR state before touching issue,2026-06-09T17:12:56Z
issue,154301,cudagraph copies in 'max-autotune' can add overhead on small model,https://github.com/pytorch/pytorch/issues/154301,has linked PR,44,"open PR references this issue; component/oncall labels: module: inductor, oncall: pt2",review linked PR state before touching issue,2026-06-09T17:12:55Z
issue,183259,InductorError on in-place multidimensional dynamic slicing with Tensor-derived slice bounds,https://github.com/pytorch/pytorch/issues/183259,has linked PR,44,"open PR references this issue; component/oncall labels: module: dynamic shapes, module: inductor, oncall: pt2",review linked PR state before touching issue,2026-06-09T17:12:54Z
issue,123339,"Inconsistency: BCELoss asserts target in [0, 1], BCEWithLogitsLoss does not",https://github.com/pytorch/pytorch/issues/123339,has linked PR,44,open PR references this issue; component/oncall labels: module: loss,review linked PR state before touching issue,2026-06-09T09:31:35Z
issue,185497,[pt2] RuntimeError: variable modified by inplace operation during backward in compiled mode (succeeds in eager),https://github.com/pytorch/pytorch/issues/185497,has linked PR,44,"open PR references this issue; component/oncall labels: module: aotdispatch, module: pt2-dispatcher, oncall: pt2",review linked PR state before touching issue,2026-06-08T17:11:15Z
issue,186556,torch.ldexp returns incorrect gradient w.r.t. mantissa for negative exponents,https://github.com/pytorch/pytorch/issues/186556,has linked PR,44,"open PR references this issue; component/oncall labels: module: autograd, module: correctness (silent), module: derivatives, module: windows",review linked PR state before touching issue,2026-06-08T17:07:37Z
issue,186176,"AOTInductor serializes entire underlying storage for each constant tensor, causing severe bloat for shared-storage weights",https://github.com/pytorch/pytorch/issues/186176,has linked PR,44,"open PR references this issue; component/oncall labels: module: aotinductor, module: serialization, oncall: export, oncall: pt2",review linked PR state before touching issue,2026-06-08T16:11:57Z
issue,186572,[Inductor] Checkpointed reductions can choose different configs between forward and recompute,https://github.com/pytorch/pytorch/issues/186572,has linked PR,44,"open PR references this issue; component/oncall labels: module: activation checkpointing, module: correctness (silent), module: inductor, module: reductions",review linked PR state before touching issue,2026-06-08T15:00:27Z
issue,185509,"[inductor] Silent incorrect gradients with interpolate, conv2d, max_pool2d and torch.where",https://github.com/pytorch/pytorch/issues/185509,has linked PR,44,"open PR references this issue; component/oncall labels: module: correctness (silent), module: inductor, oncall: pt2",review linked PR state before touching issue,2026-06-08T14:38:35Z
issue,185510,[Dynamo] TypeError: cannot determine truth value of Relational during stride sorting with dynamic=True,https://github.com/pytorch/pytorch/issues/185510,has linked PR,44,"open PR references this issue; component/oncall labels: module: dynamic shapes, module: inductor, oncall: pt2",review linked PR state before touching issue,2026-06-08T14:29:06Z
issue,185647,Unexpected storage from MSE loss output,https://github.com/pytorch/pytorch/issues/185647,has linked PR,44,"open PR references this issue; component/oncall labels: module: loss, module: memory usage",review linked PR state before touching issue,2026-06-08T13:31:54Z
issue,89459,Higher order derivatives of sinc explode,https://github.com/pytorch/pytorch/issues/89459,has linked PR,44,open PR references this issue; component/oncall labels: module: derivatives,review linked PR state before touching issue,2026-06-07T01:04:24Z
issue,178096,torch.compile(dynamic=True) on CUDA: large eager vs compiled mismatch for BatchNorm2d + Conv2d,https://github.com/pytorch/pytorch/issues/178096,has linked PR,44,"open PR references this issue; component/oncall labels: module: inductor, module: pt2 accuracy, oncall: pt2",review linked PR state before touching issue,2026-06-06T08:44:11Z
issue,171249,[Bug] softplus returns inf for large beta while other frameworks return finite value,https://github.com/pytorch/pytorch/issues/171249,has linked PR,44,"open PR references this issue; component/oncall labels: module: edge cases, module: nans and infs, module: nn",review linked PR state before touching issue,2026-06-06T04:14:15Z
issue,160057,Error raised when running torch.compile under FakeTensorMode context,https://github.com/pytorch/pytorch/issues/160057,has linked PR,44,open PR references this issue; component/oncall labels: oncall: pt2,review linked PR state before touching issue,2026-06-05T18:20:44Z
issue,140497,C++ standard library functions not found when compiling c10d debug handlers,https://github.com/pytorch/pytorch/issues/140497,has linked PR,44,"open PR references this issue; component/oncall labels: oncall: distributed, oncall: distributed infra",review linked PR state before touching issue,2026-06-05T17:01:35Z
issue,186374,[CUDA] `TestCudaAllocator.test_garbage_collect_expandable` failing on Orin,https://github.com/pytorch/pytorch/issues/186374,has linked PR,44,"open PR references this issue; component/oncall labels: module: cuda, module: cudacachingallocator, module: jetson, module: tests",review linked PR state before touching issue,2026-06-05T14:35:57Z
issue,186028,"[dynamic] torch.linalg.cond(p=""fro"") fails with symbolic sizes while dynamic=False succeeds",https://github.com/pytorch/pytorch/issues/186028,has linked PR,44,"open PR references this issue; component/oncall labels: module: dynamic shapes, module: faketensor, module: linear algebra, module: pt2-dispatcher",review linked PR state before touching issue,2026-06-05T12:17:46Z
issue,186354,Dynamo graph break in native bmm_outer_product eager router due to _is_cow_tensor / call_boxed,https://github.com/pytorch/pytorch/issues/186354,has linked PR,44,"open PR references this issue; component/oncall labels: module: dynamo, module: graph breaks, oncall: pt2",review linked PR state before touching issue,2026-06-05T10:01:51Z
issue,186140,fix qualname for OpOverload/OpOverloadPacket,https://github.com/pytorch/pytorch/issues/186140,has linked PR,44,"open PR references this issue; component/oncall labels: module: custom-operators, module: pt2-dispatcher, oncall: pt2",review linked PR state before touching issue,2026-06-04T18:50:45Z
issue,185888,[Inductor] as_strided_ metadata mutation is not reflected on original tensor,https://github.com/pytorch/pytorch/issues/185888,has linked PR,44,"open PR references this issue; component/oncall labels: module: dynamo, module: meta tensors, oncall: pt2",review linked PR state before touching issue,2026-06-04T17:01:40Z
issue,185975,torch.accelerator.default_stream API is missing,https://github.com/pytorch/pytorch/issues/185975,has linked PR,44,"open PR references this issue; component/oncall labels: module: accelerator, module: privateuse1",review linked PR state before touching issue,2026-06-04T16:18:26Z
issue,185885,[Inductor] torch.index_add accepts negative index while eager raises out-of-bounds error,https://github.com/pytorch/pytorch/issues/185885,has linked PR,44,"open PR references this issue; component/oncall labels: module: inductor, oncall: pt2",review linked PR state before touching issue,2026-06-04T15:03:14Z
issue,186136,The added test TestCompileWorker.test_shutdown_terminates_sidecar_worker_pool (and the corresponding version in TestCompileWorkerWithTimer) are flaky internally due to hitting timeout.,https://github.com/pytorch/pytorch/issues/186136,has linked PR,44,"open PR references this issue; component/oncall labels: module: inductor, module: multiprocessing, module: tests, oncall: pt2",review linked PR state before touching issue,2026-06-04T04:28:27Z
issue,184783,`set_multithreading_enabled` still spawns autograd threads,https://github.com/pytorch/pytorch/issues/184783,has linked PR,44,open PR references this issue; component/oncall labels: module: autograd,review linked PR state before touching issue,2026-06-03T19:37:56Z
issue,185480,[Inductor][CUDA][bf16] `F.celu` returns `+0.0` where eager returns a negative bf16 subnormal,https://github.com/pytorch/pytorch/issues/185480,has linked PR,44,"open PR references this issue; component/oncall labels: module: bfloat16, module: correctness (silent), module: cuda, module: edge cases",review linked PR state before touching issue,2026-06-02T21:01:32Z
issue,185470,[Inductor] `F.threshold` bf16 miscompilation: scalar threshold not cast to tensor dtype,https://github.com/pytorch/pytorch/issues/185470,has linked PR,44,"open PR references this issue; component/oncall labels: module: bfloat16, module: correctness (silent), module: decompositions, oncall: pt2",review linked PR state before touching issue,2026-06-02T21:01:14Z
issue,185955,Integrate with ONNX 1.22.0 release branch,https://github.com/pytorch/pytorch/issues/185955,has linked PR,44,open PR references this issue; component/oncall labels: module: onnx,review linked PR state before touching issue,2026-06-02T20:20:11Z
issue,185464,torch.compile produces wrong results for F.multilabel_margin_loss when target contains -1,https://github.com/pytorch/pytorch/issues/185464,has linked PR,44,"open PR references this issue; component/oncall labels: module: correctness (silent), module: decompositions, module: loss, oncall: pt2",review linked PR state before touching issue,2026-06-02T18:07:50Z
issue,185185,torch.compile(fullgraph=True) fails with PendingUnbackedSymbolNotFound when tensor-derived scalar values are used as slice bounds,https://github.com/pytorch/pytorch/issues/185185,has linked PR,44,"open PR references this issue; component/oncall labels: module: dynamic shapes, oncall: pt2",review linked PR state before touching issue,2026-06-02T17:25:52Z
issue,185331,"[pipelining] Pipelines reuse recv buffer tensors and directly pass them to the stage, causing backward hooks on stage boundary activations to persist between steps",https://github.com/pytorch/pytorch/issues/185331,has linked PR,44,"open PR references this issue; component/oncall labels: module: autograd, module: pipelining, oncall: distributed, oncall: distributed parallelisms",review linked PR state before touching issue,2026-06-02T17:23:06Z
issue,185780,[pipelining] Add downstream tests to test pipelining integration with popular HF transformers architectures,https://github.com/pytorch/pytorch/issues/185780,has linked PR,44,"open PR references this issue; component/oncall labels: module: pipelining, oncall: distributed, oncall: distributed parallelisms",review linked PR state before touching issue,2026-06-01T19:46:25Z
issue,184031,torch.nn.functional.gumbel_softmax returns NaN when tau=0 (division by zero),https://github.com/pytorch/pytorch/issues/184031,has linked PR,44,"open PR references this issue; component/oncall labels: module: correctness (silent), module: edge cases, module: nn",review linked PR state before touching issue,2026-06-01T17:08:53Z
issue,185006,ConstantLR accepts factor=0 and crashes with ZeroDivisionError after total_iters steps,https://github.com/pytorch/pytorch/issues/185006,has linked PR,44,"open PR references this issue; component/oncall labels: module: error checking, module: lrscheduler, module: optimizer",review linked PR state before touching issue,2026-05-30T11:23:50Z
issue,185640,"`make_fx` generates broken input codegen for a function with a `(tuple, dict)` positional signature",https://github.com/pytorch/pytorch/issues/185640,has linked PR,44,"open PR references this issue; component/oncall labels: module: fx, module: proxytensor, module: pt2-dispatcher, oncall: pt2",review linked PR state before touching issue,2026-05-29T16:32:38Z
issue,185543,[inductor] Gradient mismatch between eager and inductor for torch.quantile on tied values,https://github.com/pytorch/pytorch/issues/185543,has linked PR,44,"open PR references this issue; component/oncall labels: module: decompositions, module: derivatives, module: inductor, oncall: pt2",review linked PR state before touching issue,2026-05-29T14:44:20Z
issue,185630,[PT2] RuntimeError with garbage shape in apply_view_meta_sequence when using torch.compile(dynamic=True),https://github.com/pytorch/pytorch/issues/185630,has linked PR,44,"open PR references this issue; component/oncall labels: module: aotdispatch, module: dynamic shapes, module: pt2-dispatcher, oncall: pt2",review linked PR state before touching issue,2026-05-29T13:39:33Z
issue,184219,serialization_id is not deterministic across machines,https://github.com/pytorch/pytorch/issues/184219,has linked PR,44,open PR references this issue; component/oncall labels: module: serialization,review linked PR state before touching issue,2026-05-29T12:50:29Z
issue,185575,"[inductor] Upsample(mode='area') fails under torch.compile(dynamic=True) — ""cannot determine truth value of Relational"" in _adaptive_avg_pool2d lowering",https://github.com/pytorch/pytorch/issues/185575,has linked PR,44,"open PR references this issue; component/oncall labels: module: dynamic shapes, module: inductor, module: interpolation, module: pooling",review linked PR state before touching issue,2026-05-29T01:20:06Z
issue,185533,"torch.compile(mode=""max-autotune"") fails with NotImplementedError: SliceView in get_stride()",https://github.com/pytorch/pytorch/issues/185533,has linked PR,44,"open PR references this issue; component/oncall labels: module: inductor, oncall: pt2",review linked PR state before touching issue,2026-05-28T20:53:58Z
issue,180346,[DTensor] clip_grad_norm_ fails when per-parameter norms come from different DeviceMesh objects,https://github.com/pytorch/pytorch/issues/180346,has linked PR,44,"open PR references this issue; component/oncall labels: module: dtensor, oncall: distributed, oncall: distributed parallelisms",review linked PR state before touching issue,2026-05-28T14:12:40Z
issue,185216,`torch.ops.symm_mem.one_shot_all_reduce` causes `CUDA_ERROR_ILLEGAL_ADDRESS`,https://github.com/pytorch/pytorch/issues/185216,has linked PR,44,"open PR references this issue; component/oncall labels: module: symm_mem, oncall: distributed",review linked PR state before touching issue,2026-05-27T23:07:31Z
issue,185296,[AOTAutograd] [DTensor] subclass output SymInt arity mismatch causes partitioner assertion,https://github.com/pytorch/pytorch/issues/185296,has linked PR,44,"open PR references this issue; component/oncall labels: module: aotdispatch, module: dtensor, module: pt2-dispatcher, oncall: distributed",review linked PR state before touching issue,2026-05-26T23:32:28Z
issue,176770,"torch.compile error in unit tests, but test passes when ran individually",https://github.com/pytorch/pytorch/issues/176770,has linked PR,44,"open PR references this issue; component/oncall labels: module: dynamic shapes, module: inductor, oncall: pt2",review linked PR state before touching issue,2026-05-26T17:55:03Z
issue,185046,[Bug] `torch.library._register_fake` crashes when any unrelated `LazyLoader` module in `sys.modules` fails to import,https://github.com/pytorch/pytorch/issues/185046,has linked PR,44,"open PR references this issue; component/oncall labels: module: custom-operators, module: library, module: pt2-dispatcher, module: vllm",review linked PR state before touching issue,2026-05-26T17:51:45Z
issue,185200,[C++] Print tensors without storage,https://github.com/pytorch/pytorch/issues/185200,has linked PR,44,"open PR references this issue; component/oncall labels: module: cpp, module: printing",review linked PR state before touching issue,2026-05-26T15:08:45Z
issue,185141,"[Dynamo] torchvision.io functions raise ""RuntimeError when making fake tensor call"" instead of graph breaking",https://github.com/pytorch/pytorch/issues/185141,has linked PR,44,"open PR references this issue; component/oncall labels: module: dynamo, oncall: pt2",review linked PR state before touching issue,2026-05-26T14:36:57Z
issue,102948,[onnx] aten::cumprod cannot be exported to ONNX,https://github.com/pytorch/pytorch/issues/102948,has linked PR,44,open PR references this issue; component/oncall labels: module: onnx,review linked PR state before touching issue,2026-05-26T14:02:18Z
issue,184842,`torch.autograd.functional.jacobian` raises `NotImplementedError` at execution time for `strategy='forward-mode'` with `vectorize=False` instead of raising `ValueError` at argument validation,https://github.com/pytorch/pytorch/issues/184842,has linked PR,44,"open PR references this issue; component/oncall labels: module: autograd, module: error checking, module: forward ad",review linked PR state before touching issue,2026-05-26T13:42:39Z
issue,183957,enum custom op input is a little weird with torch.library.define,https://github.com/pytorch/pytorch/issues/183957,has linked PR,44,"open PR references this issue; component/oncall labels: module: custom-operators, module: library, module: pt2-dispatcher, oncall: pt2",review linked PR state before touching issue,2026-05-26T13:32:46Z
issue,184572,`torch.maximum` / `torch.clamp_min` gradients at tie point violate PyTorch's documented subgradient convention and are inconsistent with `F.relu`,https://github.com/pytorch/pytorch/issues/184572,has linked PR,44,"open PR references this issue; component/oncall labels: module: autograd, module: edge cases",review linked PR state before touching issue,2026-05-26T13:20:37Z
issue,141184,Dynamo graph break on RNN/LSTM doesn't resume the function correctly.,https://github.com/pytorch/pytorch/issues/141184,has linked PR,44,"open PR references this issue; component/oncall labels: module: dynamo, oncall: pt2",review linked PR state before touching issue,2026-05-25T12:57:08Z
issue,185023,register_buffer-created attributes and Parameter.data incorrectly typed as Tensor | Module,https://github.com/pytorch/pytorch/issues/185023,has linked PR,44,"open PR references this issue; component/oncall labels: module: nn, module: typing",review linked PR state before touching issue,2026-05-24T12:49:58Z
issue,185022,ModuleList not recognized as Iterable by type checkers,https://github.com/pytorch/pytorch/issues/185022,has linked PR,44,"open PR references this issue; component/oncall labels: module: nn, module: typing",review linked PR state before touching issue,2026-05-23T22:55:49Z
issue,184890,vmap of autograd.grad fails on the output of model run on AMP,https://github.com/pytorch/pytorch/issues/184890,has linked PR,44,"open PR references this issue; component/oncall labels: module: amp (automated mixed precision), module: autograd, module: functorch, module: norms and normalization",review linked PR state before touching issue,2026-05-22T18:04:59Z
issue,183971,`ldd libtorch_nvshmem.so` in 2.12 libtorch results in `libnvshmem_host.so.3 => not found`,https://github.com/pytorch/pytorch/issues/183971,has linked PR,44,"open PR references this issue; component/oncall labels: module: cpp, module: cuda",review linked PR state before touching issue,2026-05-22T15:40:59Z
issue,184845,"`torch.nn.functional.pad` raises `NotImplementedError` for valid `(ndim, pad_size)` combinations in non-constant modes — constraint not documented in the API",https://github.com/pytorch/pytorch/issues/184845,has linked PR,44,"open PR references this issue; component/oncall labels: module: docs, module: error checking, module: nn, module: padding",review linked PR state before touching issue,2026-05-22T03:39:32Z
issue,184642,`Module.to()` has no overload for `.to(memory_format=torch.channels_last)`,https://github.com/pytorch/pytorch/issues/184642,has linked PR,44,"open PR references this issue; component/oncall labels: module: memory format, module: nn, module: typing",review linked PR state before touching issue,2026-05-21T02:44:32Z
issue,161113,[UED][Falcon-7B x vLLM] selective_scan_fwd custom op runtime error,https://github.com/pytorch/pytorch/issues/161113,has linked PR,44,"open PR references this issue; component/oncall labels: module: dynamic shapes, module: inductor, module: vllm, oncall: pt2",review linked PR state before touching issue,2026-05-19T22:58:20Z
issue,183988,[inductor] `torch.compile` crashes on `x[torch.randperm(n)]` — pattern matcher missing `slice_shape` input,https://github.com/pytorch/pytorch/issues/183988,has linked PR,44,"open PR references this issue; component/oncall labels: module: advanced indexing, module: inductor, oncall: pt2",review linked PR state before touching issue,2026-05-19T18:09:43Z
issue,183899,torch.compile fails on addmm with beta=0 and non-broadcastable bias shape,https://github.com/pytorch/pytorch/issues/183899,has linked PR,44,"open PR references this issue; component/oncall labels: module: inductor, module: linear algebra, oncall: pt2",review linked PR state before touching issue,2026-05-19T18:09:30Z
issue,183887,torch.compile fails on int64 addmm/addmm_ comparison with fake tensor RuntimeError,https://github.com/pytorch/pytorch/issues/183887,has linked PR,44,"open PR references this issue; component/oncall labels: module: inductor, oncall: pt2",review linked PR state before touching issue,2026-05-19T18:09:16Z
issue,183249,`torch.compile` produces wrong results for `fractional_max_pool2d` when `_random_samples` is not provided,https://github.com/pytorch/pytorch/issues/183249,has linked PR,44,"open PR references this issue; component/oncall labels: module: correctness (silent), module: inductor, module: pooling, module: random",review linked PR state before touching issue,2026-05-19T18:08:11Z
issue,180823,"[Inductor] RuntimeError: weakref to UntypedStorage when compiling transpose(-2, -1) under inference_mode",https://github.com/pytorch/pytorch/issues/180823,has linked PR,44,"open PR references this issue; component/oncall labels: module: inductor, oncall: pt2",review linked PR state before touching issue,2026-05-19T18:07:29Z
issue,179368,torch.compile + view_as_complex: RuntimeError: Tensor must have a last dimension with stride 1,https://github.com/pytorch/pytorch/issues/179368,has linked PR,44,"open PR references this issue; component/oncall labels: module: complex, module: inductor, module: memory format, oncall: pt2",review linked PR state before touching issue,2026-05-19T18:07:20Z
issue,176929,Addcmul CPU bitwise numerics not matching eager w/ torch.compile,https://github.com/pytorch/pytorch/issues/176929,has linked PR,44,"open PR references this issue; component/oncall labels: module: inductor, oncall: pt2",review linked PR state before touching issue,2026-05-19T18:06:50Z
issue,172942,cuda graphs dont play nice with numpy numbers,https://github.com/pytorch/pytorch/issues/172942,has linked PR,44,"open PR references this issue; component/oncall labels: module: cuda graphs, module: inductor, oncall: pt2",review linked PR state before touching issue,2026-05-19T18:06:29Z
issue,166604,[inductor][pattern_matcher] RuntimeError in joint graph passes when compiling simple `randperm` + advanced indexing pipeline,https://github.com/pytorch/pytorch/issues/166604,has linked PR,44,"open PR references this issue; component/oncall labels: module: inductor, oncall: pt2",review linked PR state before touching issue,2026-05-19T18:05:52Z
issue,160124,FX graph segments for multiple triton kernels from reduction node are broken,https://github.com/pytorch/pytorch/issues/160124,has linked PR,44,"open PR references this issue; component/oncall labels: module: fx, module: inductor, oncall: pt2",review linked PR state before touching issue,2026-05-19T18:05:31Z
issue,154146,inductor unbacked codegen results in undefined inputs,https://github.com/pytorch/pytorch/issues/154146,has linked PR,44,"open PR references this issue; component/oncall labels: module: dynamic shapes, module: inductor, oncall: pt2",review linked PR state before touching issue,2026-05-19T18:05:04Z
issue,148751,[Inductor] Torch inductor pattern match breaks topological sort after replace the pattern,https://github.com/pytorch/pytorch/issues/148751,has linked PR,44,"open PR references this issue; component/oncall labels: module: inductor, oncall: pt2",review linked PR state before touching issue,2026-05-19T18:04:52Z
issue,138301,max_block['X'] not synced with max XBLOCK size (error shows up in `create_block_mask`),https://github.com/pytorch/pytorch/issues/138301,has linked PR,44,open PR references this issue; component/oncall labels: module: inductor,review linked PR state before touching issue,2026-05-19T18:04:19Z
issue,174472,[Inductor][CSE] `split_with_sizes` not deduplicated on B200 with FP8 quantization,https://github.com/pytorch/pytorch/issues/174472,has linked PR,44,"open PR references this issue; component/oncall labels: module: inductor, module: vllm, oncall: pt2",review linked PR state before touching issue,2026-05-19T15:22:23Z
issue,170798,nn.functional.pad silently changes the padding value if input dtype is int64 and padding value is larger than 2^53,https://github.com/pytorch/pytorch/issues/170798,has linked PR,44,"open PR references this issue; component/oncall labels: module: 64-bit, module: correctness (silent), module: edge cases, module: nn",review linked PR state before touching issue,2026-05-19T15:17:56Z
issue,170737,Expect to open the is_privateuseone branch in the LSTM interface of PyTorch's RNN.cpp to develop corresponding LSTM functionality on NPU.,https://github.com/pytorch/pytorch/issues/170737,has linked PR,44,open PR references this issue; component/oncall labels: module: privateuse1,review linked PR state before touching issue,2026-05-19T14:25:29Z
issue,184282,LazyBatchNorm2d complex64 input has inconsistent CPU and CUDA errors,https://github.com/pytorch/pytorch/issues/184282,has linked PR,44,"open PR references this issue; component/oncall labels: module: complex, module: cudnn, module: error checking, module: nn",review linked PR state before touching issue,2026-05-19T07:04:21Z
issue,151649,Inductor pattern matcher replaces aten.reshape with aten.view in pattern,https://github.com/pytorch/pytorch/issues/151649,has linked PR,44,"open PR references this issue; component/oncall labels: module: inductor, module: vllm, oncall: pt2",review linked PR state before touching issue,2026-05-18T21:43:54Z
issue,133254,Shared memory out of resource when using flex attention,https://github.com/pytorch/pytorch/issues/133254,has linked PR,44,"open PR references this issue; component/oncall labels: module: flex attention, module: inductor, oncall: pt2",review linked PR state before touching issue,2026-05-18T21:43:02Z
issue,179272,`torch.compile` returns different output stride/contiguity from eager for upsample_nearest3d on non-contiguous input,https://github.com/pytorch/pytorch/issues/179272,has linked PR,44,"open PR references this issue; component/oncall labels: module: inductor, module: interpolation, oncall: pt2",review linked PR state before touching issue,2026-05-18T14:11:07Z
issue,178040,`torch.compile` raises RuntimeError on valid `torch.addmm` with shape mismatch where eager succeeds,https://github.com/pytorch/pytorch/issues/178040,has linked PR,44,"open PR references this issue; component/oncall labels: module: inductor, module: linear algebra, oncall: pt2",review linked PR state before touching issue,2026-05-18T06:42:07Z
issue,175833,Duplicate SDPA Input Constraints,https://github.com/pytorch/pytorch/issues/175833,has linked PR,44,"open PR references this issue; component/oncall labels: module: inductor, module: performance, module: sdpa, oncall: pt2",review linked PR state before touching issue,2026-05-18T06:40:25Z
issue,182940,Eager fake-tensor execution on symbolic torch.export metadata can specialize shapes and add guards outside tracing,https://github.com/pytorch/pytorch/issues/182940,has linked PR,44,"open PR references this issue; component/oncall labels: module: dynamic shapes, module: faketensor, module: pt2-dispatcher, oncall: pt2",review linked PR state before touching issue,2026-05-18T04:59:34Z
issue,180156,GPU two-pass variance: intermediate squared values overflow float32. CPU Welford online algorithm avoids overflow.,https://github.com/pytorch/pytorch/issues/180156,has linked PR,44,"open PR references this issue; component/oncall labels: module: cuda, module: edge cases, module: numerical-stability",review linked PR state before touching issue,2026-05-18T03:48:37Z
issue,182403,local_map deferred inlining can expose index_put_ and force default_partition fallback,https://github.com/pytorch/pytorch/issues/182403,has linked PR,44,"open PR references this issue; component/oncall labels: module: aotdispatch, module: functionalization, module: higher order operators, module: performance",review linked PR state before touching issue,2026-05-17T09:10:15Z
issue,184084,Expose public API for clearing cuBLAS workspaces (currently only private torch._C._cuda_clearCublasWorkspaces),https://github.com/pytorch/pytorch/issues/184084,has linked PR,44,"open PR references this issue; component/oncall labels: module: cublas, module: cuda, module: cudacachingallocator",review linked PR state before touching issue,2026-05-17T08:45:01Z
issue,85852,`torch.mm` produces wrong result on cpu when using in-place computation,https://github.com/pytorch/pytorch/issues/85852,has linked PR,44,open PR references this issue; component/oncall labels: module: docs,review linked PR state before touching issue,2026-05-17T03:28:46Z
issue,183901,torch.compile produces int32-overflowed results for int64 arange multiplication on CUDA,https://github.com/pytorch/pytorch/issues/183901,has linked PR,44,"open PR references this issue; component/oncall labels: module: correctness (silent), module: cuda, module: inductor, module: int overflow",review linked PR state before touching issue,2026-05-17T02:51:52Z
issue,183120,torch.compile Inductor produces wrong output for BatchNorm2d + ELU + GroupNorm + log clamp chain,https://github.com/pytorch/pytorch/issues/183120,has linked PR,44,"open PR references this issue; component/oncall labels: module: correctness (silent), module: inductor, module: norms and normalization, oncall: pt2",review linked PR state before touching issue,2026-05-17T02:50:57Z
issue,180651,"[Inductor][flex attention] flex_attention autotuner uses misleading block-mask, may causing overhead and slower configs",https://github.com/pytorch/pytorch/issues/180651,has linked PR,44,"open PR references this issue; component/oncall labels: module: flex attention, module: inductor, module: performance, oncall: pt2",review linked PR state before touching issue,2026-05-17T02:47:52Z
issue,184037,torch.nn.LPPool2d returns NaN for non-integer norm_type on signed input (x**p instead of |x|**p),https://github.com/pytorch/pytorch/issues/184037,has linked PR,44,open PR references this issue; component/oncall labels: module: nn,review linked PR state before touching issue,2026-05-16T22:45:13Z
issue,146896,Assertion error in Flex Attention backward pass when indexing a parameter,https://github.com/pytorch/pytorch/issues/146896,has linked PR,44,"open PR references this issue; component/oncall labels: module: flex attention, module: higher order operators, module: pt2-dispatcher, oncall: pt2",review linked PR state before touching issue,2026-05-15T02:11:33Z
issue,149767,`flex_attention` slower than manual attention implementation,https://github.com/pytorch/pytorch/issues/149767,has linked PR,44,"open PR references this issue; component/oncall labels: module: flex attention, module: higher order operators, module: pt2-dispatcher, oncall: pt2",review linked PR state before touching issue,2026-05-15T02:03:49Z
issue,149586,UserWarning: Dynamo does not know how to trace the builtin `None.pybind11_object.__new__.`,https://github.com/pytorch/pytorch/issues/149586,has linked PR,44,"open PR references this issue; component/oncall labels: module: compiled autograd, module: dynamo, module: flex attention, module: higher order operators",review linked PR state before touching issue,2026-05-14T17:57:32Z
issue,147135,"Don't print fw_metadata in ""Found a graph input that requires gradients""",https://github.com/pytorch/pytorch/issues/147135,has linked PR,44,"open PR references this issue; component/oncall labels: module: aotdispatch, module: pt2-dispatcher, oncall: pt2",review linked PR state before touching issue,2026-05-14T17:41:54Z
issue,148695,dynamo fakification errors with opaquetensorimpl,https://github.com/pytorch/pytorch/issues/148695,has linked PR,44,"open PR references this issue; component/oncall labels: module: autograd, module: faketensor, module: pt2-dispatcher, oncall: pt2",review linked PR state before touching issue,2026-05-14T17:41:40Z
issue,182928,[Inductor] Filter extremely bad triton configs by device information for triton heuristics,https://github.com/pytorch/pytorch/issues/182928,has linked PR,44,"open PR references this issue; component/oncall labels: module: inductor, module: performance, oncall: pt2",review linked PR state before touching issue,2026-05-14T06:55:57Z
issue,128160,[dynamo] Dynamo traces through __torch_dispatch__ on custom tensor subclasses,https://github.com/pytorch/pytorch/issues/128160,has linked PR,44,"open PR references this issue; component/oncall labels: module: dynamo, module: pt2-dispatcher, oncall: pt2",review linked PR state before touching issue,2026-05-14T04:06:16Z
issue,145899,torch.compile + torch.autograd.grad: grad can be implicitly created only for scalar outputs,https://github.com/pytorch/pytorch/issues/145899,has linked PR,44,"open PR references this issue; component/oncall labels: module: aotdispatch, module: dynamo, module: pt2-dispatcher, oncall: pt2",review linked PR state before touching issue,2026-05-13T23:36:12Z
issue,183621,[MPS] For accumulating ops allow the correctness tolerance to scale within the envelope of the reference implementations drift,https://github.com/pytorch/pytorch/issues/183621,has linked PR,44,"open PR references this issue; component/oncall labels: module: half, module: mps, module: tests",review linked PR state before touching issue,2026-05-13T21:42:14Z
issue,144376,torch.compile post_accumulate_grad_hook ordering is wrong for tiebreakers,https://github.com/pytorch/pytorch/issues/144376,has linked PR,44,"open PR references this issue; component/oncall labels: module: aotdispatch, module: pt2-dispatcher, oncall: pt2",review linked PR state before touching issue,2026-05-13T04:02:15Z
issue,183396,`torch.compile` fails on `nn.CTCLoss` with missing fake impl for `aten._use_cudnn_ctc_loss.Tensor`,https://github.com/pytorch/pytorch/issues/183396,has linked PR,44,"open PR references this issue; component/oncall labels: module: cudnn, module: faketensor, module: loss, module: pt2-dispatcher",review linked PR state before touching issue,2026-05-12T01:44:24Z
issue,137635,[MPS] `torch.linspace` produces different results that CPU implementation for integral types,https://github.com/pytorch/pytorch/issues/137635,has linked PR,44,"open PR references this issue; component/oncall labels: module: correctness (silent), module: mps",review linked PR state before touching issue,2026-05-11T22:52:39Z
issue,183266,[NestedTensor] torch.compile fails when accessing _max_seqlen with GuardOnDataDependentSymNode,https://github.com/pytorch/pytorch/issues/183266,has linked PR,44,"open PR references this issue; component/oncall labels: module: dynamic shapes, module: dynamo, module: nestedtensor, oncall: pt2",review linked PR state before touching issue,2026-05-11T11:59:15Z
issue,181806,CUDA torch.signbit returns false for negative float16 NaN,https://github.com/pytorch/pytorch/issues/181806,has linked PR,44,"open PR references this issue; component/oncall labels: module: correctness (silent), module: cuda, module: edge cases, module: half",review linked PR state before touching issue,2026-05-11T05:19:13Z
issue,183015,`torch.compile(torch.vmap(torch.func.hessian(f)))` crashes with TorchRuntimeError,https://github.com/pytorch/pytorch/issues/183015,has linked PR,44,"open PR references this issue; component/oncall labels: module: functorch, oncall: pt2",review linked PR state before touching issue,2026-05-09T00:58:31Z
issue,175608,torch.compile fails to trace try-except semantic and throw `RuntimeError`,https://github.com/pytorch/pytorch/issues/175608,has linked PR,44,"open PR references this issue; component/oncall labels: module: dynamo, oncall: pt2",review linked PR state before touching issue,2026-05-08T17:10:56Z
issue,154306,graph recording observed an input tensor deallocate during graph recording that did not occur during replay,https://github.com/pytorch/pytorch/issues/154306,has linked PR,44,"open PR references this issue; component/oncall labels: module: cuda graphs, module: inductor, oncall: pt2",review linked PR state before touching issue,2026-05-08T08:27:57Z
issue,161765,Add torch compile Dynamo support for protobuf,https://github.com/pytorch/pytorch/issues/161765,has linked PR,44,"open PR references this issue; component/oncall labels: module: dynamo, oncall: pt2",review linked PR state before touching issue,2026-05-07T20:26:49Z
issue,182706,Dynamo is leaking internal state during optimizer tracing,https://github.com/pytorch/pytorch/issues/182706,has linked PR,44,"open PR references this issue; component/oncall labels: module: dynamo, module: pt2 optimizer, oncall: pt2",review linked PR state before touching issue,2026-05-06T20:47:50Z
issue,116203,Refactor precision handling in test/test_optim.py,https://github.com/pytorch/pytorch/issues/116203,has linked PR,44,open PR references this issue; component/oncall labels: module: optimizer,review linked PR state before touching issue,2026-05-06T18:47:51Z
issue,77176,Improve the overall design of MPSGraphCache,https://github.com/pytorch/pytorch/issues/77176,has linked PR,44,"open PR references this issue; component/oncall labels: module: backend, module: mps",review linked PR state before touching issue,2026-05-06T10:35:11Z
issue,181565,[C++ API][Packaging] nlohmann/json.hpp is exposed again by public headers in PyTorch 2.10 wheels,https://github.com/pytorch/pytorch/issues/181565,has linked PR,44,"open PR references this issue; component/oncall labels: module: binaries, module: cpp",review linked PR state before touching issue,2026-05-06T03:00:51Z
issue,182430,Inductor generates redundant upcast_to_fp32,https://github.com/pytorch/pytorch/issues/182430,has linked PR,44,"open PR references this issue; component/oncall labels: module: inductor, oncall: pt2",review linked PR state before touching issue,2026-05-05T13:31:23Z
issue,180397,[MPS] Add graph capture/replay API to eliminate per-op CPU dispatch overhead,https://github.com/pytorch/pytorch/issues/180397,has linked PR,44,"open PR references this issue; component/oncall labels: module: mps, module: performance",review linked PR state before touching issue,2026-05-05T08:30:25Z
issue,182318,Dynamo codegen leaks torch function enabled state via _EnableTorchFunction destructor,https://github.com/pytorch/pytorch/issues/182318,has linked PR,44,"open PR references this issue; component/oncall labels: module: __torch_function__, module: correctness (silent), module: dynamo, module: graph breaks",review linked PR state before touching issue,2026-05-05T01:33:14Z
issue,167568,Autoparallel's local_map doesn't functionalize triton kernels properly,https://github.com/pytorch/pytorch/issues/167568,has linked PR,44,"open PR references this issue; component/oncall labels: module: higher order operators, module: pt2-dispatcher, oncall: pt2",review linked PR state before touching issue,2026-05-04T23:16:19Z
issue,174017,Custom autograd Functions are not aware of partial backward pass execution,https://github.com/pytorch/pytorch/issues/174017,has linked PR,44,open PR references this issue; component/oncall labels: module: autograd,review linked PR state before touching issue,2026-05-04T22:40:16Z
issue,96409,"Initialization on `meta` device failing for models containing `nn.utils.weight_norm`, with `NotImplementedError: Could not run 'aten::_weight_norm_interface' with arguments from the 'Meta' backend.`",https://github.com/pytorch/pytorch/issues/96409,has linked PR,44,open PR references this issue; component/oncall labels: module: meta tensors,review linked PR state before touching issue,2026-05-04T19:59:17Z
issue,182036,torch.while_loop()'s body_fn is not allowed to return nothing,https://github.com/pytorch/pytorch/issues/182036,has linked PR,44,"open PR references this issue; component/oncall labels: module: higher order operators, module: pt2-dispatcher, oncall: pt2",review linked PR state before touching issue,2026-05-04T16:32:35Z
issue,181805,torch.fmax with uint8 out produces inconsistent overflow casting on CPU and CUDA,https://github.com/pytorch/pytorch/issues/181805,has linked PR,44,"open PR references this issue; component/oncall labels: module: cuda, module: python frontend",review linked PR state before touching issue,2026-05-03T22:46:12Z
issue,181807,torch.linspace with integer dtype produces inconsistent results on CPU and CUDA,https://github.com/pytorch/pytorch/issues/181807,has linked PR,44,"open PR references this issue; component/oncall labels: module: cpu, module: cuda, module: edge cases, module: tensor creation",review linked PR state before touching issue,2026-05-01T14:49:59Z
issue,181576,torch.meshgrid raises RuntimeError if no arrays passed,https://github.com/pytorch/pytorch/issues/181576,has linked PR,44,"open PR references this issue; component/oncall labels: module: error checking, module: python array api",review linked PR state before touching issue,2026-05-01T04:35:33Z
issue,181891,torch.cond()'s true_fn and false_fn are not allowed to return None,https://github.com/pytorch/pytorch/issues/181891,has linked PR,44,"open PR references this issue; component/oncall labels: module: cuda graphs, module: higher order operators, module: pt2-dispatcher, oncall: pt2",review linked PR state before touching issue,2026-04-29T20:01:43Z
issue,181347,"[Feedback about torch.multinomial][Docs] Inconsistency in torch.multinomial: Note says ""lower than"" but example allows ""equal to"" for num_samples without replacement",https://github.com/pytorch/pytorch/issues/181347,has linked PR,44,open PR references this issue; component/oncall labels: module: docs,review linked PR state before touching issue,2026-04-29T16:09:01Z
issue,180833,[MPS] Replace isin MPSGraph implementation with native Metal kernel and binary-search path for large inputs,https://github.com/pytorch/pytorch/issues/180833,has linked PR,44,"open PR references this issue; component/oncall labels: module: mps, module: performance",review linked PR state before touching issue,2026-04-29T11:03:58Z
issue,181177,[aot_autograd] RuntimeError: derivative for aten::_scaled_dot_product_flash_attention_for_cpu_backward is not implemented,https://github.com/pytorch/pytorch/issues/181177,has linked PR,44,"open PR references this issue; component/oncall labels: module: aotdispatch, module: cpu, module: functorch, module: pt2-dispatcher",review linked PR state before touching issue,2026-04-28T14:53:56Z
issue,181725,[MPS] nn.MultiheadAttention is ~9x slower than direct F.scaled_dot_product_attention with bit-identical output (B=1),https://github.com/pytorch/pytorch/issues/181725,has linked PR,44,"open PR references this issue; component/oncall labels: module: mps, module: nn, module: performance",review linked PR state before touching issue,2026-04-28T07:34:26Z
issue,181304,torch.compile fails with FakeTensor Assertion Error in torch/_subclasses/fake_tensor.py,https://github.com/pytorch/pytorch/issues/181304,has linked PR,44,"open PR references this issue; component/oncall labels: module: aotdispatch, module: faketensor, module: functorch, module: pt2-dispatcher",review linked PR state before touching issue,2026-04-27T14:45:44Z
issue,163836,[OpenReg] Using InlineEvent (or its base) for Carrying More Event Metadata in DeviceGuardImplInterface APIs,https://github.com/pytorch/pytorch/issues/163836,has linked PR,44,"open PR references this issue; component/oncall labels: module: openreg, module: privateuse1",review linked PR state before touching issue,2026-04-27T11:19:23Z
issue,181374,"`torch.empty((1,1), device=""cpu"", pin_memory=True).device` gives back MPS device on mac",https://github.com/pytorch/pytorch/issues/181374,has linked PR,44,"open PR references this issue; component/oncall labels: module: macos, module: mps",review linked PR state before touching issue,2026-04-27T09:10:40Z
issue,180377,[Silent Bug] torch.linalg.qr allows the same tensor for both Q and R in the 'out' argument when the input is square,https://github.com/pytorch/pytorch/issues/180377,has linked PR,44,"open PR references this issue; component/oncall labels: module: correctness (silent), module: edge cases, module: error checking, module: linear algebra",review linked PR state before touching issue,2026-04-27T08:33:50Z
issue,131765,[typing] Add missing `__all__` to modules.,https://github.com/pytorch/pytorch/issues/131765,has linked PR,44,"open PR references this issue; component/oncall labels: module: optimizer, module: typing",review linked PR state before touching issue,2026-04-25T03:28:51Z
issue,181474,Add a document for reusing the testcases in PyTorch.,https://github.com/pytorch/pytorch/issues/181474,has linked PR,44,"open PR references this issue; component/oncall labels: module: docs, module: openreg, module: tests",review linked PR state before touching issue,2026-04-25T03:26:38Z
issue,146508,Something very wrong with float16 CPU implementation,https://github.com/pytorch/pytorch/issues/146508,has linked PR,44,"open PR references this issue; component/oncall labels: module: cpu, module: half, module: performance",review linked PR state before touching issue,2026-04-24T01:49:37Z
issue,145093,Inductor aten.clone lowering ignores Conjugate and Negative dispatch keys,https://github.com/pytorch/pytorch/issues/145093,has linked PR,44,"open PR references this issue; component/oncall labels: module: correctness (silent), module: inductor, module: pt2-dispatcher, oncall: pt2",review linked PR state before touching issue,2026-04-23T21:17:36Z
issue,166209,Add helper functions to massage common 3D+ params into 2D for Muon,https://github.com/pytorch/pytorch/issues/166209,has linked PR,44,open PR references this issue; component/oncall labels: module: optimizer,review linked PR state before touching issue,2026-04-23T12:53:10Z
issue,180957,Missing CSE in TorchInductor,https://github.com/pytorch/pytorch/issues/180957,has linked PR,44,"open PR references this issue; component/oncall labels: module: inductor, module: performance, oncall: pt2",review linked PR state before touching issue,2026-04-21T03:52:54Z
issue,180467,remove TORCHDYNAMO_DYNAMIC_SHAPES=1 from docs,https://github.com/pytorch/pytorch/issues/180467,has linked PR,44,"open PR references this issue; component/oncall labels: module: docs, module: dynamic shapes, oncall: pt2",review linked PR state before touching issue,2026-04-20T20:41:08Z
issue,180855,"`""index_copy_cpu""` not implemented for `'Float8_e5m2'`",https://github.com/pytorch/pytorch/issues/180855,has linked PR,44,"open PR references this issue; component/oncall labels: module: advanced indexing, module: cpu, module: floatx (formerly float8)",review linked PR state before touching issue,2026-04-20T16:48:43Z
issue,180762,torch.utils.cpp_extension.load should support CUDA device linking (-dlink) similar to CUDAExtension,https://github.com/pytorch/pytorch/issues/180762,has linked PR,44,"open PR references this issue; component/oncall labels: module: cpp-extensions, module: cuda",review linked PR state before touching issue,2026-04-19T06:49:41Z
issue,180649,torch.linalg._powsum docstring advertises out= but it is not implemented,https://github.com/pytorch/pytorch/issues/180649,has linked PR,44,"open PR references this issue; component/oncall labels: module: docs, module: linear algebra",review linked PR state before touching issue,2026-04-17T06:29:57Z
issue,180624,[RFC] Decouple `torch.utils.cpp_extension` from setuptools and align with the scikit-build-core build backend,https://github.com/pytorch/pytorch/issues/180624,has linked PR,44,"open PR references this issue; component/oncall labels: module: build, module: cpp-extensions",review linked PR state before touching issue,2026-04-16T23:16:35Z
issue,180447,Poor BF16 torch.compile + Freezing perf on AArch64 CPUs,https://github.com/pytorch/pytorch/issues/180447,has linked PR,44,"open PR references this issue; component/oncall labels: module: arm, module: bfloat16, module: cpu, module: inductor",review linked PR state before touching issue,2026-04-16T08:01:30Z
issue,165650,"`torch.remainder` with integer divisor 0 should raise, but returns large values in CUDA",https://github.com/pytorch/pytorch/issues/165650,has linked PR,44,"open PR references this issue; component/oncall labels: module: cuda, module: error checking, module: nans and infs",review linked PR state before touching issue,2026-04-15T08:30:35Z
issue,179608,[MPS] avg_pool1d produces negative values from non-negative input when zeros follow large floats,https://github.com/pytorch/pytorch/issues/179608,has linked PR,44,"open PR references this issue; component/oncall labels: module: correctness (silent), module: edge cases, module: mps, module: pooling",review linked PR state before touching issue,2026-04-13T17:28:05Z
issue,180119,Feedback about torch.Tensor.index_add_,https://github.com/pytorch/pytorch/issues/180119,has linked PR,44,open PR references this issue; component/oncall labels: module: docs,review linked PR state before touching issue,2026-04-11T06:53:33Z
issue,112044,DDE in pack_padded_sequence/pad_packed_sequence decomposition.,https://github.com/pytorch/pytorch/issues/112044,has linked PR,44,"open PR references this issue; component/oncall labels: module: dynamic shapes, module: fx, module: rnn, oncall: pt2",review linked PR state before touching issue,2026-04-09T19:37:07Z
issue,80025,`torch.special.gammainc` backward pass with respect to the first argument,https://github.com/pytorch/pytorch/issues/80025,has linked PR,44,"open PR references this issue; component/oncall labels: module: autograd, module: distributions",review linked PR state before touching issue,2026-04-09T19:02:46Z
issue,175211,CUDA/ROCm/Accelerator testing should replace get_device_capability() with feature queries,https://github.com/pytorch/pytorch/issues/175211,has linked PR,44,"open PR references this issue; component/oncall labels: module: accelerator, module: cuda, module: rocm, module: tests",review linked PR state before touching issue,2026-04-08T20:04:40Z
issue,178379,"[bug]`torch.linalg.solve_triangular(A, B, *, out=B)`: wrong result with `B.is_neg() == True`",https://github.com/pytorch/pytorch/issues/178379,has linked PR,44,"open PR references this issue; component/oncall labels: module: correctness (silent), module: cuda, module: linear algebra",review linked PR state before touching issue,2026-04-07T13:15:17Z
issue,179128,FSDP2 HSDP: AllReduceState holds FP32 reduce buffers across all layers during backward,https://github.com/pytorch/pytorch/issues/179128,has linked PR,44,"open PR references this issue; component/oncall labels: module: fsdp, oncall: distributed",review linked PR state before touching issue,2026-04-07T03:05:25Z
issue,179259,test_torch.py does not run any tests on MPS despite number of `skipIfMPS` decorators,https://github.com/pytorch/pytorch/issues/179259,has linked PR,44,"open PR references this issue; component/oncall labels: module: mps, module: tests",review linked PR state before touching issue,2026-04-04T02:31:35Z
issue,178977,`device_id` not honored when creating `process_group` with `$TORCH_DISTRIBUTED_DEBUG=DETAIL`,https://github.com/pytorch/pytorch/issues/178977,has linked PR,44,"open PR references this issue; component/oncall labels: module: c10d, oncall: distributed",review linked PR state before touching issue,2026-04-03T21:20:12Z
issue,169477,torch.compile on @torch.inferece_mode() cause RuntimeError: <weakref ... to 'torch.storage.UntypedStorage' ...>,https://github.com/pytorch/pytorch/issues/169477,has linked PR,44,"open PR references this issue; component/oncall labels: module: aotdispatch, module: pt2-dispatcher, oncall: pt2",review linked PR state before touching issue,2026-04-03T18:45:02Z
issue,175467,[DTensor] torch.export.export() fails with RuntimeError: wrapped_call_impl - DTensorSpec not registered as pytree constant,https://github.com/pytorch/pytorch/issues/175467,has linked PR,44,"open PR references this issue; component/oncall labels: module: dtensor, oncall: distributed",review linked PR state before touching issue,2026-04-02T18:25:44Z
issue,178677,`torch.compile` silently succeeds on `TransformerEncoder` with all-masked `src_key_padding_mask` where eager raises RuntimeError,https://github.com/pytorch/pytorch/issues/178677,has linked PR,44,"open PR references this issue; component/oncall labels: module: correctness (silent), module: dynamo, module: edge cases, module: nestedtensor",review linked PR state before touching issue,2026-03-31T14:41:26Z
issue,108798,Meta kernel not raising,https://github.com/pytorch/pytorch/issues/108798,has linked PR,44,"open PR references this issue; component/oncall labels: module: dynamo, oncall: pt2",review linked PR state before touching issue,2026-03-30T18:16:23Z
issue,176211,TypeError in guard when working with list of tensors,https://github.com/pytorch/pytorch/issues/176211,has linked PR,44,"open PR references this issue; component/oncall labels: module: dynamo, module: guards, oncall: pt2",review linked PR state before touching issue,2026-03-29T10:27:51Z
issue,178128,conv_transpose3d meta execution produces output for invalid bias shape instead of matching eager validation,https://github.com/pytorch/pytorch/issues/178128,has linked PR,44,"open PR references this issue; component/oncall labels: module: convolution, module: error checking, module: meta tensors",review linked PR state before touching issue,2026-03-27T06:45:32Z
issue,178127,torch.conv_transpose2d meta execution fails to validate invalid output_padding/stride/dilation combination,https://github.com/pytorch/pytorch/issues/178127,has linked PR,44,"open PR references this issue; component/oncall labels: module: convolution, module: error checking, module: meta tensors, oncall: pt2",review linked PR state before touching issue,2026-03-27T06:45:28Z
issue,178125,conv_transpose1d meta implementation allows invalid output_padding that real CUDA execution rejects,https://github.com/pytorch/pytorch/issues/178125,has linked PR,44,"open PR references this issue; component/oncall labels: module: convolution, module: error checking, module: meta tensors",review linked PR state before touching issue,2026-03-27T06:45:23Z
issue,178388,[torch.compile] Dynamo fails to trace assignment of custom tensor subclass to module attribute:self.custom_weights = CustomTensor(self.layer1.weight.data)``,https://github.com/pytorch/pytorch/issues/178388,has linked PR,44,"open PR references this issue; component/oncall labels: module: dynamo, oncall: pt2",review linked PR state before touching issue,2026-03-25T14:26:34Z
issue,178317,[inductor] Tolerances in test_remove_no_ops have to be lowered with native matmul,https://github.com/pytorch/pytorch/issues/178317,has linked PR,44,"open PR references this issue; component/oncall labels: module: inductor, module: tests, oncall: pt2",review linked PR state before touching issue,2026-03-24T19:35:22Z
issue,134385,FlopCounterMode doesn't support HOP,https://github.com/pytorch/pytorch/issues/134385,has linked PR,44,"open PR references this issue; component/oncall labels: module: flop counter, module: higher order operators, module: pt2-dispatcher, oncall: pt2",review linked PR state before touching issue,2026-03-24T18:39:11Z
issue,177968,Add custom precision or tolerance support for out of tree backends,https://github.com/pytorch/pytorch/issues/177968,has linked PR,44,"open PR references this issue; component/oncall labels: module: backend, module: tests",review linked PR state before touching issue,2026-03-24T14:08:26Z
issue,178248,[TorchDynamo] Failed to trace builtin operator 'setattr' when registering forward hook inside forward method,https://github.com/pytorch/pytorch/issues/178248,has linked PR,44,"open PR references this issue; component/oncall labels: module: dynamo, oncall: pt2",review linked PR state before touching issue,2026-03-24T07:23:54Z
issue,161863,torch.testing.assert_close doesn't accept tensor-likes,https://github.com/pytorch/pytorch/issues/161863,has linked PR,44,"open PR references this issue; component/oncall labels: module: python frontend, module: testing",review linked PR state before touching issue,2026-03-24T00:31:37Z
issue,177237,torch.conv1d incorrectly accepts invalid weight shape in meta backend while raising correct error in eager execution,https://github.com/pytorch/pytorch/issues/177237,has linked PR,44,"open PR references this issue; component/oncall labels: module: convolution, module: correctness (silent), module: dynamo, module: error checking",review linked PR state before touching issue,2026-03-23T23:16:38Z
issue,128712,Add Swiglu activation function,https://github.com/pytorch/pytorch/issues/128712,has linked PR,44,open PR references this issue; component/oncall labels: module: nn,review linked PR state before touching issue,2026-03-23T22:17:14Z
issue,172324,[OpenReg] Add documentation for standardized testing patterns.,https://github.com/pytorch/pytorch/issues/172324,has linked PR,44,"open PR references this issue; component/oncall labels: module: docs, module: openreg",review linked PR state before touching issue,2026-03-23T07:48:42Z
issue,160094,torch.library.infer_schema doesn't have full support for list types,https://github.com/pytorch/pytorch/issues/160094,has linked PR,44,"open PR references this issue; component/oncall labels: module: custom-operators, module: pt2-dispatcher, module: vllm, oncall: pt2",review linked PR state before touching issue,2026-03-22T23:24:57Z
issue,177750,[functorch] vmap and functionalize under inference_mode,https://github.com/pytorch/pytorch/issues/177750,has linked PR,44,"open PR references this issue; component/oncall labels: module: correctness (silent), module: functionalization, module: functorch, module: pt2-dispatcher",review linked PR state before touching issue,2026-03-20T21:37:00Z
issue,160939,Strange failure modes of while_loop under torch.compile -- possibly due to internal array stride,https://github.com/pytorch/pytorch/issues/160939,has linked PR,44,"open PR references this issue; component/oncall labels: module: higher order operators, module: pt2-dispatcher, oncall: pt2",review linked PR state before touching issue,2026-03-20T18:39:03Z
issue,91898,Lazily start worker threads in the autograd engine,https://github.com/pytorch/pytorch/issues/91898,has linked PR,44,open PR references this issue; component/oncall labels: module: autograd,review linked PR state before touching issue,2026-03-19T10:40:27Z
issue,91468,torch.compile incorrect when imperative autograd APIs are used,https://github.com/pytorch/pytorch/issues/91468,has linked PR,44,"open PR references this issue; component/oncall labels: module: aotdispatch, module: autograd, module: correctness (silent), module: pt2-dispatcher",review linked PR state before touching issue,2026-03-18T21:56:26Z
issue,177712,[dynamo] [torch.compile] sdpa become slower after torch._dynamo.run in dynamo benchmark when attention_mask is None,https://github.com/pytorch/pytorch/issues/177712,has linked PR,44,"open PR references this issue; component/oncall labels: module: dynamo, module: performance, module: sdpa, oncall: pt2",review linked PR state before touching issue,2026-03-18T02:54:15Z
issue,167070,Inductor doesn't accept Sympy.Eq as input,https://github.com/pytorch/pytorch/issues/167070,has linked PR,44,"open PR references this issue; component/oncall labels: module: dynamic shapes, module: inductor, module: pt2-dispatcher, oncall: pt2",review linked PR state before touching issue,2026-03-17T23:57:06Z
issue,177127,torch.multinomial: add flag to skip input validation (66% of GPU time),https://github.com/pytorch/pytorch/issues/177127,has linked PR,44,open PR references this issue; component/oncall labels: module: performance,review linked PR state before touching issue,2026-03-17T19:21:10Z
issue,116202,"Dynamo'd `test_mixed_device_dtype` needs higher tolerances for Adagrad, RMSProp",https://github.com/pytorch/pytorch/issues/116202,has linked PR,44,"open PR references this issue; component/oncall labels: module: dynamo, module: optimizer, module: tests, oncall: pt2",review linked PR state before touching issue,2026-03-17T19:17:55Z
issue,167098,request for faster torch.compile kernels for casting to nvfp4,https://github.com/pytorch/pytorch/issues/167098,has linked PR,44,open PR references this issue; component/oncall labels: oncall: pt2,review linked PR state before touching issue,2026-03-17T14:03:40Z
issue,153825,'torch.sparse.to_sparse_semi_structured' significantly worsens performance on H100 GPUs,https://github.com/pytorch/pytorch/issues/153825,has linked PR,44,"open PR references this issue; component/oncall labels: module: cuda, module: performance, module: sparse",review linked PR state before touching issue,2026-03-16T20:01:15Z
issue,177265,Compiling CppExtensions in parallel causes race condition,https://github.com/pytorch/pytorch/issues/177265,has linked PR,44,"open PR references this issue; component/oncall labels: module: build, module: cpp-extensions",review linked PR state before touching issue,2026-03-16T17:15:29Z
issue,175858,RFC: `torch._dynamo.mark_dynamic_dict` — skip key-set guards on dict mutations,https://github.com/pytorch/pytorch/issues/175858,has linked PR,44,"open PR references this issue; component/oncall labels: module: compile ux, module: compile-time, module: dynamo, oncall: pt2",review linked PR state before touching issue,2026-03-15T00:07:25Z
issue,176244,Bug: gaussian_nll_loss — possible issue with broadcastable variance shapes having multiple unit dimensions,https://github.com/pytorch/pytorch/issues/176244,has linked PR,44,"open PR references this issue; component/oncall labels: module: error checking, module: nn, module: python frontend",review linked PR state before touching issue,2026-03-13T16:41:46Z
issue,177327,AArch64 Unit Test Failure - test_int8_woq_mm_*,https://github.com/pytorch/pytorch/issues/177327,has linked PR,44,"open PR references this issue; component/oncall labels: module: arm, module: cpu, module: inductor, module: linear algebra",review linked PR state before touching issue,2026-03-12T22:19:44Z
issue,177250,AArch64 Unit Test Failure - TestNN test_upsampling_bfloat16,https://github.com/pytorch/pytorch/issues/177250,has linked PR,44,"open PR references this issue; component/oncall labels: module: arm, module: autograd, module: nn",review linked PR state before touching issue,2026-03-12T11:37:25Z
issue,127112,[dynamo] Module tracker test errors in dynamo at runtime,https://github.com/pytorch/pytorch/issues/127112,has linked PR,44,"open PR references this issue; component/oncall labels: module: dynamo, oncall: pt2",review linked PR state before touching issue,2026-03-11T21:26:22Z
issue,124344,Inconsistent argument parsing behavior for torch.ops bindings between Tensor and FakeTensor,https://github.com/pytorch/pytorch/issues/124344,has linked PR,44,"open PR references this issue; component/oncall labels: module: dynamic shapes, oncall: pt2",review linked PR state before touching issue,2026-03-11T21:19:25Z
issue,177109,[RFC] Make torch.library.custom_op faster in the common case,https://github.com/pytorch/pytorch/issues/177109,has linked PR,44,"open PR references this issue; component/oncall labels: module: custom-operators, module: performance, module: pt2-dispatcher, oncall: pt2",review linked PR state before touching issue,2026-03-11T17:44:41Z
issue,138141,torch.cond with shape condition in predicate doesn't allow for assertions,https://github.com/pytorch/pytorch/issues/138141,has linked PR,44,"open PR references this issue; component/oncall labels: module: dynamic shapes, module: higher order operators, module: pt2-dispatcher, oncall: pt2",review linked PR state before touching issue,2026-03-11T17:14:14Z
issue,154329,MPS Memory Leak,https://github.com/pytorch/pytorch/issues/154329,has linked PR,44,"open PR references this issue; component/oncall labels: module: memory usage, module: mps",review linked PR state before touching issue,2026-03-11T05:49:52Z
issue,145773,Dynamo performance test benchmarks reuse model state between eager and compiled,https://github.com/pytorch/pytorch/issues/145773,has linked PR,44,"open PR references this issue; component/oncall labels: module: benchmark, module: cuda graphs, module: dynamo, oncall: pt2",review linked PR state before touching issue,2026-03-10T22:09:29Z
issue,175001,torch.cond() and cuda graph mega issue,https://github.com/pytorch/pytorch/issues/175001,has linked PR,44,"open PR references this issue; component/oncall labels: module: cuda graphs, module: higher order operators, module: pt2-dispatcher, oncall: pt2",review linked PR state before touching issue,2026-03-10T22:07:04Z
issue,150063,Torch.check does not preserve original error message in the deferred runtime assert,https://github.com/pytorch/pytorch/issues/150063,has linked PR,44,"open PR references this issue; component/oncall labels: module: dynamic shapes, module: dynamo, oncall: pt2",review linked PR state before touching issue,2026-03-10T16:04:27Z
issue,138795,compile + allgather with group will fail for stack-style allgather,https://github.com/pytorch/pytorch/issues/138795,has linked PR,44,"open PR references this issue; component/oncall labels: module: vllm, oncall: distributed, oncall: pt2",review linked PR state before touching issue,2026-03-10T05:23:43Z
issue,142228,Scaled Dot-Product Attention Invalid Configuration Error on Large batch size,https://github.com/pytorch/pytorch/issues/142228,has linked PR,44,open PR references this issue; component/oncall labels: module: sdpa,review linked PR state before touching issue,2026-03-06T08:29:48Z
issue,152656,"Add explicit error message for def infer_size(a, b): that specificy that non broadcast path was picked due to unbacked existing in both inputs.",https://github.com/pytorch/pytorch/issues/152656,has linked PR,44,"open PR references this issue; component/oncall labels: module: dynamic shapes, oncall: pt2",review linked PR state before touching issue,2026-03-05T06:52:55Z
issue,167729,Dynamo should support torch.autograd.grad,https://github.com/pytorch/pytorch/issues/167729,has linked PR,44,"open PR references this issue; component/oncall labels: module: dynamo, oncall: pt2",review linked PR state before touching issue,2026-03-05T01:49:51Z
issue,169538,`torch.compile` fails on `addmm_` with int64 tensors due to fake-tensor meta error (`expand: attempting to expand a dimension of length 2 -> 1`),https://github.com/pytorch/pytorch/issues/169538,has linked PR,44,"open PR references this issue; component/oncall labels: module: dynamo, oncall: pt2",review linked PR state before touching issue,2026-03-05T01:16:32Z
issue,117122,Improve behaviour of `torch.linalg.lstsq` on CUDA GPU for rank defficient matrices,https://github.com/pytorch/pytorch/issues/117122,has linked PR,44,"open PR references this issue; component/oncall labels: module: cuda, module: linear algebra",review linked PR state before touching issue,2026-03-04T16:20:50Z
issue,176381,Output difference anomaly under equivalent migration in RReLU operator,https://github.com/pytorch/pytorch/issues/176381,has linked PR,44,open PR references this issue; component/oncall labels: module: nn,review linked PR state before touching issue,2026-03-04T05:00:19Z
issue,176296,[MPS] Binary ops on uint16/uint32/uint64 produce garbage values,https://github.com/pytorch/pytorch/issues/176296,has linked PR,44,"open PR references this issue; component/oncall labels: module: correctness (silent), module: mps, module: unsigned int",review linked PR state before touching issue,2026-03-03T23:00:54Z
issue,175477,make_fx symbolic tracing produces wrong 2nd-order gradients through torch.sqrt,https://github.com/pytorch/pytorch/issues/175477,has linked PR,44,"open PR references this issue; component/oncall labels: module: autograd, module: decompositions, module: fx, oncall: pt2",review linked PR state before touching issue,2026-03-03T19:41:05Z
issue,175408,custom_op + TensorSubclass + torch.compile interaction : Better error message or doc,https://github.com/pytorch/pytorch/issues/175408,has linked PR,44,"open PR references this issue; component/oncall labels: module: custom-operators, module: pt2-dispatcher, oncall: pt2",review linked PR state before touching issue,2026-03-03T14:05:41Z
issue,175292,[dynamo] Metaclass not supported correctly,https://github.com/pytorch/pytorch/issues/175292,has linked PR,44,"open PR references this issue; component/oncall labels: module: dynamo, oncall: pt2",review linked PR state before touching issue,2026-03-02T21:48:45Z
issue,175267,[dynamo] Audit `x is y` code path,https://github.com/pytorch/pytorch/issues/175267,has linked PR,44,"open PR references this issue; component/oncall labels: module: dynamo, oncall: pt2",review linked PR state before touching issue,2026-03-02T21:48:21Z
issue,175264,[dynamo] Investigate why we create VariableTrackers for only sourceful objects in UserDefinedClassVariable,https://github.com/pytorch/pytorch/issues/175264,has linked PR,44,"open PR references this issue; component/oncall labels: module: dynamo, oncall: pt2",review linked PR state before touching issue,2026-03-02T21:48:11Z
issue,174668,torch.histc is not consistent,https://github.com/pytorch/pytorch/issues/174668,has linked PR,44,"open PR references this issue; component/oncall labels: module: edge cases, module: half",review linked PR state before touching issue,2026-03-02T18:23:05Z
issue,175994,CI: Dead CentOS code in docker setup scripts,https://github.com/pytorch/pytorch/issues/175994,has linked PR,44,open PR references this issue; component/oncall labels: module: ci,review linked PR state before touching issue,2026-02-27T21:24:56Z
issue,175968,[opaque object] Running the compiled function with opaque object for the second time fails.,https://github.com/pytorch/pytorch/issues/175968,has linked PR,44,"open PR references this issue; component/oncall labels: module: custom-operators, module: inductor, module: pt2-dispatcher, oncall: pt2",review linked PR state before touching issue,2026-02-27T12:48:45Z
issue,174183,[test] Add error_inputs for nn.Conv2d module,https://github.com/pytorch/pytorch/issues/174183,has linked PR,44,"open PR references this issue; component/oncall labels: module: nn, module: tests",review linked PR state before touching issue,2026-02-26T13:05:43Z
issue,175521,torch.vmap working with non-tensor leaves in output pytrees,https://github.com/pytorch/pytorch/issues/175521,has linked PR,44,"open PR references this issue; component/oncall labels: module: functorch, module: pytree, module: vmap",review linked PR state before touching issue,2026-02-26T09:01:43Z
issue,175368,`F.embedding_bag` segfaults when intermediate offsets exceed indices length,https://github.com/pytorch/pytorch/issues/175368,has linked PR,44,"open PR references this issue; component/oncall labels: module: error checking, module: nn",review linked PR state before touching issue,2026-02-26T06:25:18Z
issue,175489,"CPU `F.interpolate(..., antialias=True)` raises NotImplementedError for bfloat16/float16 (bilinear/bicubic)",https://github.com/pytorch/pytorch/issues/175489,has linked PR,44,"open PR references this issue; component/oncall labels: module: bfloat16, module: cpu, module: half, module: nn",review linked PR state before touching issue,2026-02-23T18:45:02Z
issue,174984,`torch.fft.rfft` errors out with MKL FFT error when given an empty tensor,https://github.com/pytorch/pytorch/issues/174984,has linked PR,44,"open PR references this issue; component/oncall labels: module: edge cases, module: empty tensor, module: error checking, module: fft",review linked PR state before touching issue,2026-02-23T17:49:21Z
issue,175522,while_loop vmap compatibility,https://github.com/pytorch/pytorch/issues/175522,has linked PR,44,"open PR references this issue; component/oncall labels: module: functorch, module: higher order operators, module: pt2-dispatcher, module: vmap",review linked PR state before touching issue,2026-02-23T10:04:27Z
issue,139707,Adding backend-specific autograd kernel for aten builtin operators breaks `torch.compile`,https://github.com/pytorch/pytorch/issues/139707,has linked PR,44,"open PR references this issue; component/oncall labels: module: custom-operators, module: pt2-dispatcher, oncall: pt2",review linked PR state before touching issue,2026-02-23T04:00:20Z
issue,138264,`_native_batch_norm_legit`: `aot_eager` backend incorrectly raising an error.,https://github.com/pytorch/pytorch/issues/138264,has linked PR,44,"open PR references this issue; component/oncall labels: module: pt2-dispatcher, oncall: pt2",review linked PR state before touching issue,2026-02-22T20:30:55Z
issue,138229,[cond] Cond lifts duplicate symint inputs,https://github.com/pytorch/pytorch/issues/138229,has linked PR,44,"open PR references this issue; component/oncall labels: module: higher order operators, module: pt2-dispatcher, oncall: pt2",review linked PR state before touching issue,2026-02-22T20:27:43Z
issue,122386,[histogram] expand: attempting to expand a dimension of length 6!,https://github.com/pytorch/pytorch/issues/122386,has linked PR,44,"open PR references this issue; component/oncall labels: module: functionalization, module: pt2-dispatcher, oncall: pt2",review linked PR state before touching issue,2026-02-22T20:21:40Z
issue,133512,functionalization doesn't faithfully reproduce strides of intermediate values,https://github.com/pytorch/pytorch/issues/133512,has linked PR,44,"open PR references this issue; component/oncall labels: module: functionalization, module: pt2-dispatcher, oncall: pt2",review linked PR state before touching issue,2026-02-22T04:30:40Z
issue,132896,AOTAutograd.process_inputs doesn't work on nested subclasses,https://github.com/pytorch/pytorch/issues/132896,has linked PR,44,"open PR references this issue; component/oncall labels: module: aotdispatch, module: pt2-dispatcher, oncall: pt2",review linked PR state before touching issue,2026-02-22T04:00:05Z
issue,122860,Do not use assert for user triggered issues,https://github.com/pytorch/pytorch/issues/122860,has linked PR,44,"open PR references this issue; component/oncall labels: module: functorch, module: pt2-dispatcher, oncall: export, oncall: pt2",review linked PR state before touching issue,2026-02-22T03:48:28Z
issue,117555,Update decompositions registered to `Autograd` dispatch key to also register `CompositeImplicitAutograd`,https://github.com/pytorch/pytorch/issues/117555,has linked PR,44,"open PR references this issue; component/oncall labels: module: aotdispatch, module: decompositions, module: pt2-dispatcher, module: python dispatcher",review linked PR state before touching issue,2026-02-22T03:38:48Z
issue,174541,torch.compile: ConstDictVariable should support runtime dict lookup for dynamic keys to avoid recompilation,https://github.com/pytorch/pytorch/issues/174541,has linked PR,44,"open PR references this issue; component/oncall labels: module: compile-time, module: dynamo, oncall: pt2",review linked PR state before touching issue,2026-02-21T16:09:37Z
issue,90923,dynamo.optimizations.training.aot_autograd does not trace correct overload,https://github.com/pytorch/pytorch/issues/90923,has linked PR,44,"open PR references this issue; component/oncall labels: module: aotdispatch, module: pt2-dispatcher, oncall: export, oncall: pt2",review linked PR state before touching issue,2026-02-21T05:00:03Z
issue,124181,AOTAutograd breaks diffusers VAE decoder,https://github.com/pytorch/pytorch/issues/124181,has linked PR,44,"open PR references this issue; component/oncall labels: module: pt2-dispatcher, oncall: pt2",review linked PR state before touching issue,2026-02-21T02:49:33Z
issue,122029,AOTAutograd has high fixed overheads,https://github.com/pytorch/pytorch/issues/122029,has linked PR,44,"open PR references this issue; component/oncall labels: module: aotdispatch, module: pt2-dispatcher, oncall: pt2",review linked PR state before touching issue,2026-02-21T02:20:57Z
issue,90507,Functionalization on inplace_views should properly reflect autograd metadata,https://github.com/pytorch/pytorch/issues/90507,has linked PR,44,"open PR references this issue; component/oncall labels: module: bootcamp, module: functionalization, module: pt2-dispatcher, oncall: pt2",review linked PR state before touching issue,2026-02-20T19:55:28Z
issue,113899,Tensor.requires_grad_() does not trace with make_fx,https://github.com/pytorch/pytorch/issues/113899,has linked PR,44,"open PR references this issue; component/oncall labels: module: proxytensor, module: pt2-dispatcher, oncall: pt2",review linked PR state before touching issue,2026-02-20T19:43:56Z
issue,114415,AOTAutograd should detect aliasing of inputs that happens *below* subclasses,https://github.com/pytorch/pytorch/issues/114415,has linked PR,44,"open PR references this issue; component/oncall labels: module: __torch_dispatch__, module: aotdispatch, module: pt2-dispatcher, oncall: pt2",review linked PR state before touching issue,2026-02-20T18:36:37Z
issue,140707,mutating a buffer to change its requires_grad-ness can cause RuntimeError,https://github.com/pytorch/pytorch/issues/140707,has linked PR,44,"open PR references this issue; component/oncall labels: module: aotdispatch, module: pt2-dispatcher, oncall: pt2",review linked PR state before touching issue,2026-02-20T03:43:28Z
issue,173816,make should_preserve_node_meta and relevant variable TLS,https://github.com/pytorch/pytorch/issues/173816,has linked PR,44,"open PR references this issue; component/oncall labels: module: dynamo, module: fx, oncall: pt2",review linked PR state before touching issue,2026-02-19T10:37:47Z
issue,175324,Should `overrides.is_tensor_like` return a `TypeGuard[Tensor]`?,https://github.com/pytorch/pytorch/issues/175324,has linked PR,44,"open PR references this issue; component/oncall labels: module: autograd, module: typing",review linked PR state before touching issue,2026-02-19T10:24:07Z
issue,172538,Support native structured kernel and out_as_primary: true for PrivateUse1,https://github.com/pytorch/pytorch/issues/172538,has linked PR,44,open PR references this issue; component/oncall labels: module: privateuse1,review linked PR state before touching issue,2026-02-18T20:24:08Z
issue,145383,Windows Pytorch compiler crash some version of cl.exe. Fix provided,https://github.com/pytorch/pytorch/issues/145383,has linked PR,44,"open PR references this issue; component/oncall labels: module: windows, oncall: pt2",review linked PR state before touching issue,2026-02-18T01:34:13Z
issue,172026,`torch.compile(torch.func.vjp)` crashes when creating a new `Tensor`,https://github.com/pytorch/pytorch/issues/172026,has linked PR,44,"open PR references this issue; component/oncall labels: module: autograd, module: functorch, oncall: pt2",review linked PR state before touching issue,2026-02-17T18:53:05Z
issue,174682,spin lint doesn't regenerate cuda headers,https://github.com/pytorch/pytorch/issues/174682,has linked PR,44,"open PR references this issue; component/oncall labels: module: build, module: lint, module: spin",review linked PR state before touching issue,2026-02-17T16:52:13Z
issue,174176,Inductor crashes with INTERNAL ASSERT in FunctionalizeFallbackKernel,https://github.com/pytorch/pytorch/issues/174176,has linked PR,44,"open PR references this issue; component/oncall labels: module: functionalization, module: pt2-dispatcher, oncall: pt2",review linked PR state before touching issue,2026-02-16T19:39:12Z
issue,174371,"[inductor][aot_eager] Incorrect results with torch.compile when using clone, as_strided followed by add_",https://github.com/pytorch/pytorch/issues/174371,has linked PR,44,"open PR references this issue; component/oncall labels: module: functionalization, module: pt2-dispatcher, oncall: pt2",review linked PR state before touching issue,2026-02-16T09:31:33Z
issue,140000,torch.argmax slow on CPU,https://github.com/pytorch/pytorch/issues/140000,has linked PR,44,"open PR references this issue; component/oncall labels: module: cpu, module: performance",review linked PR state before touching issue,2026-02-14T21:46:49Z
issue,148651,Avoid fork for TORCHINDUCTOR_COMPILE_THREADS > 1,https://github.com/pytorch/pytorch/issues/148651,has linked PR,44,"open PR references this issue; component/oncall labels: module: inductor, oncall: pt2",review linked PR state before touching issue,2026-02-13T22:31:42Z
issue,171355,torch.utils.checkpoint.CheckpointError with qwen3 MoE,https://github.com/pytorch/pytorch/issues/171355,has linked PR,44,open PR references this issue; component/oncall labels: module: activation checkpointing,review linked PR state before touching issue,2026-02-12T13:36:07Z
issue,174625,document `torch.cuda.memory.mem_get_info` in,https://github.com/pytorch/pytorch/issues/174625,has linked PR,44,"open PR references this issue; component/oncall labels: module: cuda, module: docs",review linked PR state before touching issue,2026-02-12T00:05:09Z
issue,168972,Aarch64 unit test failures - test/test_cpu_repro.py CPUReproTests.test_float32_to_uint8,https://github.com/pytorch/pytorch/issues/168972,has linked PR,44,"open PR references this issue; component/oncall labels: module: arm, module: inductor, oncall: pt2",review linked PR state before touching issue,2026-02-11T10:37:54Z
issue,169188,"`torch.compile(backend=""tvm"")` produces wrong result on simple `nn.Parameter` addition (eager = 2.0, compiled = 0.0)",https://github.com/pytorch/pytorch/issues/169188,has linked PR,44,"open PR references this issue; component/oncall labels: module: backend, oncall: pt2",review linked PR state before touching issue,2026-02-07T15:44:11Z
issue,169111,"`torch.compile(backend=""tvm"", fullgraph=True, dynamic=True)` fails on PyTorch 2.9.0 with `RuntimeError: Tracer cannot infer type ... Only tensors ... supported, but got SymInt`",https://github.com/pytorch/pytorch/issues/169111,has linked PR,44,"open PR references this issue; component/oncall labels: module: backend, oncall: pt2",review linked PR state before touching issue,2026-02-07T15:06:51Z
issue,174177,[test] Add error_inputs for nn.Linear module,https://github.com/pytorch/pytorch/issues/174177,has linked PR,44,open PR references this issue; component/oncall labels: module: tests,review linked PR state before touching issue,2026-02-04T15:23:46Z
issue,173912,Wrong mode in `torch.distributions.Kumaraswamy`,https://github.com/pytorch/pytorch/issues/173912,has linked PR,44,open PR references this issue; component/oncall labels: module: distributions,review linked PR state before touching issue,2026-02-02T18:29:03Z
issue,173578,FunctionIdSet should store weakrefs instead of object ids.,https://github.com/pytorch/pytorch/issues/173578,has linked PR,44,"open PR references this issue; component/oncall labels: module: dynamo, oncall: pt2",review linked PR state before touching issue,2026-01-30T12:13:52Z
issue,146249,Replace `*args : Any` with `typing_extensions.TypeVarTuple`,https://github.com/pytorch/pytorch/issues/146249,has linked PR,44,open PR references this issue; component/oncall labels: module: typing,review linked PR state before touching issue,2026-01-30T12:13:17Z
issue,173881,[Build][ppc64le] Fixes required for compiling PyTorch 2.5.1 on PowerPC with GCC 12,https://github.com/pytorch/pytorch/issues/173881,has linked PR,44,"open PR references this issue; component/oncall labels: module: build, module: power, oncall: releng",review linked PR state before touching issue,2026-01-30T04:51:40Z
issue,170869,Feedback about torch.addcdiv,https://github.com/pytorch/pytorch/issues/170869,has linked PR,44,"open PR references this issue; component/oncall labels: module: docs, module: edge cases",review linked PR state before touching issue,2026-01-29T13:17:37Z
issue,173052,[Inconsistency] aten.linalg_pinv returns inconsistent output shape between eager and inductor,https://github.com/pytorch/pytorch/issues/173052,has linked PR,44,"open PR references this issue; component/oncall labels: module: decompositions, module: faketensor, module: pt2-dispatcher, oncall: pt2",review linked PR state before touching issue,2026-01-29T10:50:06Z
issue,173252,[dynamo] LazyConvTranspose1d fails under torch.compile(dynamic=True) with “SymIntArrayRef expected to contain only concrete integers”,https://github.com/pytorch/pytorch/issues/173252,has linked PR,44,"open PR references this issue; component/oncall labels: module: dynamo, oncall: pt2",review linked PR state before touching issue,2026-01-27T20:21:03Z
issue,173049,OOM error suggests expandable_segments even when enabled,https://github.com/pytorch/pytorch/issues/173049,has linked PR,44,"open PR references this issue; component/oncall labels: module: cuda, module: cudacachingallocator, module: error checking",review linked PR state before touching issue,2026-01-22T22:24:32Z
issue,172956,Mutation during backward on tensor saved for backward can create invalid graph,https://github.com/pytorch/pytorch/issues/172956,has linked PR,44,"open PR references this issue; component/oncall labels: module: pt2-dispatcher, oncall: pt2",review linked PR state before touching issue,2026-01-21T17:53:59Z
issue,167818,undefined symbol for `at::meta::_index_put_impl_` when running or compiling executable on my own torch-related project.,https://github.com/pytorch/pytorch/issues/167818,has linked PR,44,"open PR references this issue; component/oncall labels: module: binaries, module: privateuse1",review linked PR state before touching issue,2026-01-21T10:14:01Z
issue,172869,[opaque obj] Figure out subclassing OpaqueBase in pybinded classes,https://github.com/pytorch/pytorch/issues/172869,has linked PR,44,"open PR references this issue; component/oncall labels: module: custom-operators, module: pt2-dispatcher, oncall: pt2",review linked PR state before touching issue,2026-01-20T22:00:15Z
issue,172428,`torch.compile` on a `custom_op`'s fake when using `torch.func.vjp` crashes,https://github.com/pytorch/pytorch/issues/172428,has linked PR,44,"open PR references this issue; component/oncall labels: module: custom-operators, module: functorch, module: pt2-dispatcher, oncall: pt2",review linked PR state before touching issue,2026-01-18T04:02:12Z
issue,170509,Dynamo fails on unit tests: test/test_torch.py.test_geometric_kstest,https://github.com/pytorch/pytorch/issues/170509,has linked PR,44,"open PR references this issue; component/oncall labels: module: dynamo, oncall: pt2",review linked PR state before touching issue,2026-01-16T00:42:27Z
issue,172321,[OpenReg] Extend test_autograd.py coverage,https://github.com/pytorch/pytorch/issues/172321,has linked PR,44,"open PR references this issue; component/oncall labels: module: ci, module: openreg",review linked PR state before touching issue,2026-01-13T15:08:31Z
issue,119191,AOTAutograd: functionalization should emit foreach_copy_() instead of copy_() when possible,https://github.com/pytorch/pytorch/issues/119191,has linked PR,44,"open PR references this issue; label says issue is suitable for a scoped external contribution; old update age or low-urgency/stale label; component/oncall labels: module: aotdispatch, module: functionalization, module: mta, module: pt2-dispatcher",review linked PR state before touching issue,2025-11-19T11:27:31Z
issue,135011,"dynamo (re)compilation issues: shape (1,1), nn.Parameter, mark_dynamic",https://github.com/pytorch/pytorch/issues/135011,has linked PR,44,"open PR references this issue; label says issue is suitable for a scoped external contribution; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",review linked PR state before touching issue,2025-10-10T16:58:07Z
issue,3867,"[Feature Request] Implement ""same"" padding for convolution operations?",https://github.com/pytorch/pytorch/issues/3867,has linked PR,44,"open PR references this issue; needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: nn",route to design/research decision before patch planning,2025-09-29T22:55:35Z
issue,91692,[discussion] Analyzing a list of tensors stored as intermediate values / saved_for_backward in autograd graph,https://github.com/pytorch/pytorch/issues/91692,has linked PR,44,open PR references this issue; label says issue is suitable for a scoped external contribution; old update age or low-urgency/stale label; component/oncall labels: module: autograd,review linked PR state before touching issue,2025-07-29T10:19:02Z
issue,129020,Expand Tag Set: views & reductions,https://github.com/pytorch/pytorch/issues/129020,has linked PR,44,open PR references this issue; label says issue is suitable for a scoped external contribution; old update age or low-urgency/stale label; component/oncall labels: module: internals,review linked PR state before touching issue,2025-05-11T15:54:54Z
issue,111414,Minifier doesn't transfer execution states like @torch.no_grad to repro,https://github.com/pytorch/pytorch/issues/111414,has linked PR,44,"open PR references this issue; needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: minifier, oncall: pt2",route to design/research decision before patch planning,2024-07-11T02:16:10Z
issue,89489,build: cmake: need to uniformize installation of libraries in CMAKE_INSTALL_LIBDIR (not lib),https://github.com/pytorch/pytorch/issues/89489,has linked PR,44,"open PR references this issue; needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: build, module: cpp",route to design/research decision before patch planning,2022-11-22T17:54:43Z
issue,62027,`x.to(memory_format=torch.contiguous_format)` does not always return a contiguous tensor,https://github.com/pytorch/pytorch/issues/62027,has linked PR,44,"open PR references this issue; needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: docs, module: memory format",route to design/research decision before patch planning,2022-10-31T17:22:24Z
pr,189096,[aot_autograd] Make CPU-side stream syncs real cross-stream barriers,https://github.com/pytorch/pytorch/pull/189096,waiting on contributor,43,review decision requests changes; CI lane labels: ciflow/inductor,wait for contributor update,2026-07-09T02:18:21Z
pr,187940,[xpu][inductor] fix the fallback behavior of maxpool backward for performance,https://github.com/pytorch/pytorch/pull/187940,waiting on contributor,43,"review decision requests changes; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk, ciflow/xpu",wait for contributor update,2026-07-09T02:08:16Z
pr,189284,Add torch.autograd.graph.node_creation_hook,https://github.com/pytorch/pytorch/pull/189284,waiting on contributor,43,review decision requests changes,wait for contributor update,2026-07-09T02:04:54Z
pr,187600,[ROCm][Inductor] Enable NHWC convs by default on CDNA with an Inductor layout-opt gate,https://github.com/pytorch/pytorch/pull/187600,waiting on contributor,43,"review decision requests changes; CI lane labels: ciflow/docker, ciflow/inductor, ciflow/inductor-perf-test-nightly-rocm-mi300, ciflow/inductor-perf-test-nightly-rocm-mi350",wait for contributor update,2026-07-09T01:54:02Z
pr,189109,[profiler][cupti] Resolve a version-gated CUPTI header and conditionally install libclang for the field-id codegen,https://github.com/pytorch/pytorch/pull/189109,waiting on contributor,43,review decision requests changes,wait for contributor update,2026-07-09T01:52:57Z
pr,188742,[ROCm][inductor] Enable scaled grouped GEMM Triton lowering,https://github.com/pytorch/pytorch/pull/188742,waiting on contributor,43,"review decision requests changes; CI lane labels: ciflow/b200, ciflow/h100, ciflow/inductor, ciflow/rocm",wait for contributor update,2026-07-09T01:04:51Z
pr,188600,[ROCm][inductor] Support grouped GEMM Triton lowering,https://github.com/pytorch/pytorch/pull/188600,waiting on contributor,43,"review decision requests changes; CI lane labels: ciflow/b200, ciflow/h100, ciflow/inductor, ciflow/rocm-mi300",wait for contributor update,2026-07-08T23:51:01Z
pr,180826,"Remove deprecated `set_group_info`, `get_group_info` and `enable_symm_mem_for_group`",https://github.com/pytorch/pytorch/pull/180826,waiting on contributor,43,"review decision requests changes; CI lane labels: ciflow/b200-symm-mem, ciflow/h100-symm-mem, ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-07-08T22:50:12Z
pr,186965,[RFC][dynamo] Add ContextVar.get() tracing support (Phase 1),https://github.com/pytorch/pytorch/pull/186965,waiting on contributor,43,"review decision requests changes; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk",wait for contributor update,2026-07-08T22:27:06Z
pr,187450,[ROCm] periodic-rocm-mi350: run only the 4-GPU distributed tests,https://github.com/pytorch/pytorch/pull/187450,waiting on contributor,43,"review decision requests changes; CI lane labels: ciflow/periodic-rocm-mi350, ciflow/trunk",wait for contributor update,2026-07-08T21:52:15Z
pr,185617,[Profiler] Refactor test_trace_validator.py,https://github.com/pytorch/pytorch/pull/185617,waiting on contributor,43,review decision requests changes,wait for contributor update,2026-07-08T21:49:37Z
pr,188004,rewrite exception handling to be more closer to CPython,https://github.com/pytorch/pytorch/pull/188004,waiting on contributor,43,"review decision requests changes; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-07-08T21:06:03Z
pr,188954,[MPS] add matrix exp on MPS,https://github.com/pytorch/pytorch/pull/188954,waiting on contributor,43,review decision requests changes; CI lane labels: ciflow/mps,wait for contributor update,2026-07-08T20:50:52Z
pr,189207,Scope cuDNN Engine 5 errata to SM 8.6 GPUs only,https://github.com/pytorch/pytorch/pull/189207,waiting on contributor,43,review decision requests changes,wait for contributor update,2026-07-08T20:50:47Z
pr,183663,[inductor] _make_launchers: skip rebuilding existing launchers,https://github.com/pytorch/pytorch/pull/183663,waiting on contributor,43,"review decision requests changes; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-07-08T20:49:42Z
pr,179425,[inductor][autotuning] add logging and config for incremental autotuning (#179425),https://github.com/pytorch/pytorch/pull/179425,waiting on contributor,43,"review decision requests changes; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk",wait for contributor update,2026-07-08T20:41:11Z
pr,182278,[CI] Add native build image for linux-riscv64,https://github.com/pytorch/pytorch/pull/182278,waiting on contributor,43,review decision requests changes; CI lane labels: ciflow/docker,wait for contributor update,2026-07-08T20:36:32Z
pr,186540,[MPS] Add segment_reduce support with native Metal kernels,https://github.com/pytorch/pytorch/pull/186540,waiting on contributor,43,"review decision requests changes; CI lane labels: ciflow/inductor, ciflow/mps",wait for contributor update,2026-07-08T19:52:04Z
pr,180247,Migrate build system from setuptools to scikit-build-core,https://github.com/pytorch/pytorch/pull/180247,waiting on contributor,43,"review decision requests changes; CI lane labels: ciflow/binaries, ciflow/docker, ciflow/dynamo, ciflow/inductor",wait for contributor update,2026-07-08T19:51:02Z
pr,180248,Remove setup.py and setuptools build helpers,https://github.com/pytorch/pytorch/pull/180248,waiting on contributor,43,review decision requests changes; CI lane labels: ciflow/binaries,wait for contributor update,2026-07-08T19:12:47Z
pr,188105,Add support for custom attributes on Exceptions,https://github.com/pytorch/pytorch/pull/188105,waiting on contributor,43,"review decision requests changes; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-07-08T18:44:03Z
pr,178737,"Enable hipsparselt version, initCusparseltBindings and related UTs",https://github.com/pytorch/pytorch/pull/178737,waiting on contributor,43,"review decision requests changes; CI lane labels: ciflow/rocm-mi300, ciflow/rocm-mi355, ciflow/rocm-nightly, ciflow/trunk",wait for contributor update,2026-07-08T17:58:34Z
pr,185701,[Test] Make test_packed_sequence.py device agnostic for out-of-tree backends,https://github.com/pytorch/pytorch/pull/185701,waiting on contributor,43,review decision requests changes,wait for contributor update,2026-07-08T15:58:34Z
pr,189183,[inductor][mtia] Fix fp16 reduction accumulator dtype mismatch,https://github.com/pytorch/pytorch/pull/189183,waiting on contributor,43,"review decision requests changes; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-07-08T06:21:15Z
pr,188747,Python c10d Backend trampoline cleanup,https://github.com/pytorch/pytorch/pull/188747,waiting on contributor,43,review decision requests changes,wait for contributor update,2026-07-08T02:21:56Z
pr,188374,[HOP] Deduplicate lifted args for switch,https://github.com/pytorch/pytorch/pull/188374,waiting on contributor,43,"review decision requests changes; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-07-07T23:15:33Z
pr,184779,"[MPS] Rewrite _bincount_mps with native Metal atomic-add (~18 s → 2.1 ms, 8,500× speedup)",https://github.com/pytorch/pytorch/pull/184779,waiting on contributor,43,review decision requests changes; CI lane labels: ciflow/mps,wait for contributor update,2026-07-07T21:14:06Z
pr,188042,[CI] Split AI advisor garbage verdict into infra_issue + garbage,https://github.com/pytorch/pytorch/pull/188042,waiting on contributor,43,review decision requests changes,wait for contributor update,2026-07-07T20:50:42Z
pr,176678,[CI][Windows] Add NVIDIA RTX workflow,https://github.com/pytorch/pytorch/pull/176678,waiting on contributor,43,"review decision requests changes; CI lane labels: ciflow/inductor, ciflow/win-arm64",wait for contributor update,2026-07-07T20:11:55Z
pr,177540,[inductor] Handle Triton cluster_dims metadata from both binary and metadata layouts,https://github.com/pytorch/pytorch/pull/177540,waiting on contributor,43,"review decision requests changes; CI lane labels: ciflow/inductor, ciflow/trunk",wait for contributor update,2026-07-07T18:13:39Z
pr,188331,[TEST] Refactor test_profiler_tree.py with hw_classification,https://github.com/pytorch/pytorch/pull/188331,waiting on contributor,43,review decision requests changes,wait for contributor update,2026-07-07T18:05:28Z
pr,187828,Add privateuse1 backend support to has_triton(),https://github.com/pytorch/pytorch/pull/187828,waiting on contributor,43,review decision requests changes,wait for contributor update,2026-07-07T17:43:48Z
pr,185813,[ROCm][Windows] Audit license-files: explicit enumeration + audit test (#183434),https://github.com/pytorch/pytorch/pull/185813,waiting on contributor,43,review decision requests changes; CI lane labels: ciflow/docker,wait for contributor update,2026-07-07T17:15:10Z
pr,187434,Support autograd for Tensor?[] (optional tensor list) outputs,https://github.com/pytorch/pytorch/pull/187434,waiting on contributor,43,review decision requests changes; CI lane labels: ciflow/inductor,wait for contributor update,2026-07-07T14:52:16Z
pr,182725,[c10d] Throttle TCPStore connect-retry warnings,https://github.com/pytorch/pytorch/pull/182725,waiting on contributor,43,review decision requests changes,wait for contributor update,2026-07-07T06:06:22Z
pr,188095,[ROCm][FlexAttention] Default PRESCALE_QK and auto-detect contiguous blocks,https://github.com/pytorch/pytorch/pull/188095,waiting on contributor,43,"review decision requests changes; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-07-07T03:23:43Z
pr,187755,[testintro] CI: post tests added/removed as a PR comment,https://github.com/pytorch/pytorch/pull/187755,waiting on contributor,43,review decision requests changes,wait for contributor update,2026-07-06T20:30:01Z
pr,186324,[FakeTensor] Propagate backend metadata through fake tensor conversion,https://github.com/pytorch/pytorch/pull/186324,waiting on contributor,43,"review decision requests changes; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-07-06T20:11:57Z
pr,186047,[Test][PrivateUse1] Enhance `get_all_devices` to include all available devices,https://github.com/pytorch/pytorch/pull/186047,waiting on contributor,43,review decision requests changes,wait for contributor update,2026-07-06T08:38:02Z
pr,166396,[xpu][test][1/N] Enable tests of test_nn.py on Intel GPU - instantiate TestNN with instantiate_device_type_tests,https://github.com/pytorch/pytorch/pull/166396,waiting on contributor,43,"review decision requests changes; CI lane labels: ciflow/mps, ciflow/pull, ciflow/xpu",wait for contributor update,2026-07-05T15:52:59Z
pr,176619,Fix: raise RuntimeError when cudagraph partitions are empty,https://github.com/pytorch/pytorch/pull/176619,waiting on contributor,43,"review decision requests changes; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk",wait for contributor update,2026-07-05T05:09:03Z
pr,178849,Enable eager tests for inductor-unittest - - Enabel more tests in test_ops.py on Intel GPU,https://github.com/pytorch/pytorch/pull/178849,waiting on contributor,43,review decision requests changes; CI lane labels: ciflow/xpu,wait for contributor update,2026-07-03T21:38:20Z
pr,187654,[dynamo] Remove hardcoded CUDA references in test_logging.py,https://github.com/pytorch/pytorch/pull/187654,waiting on contributor,43,review decision requests changes; CI lane labels: ciflow/inductor,wait for contributor update,2026-07-03T16:14:17Z
pr,176420,[XPU] Ensure dense (canonical) strides for convolution inputs,https://github.com/pytorch/pytorch/pull/176420,waiting on contributor,43,review decision requests changes,wait for contributor update,2026-07-03T10:11:35Z
pr,169916,[CPU] Support multiple ncores-per-instance settings in xeon/run_cpu.py,https://github.com/pytorch/pytorch/pull/169916,waiting on contributor,43,review decision requests changes,wait for contributor update,2026-07-03T01:18:14Z
pr,172603,"Rewrite _get_total_norm with _powsum, add skip_root flag",https://github.com/pytorch/pytorch/pull/172603,waiting on contributor,43,review decision requests changes; CI lane labels: ciflow/inductor,wait for contributor update,2026-07-02T18:08:04Z
pr,187553,[MPS] Migrate mse_loss to a native fused Metal kernel,https://github.com/pytorch/pytorch/pull/187553,waiting on contributor,43,review decision requests changes; CI lane labels: ciflow/mps,wait for contributor update,2026-07-02T16:52:03Z
pr,188395,Refactor Dynamo side effect replay into registry dispatcher,https://github.com/pytorch/pytorch/pull/188395,waiting on contributor,43,"review decision requests changes; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-07-02T16:36:17Z
pr,188608,[ci] Re-enable cuDSS installation in CI Docker images so CUDA CI can actually build and run cuDSS tests,https://github.com/pytorch/pytorch/pull/188608,waiting on contributor,43,review decision requests changes; CI lane labels: ciflow/docker,wait for contributor update,2026-07-02T15:06:46Z
pr,179284,Fix AOT autograd leaf backward mutation for auto_functionalized V2 (#179284),https://github.com/pytorch/pytorch/pull/179284,waiting on contributor,43,"review decision requests changes; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-07-02T14:16:47Z
pr,178550,[DTensor] Support sharding propagation with meta device mesh,https://github.com/pytorch/pytorch/pull/178550,waiting on contributor,43,"review decision requests changes; CI lane labels: ciflow/dtensor, ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-07-02T14:16:41Z
pr,176747,Fix root cause of extraneous termination handler error logging (eval) (#176747),https://github.com/pytorch/pytorch/pull/176747,waiting on contributor,43,review decision requests changes,wait for contributor update,2026-07-02T14:16:36Z
pr,170752,Support torch.accelerator.get_device_capability on CUDA,https://github.com/pytorch/pytorch/pull/170752,waiting on contributor,43,review decision requests changes; CI lane labels: ciflow/trunk,wait for contributor update,2026-07-01T17:15:49Z
pr,187787,[MPS] Migrate var/std to native Metal kernels (Welford),https://github.com/pytorch/pytorch/pull/187787,waiting on contributor,43,"review decision requests changes; CI lane labels: ciflow/inductor, ciflow/mps",wait for contributor update,2026-07-01T16:44:40Z
pr,187573,[TMA] Fix AOTI eager scratch buffer missing CACHE_TORCH_DEVICE declaration,https://github.com/pytorch/pytorch/pull/187573,waiting on contributor,43,"review decision requests changes; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-07-01T11:59:50Z
pr,186652,"[ROCm] Fix ""Ignoring invalid value for boolean flag AMD_SERIALIZE_KERNEL: 3""",https://github.com/pytorch/pytorch/pull/186652,waiting on contributor,43,"review decision requests changes; CI lane labels: ciflow/rocm-mi300, ciflow/trunk",wait for contributor update,2026-07-01T03:02:37Z
pr,187976,[MPS] Fix int64 amax/amin returning 0 for partial-simdgroup reductions,https://github.com/pytorch/pytorch/pull/187976,waiting on contributor,43,review decision requests changes; CI lane labels: ciflow/mps,wait for contributor update,2026-06-30T23:23:03Z
pr,187772,perf(dynamo): add opt-in guard lookup token plan,https://github.com/pytorch/pytorch/pull/187772,waiting on contributor,43,"review decision requests changes; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-06-29T21:27:08Z
pr,187223,Add targeted SDPA coverage to H100 CI,https://github.com/pytorch/pytorch/pull/187223,waiting on contributor,43,review decision requests changes,wait for contributor update,2026-06-29T20:14:11Z
pr,188365,[MPS] Remove support for macOS 14,https://github.com/pytorch/pytorch/pull/188365,waiting on contributor,43,"review decision requests changes; CI lane labels: ciflow/inductor, ciflow/mps",wait for contributor update,2026-06-29T17:18:35Z
pr,171035,Add 2D Right-hand-side support for cuDSS backend `linalg.solve`,https://github.com/pytorch/pytorch/pull/171035,waiting on contributor,43,review decision requests changes,wait for contributor update,2026-06-29T13:17:25Z
pr,188015,[SDPA] Fix 32-bit dropout RNG offset overflow in mem-efficient attention,https://github.com/pytorch/pytorch/pull/188015,waiting on contributor,43,review decision requests changes,wait for contributor update,2026-06-28T23:05:33Z
pr,185015,Add SAC-driven fully overlapped activation CPU offload (primarily aimed at GH200/GB200/GB300 superchips),https://github.com/pytorch/pytorch/pull/185015,waiting on contributor,43,"review decision requests changes; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-06-25T21:56:20Z
pr,184739,Refactor test_sdpa for out-of-tree backends,https://github.com/pytorch/pytorch/pull/184739,waiting on contributor,43,"review decision requests changes; CI lane labels: ciflow/inductor, ciflow/trunk",wait for contributor update,2026-06-25T17:53:48Z
pr,185505,[inductor] Enable ROCm CK GEMM templates under JIT cpp_wrapper,https://github.com/pytorch/pytorch/pull/185505,waiting on contributor,43,"review decision requests changes; CI lane labels: ciflow/inductor, ciflow/inductor-rocm-mi300, ciflow/torchtitan",wait for contributor update,2026-06-25T13:55:58Z
pr,187747,Replace CUDA APIs with generic torch.accelerator equivalents,https://github.com/pytorch/pytorch/pull/187747,waiting on contributor,43,"review decision requests changes; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-06-25T03:48:22Z
pr,188086,Migrate torch_export_python to ShapesSpec API,https://github.com/pytorch/pytorch/pull/188086,waiting on contributor,43,"review decision requests changes; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-06-24T21:08:38Z
pr,187603,[executorch][dynamo] Import torch._utils before Dynamo variable registration,https://github.com/pytorch/pytorch/pull/187603,waiting on contributor,43,"review decision requests changes; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-06-24T19:15:01Z
pr,182984,[Inductor] Expose config for sm carveout of templates (#182984),https://github.com/pytorch/pytorch/pull/182984,waiting on contributor,43,"review decision requests changes; CI lane labels: ciflow/b200, ciflow/h100, ciflow/inductor, ciflow/rocm-mi300",wait for contributor update,2026-06-24T19:04:36Z
pr,187754,[testintro] Add a test-introspection tool,https://github.com/pytorch/pytorch/pull/187754,waiting on contributor,43,review decision requests changes; CI lane labels: ciflow/docker,wait for contributor update,2026-06-24T14:17:39Z
pr,185910,[AARCH64] Optimize AArch64 qlinear performance by triggering blocked layout in prepacked weight,https://github.com/pytorch/pytorch/pull/185910,waiting on contributor,43,review decision requests changes,wait for contributor update,2026-06-24T13:28:28Z
pr,181138,[Inductor][NFC] Refactor fusion legality checks for `UserDefinedTritonKernel`,https://github.com/pytorch/pytorch/pull/181138,waiting on contributor,43,"review decision requests changes; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-06-24T07:21:46Z
pr,182809,[2/N] Replace data_ptr with mutable_data_ptr,https://github.com/pytorch/pytorch/pull/182809,waiting on contributor,43,"review decision requests changes; CI lane labels: ciflow/b200, ciflow/h100, ciflow/rocm-mi300",wait for contributor update,2026-06-24T06:07:07Z
pr,187696,Add DataLoader worker signal message affixes,https://github.com/pytorch/pytorch/pull/187696,waiting on contributor,43,review decision requests changes,wait for contributor update,2026-06-24T02:59:29Z
pr,178463,[Distributed] Make Ckpt Tests Backend Agnostic,https://github.com/pytorch/pytorch/pull/178463,waiting on contributor,43,review decision requests changes; CI lane labels: ciflow/torchtitan,wait for contributor update,2026-06-23T19:14:39Z
pr,180080,Bitwise operator tensorcompare.cpp,https://github.com/pytorch/pytorch/pull/180080,waiting on contributor,43,review decision requests changes,wait for contributor update,2026-06-21T20:56:43Z
pr,173258,[CUDA] Math.cuh add missing return statement,https://github.com/pytorch/pytorch/pull/173258,waiting on contributor,43,review decision requests changes; CI lane labels: ciflow/trunk,wait for contributor update,2026-06-20T09:40:27Z
pr,174491,fix: prevent issue with destruction of std::shared_timed_mutex,https://github.com/pytorch/pytorch/pull/174491,waiting on contributor,43,review decision requests changes; CI lane labels: ciflow/trunk,wait for contributor update,2026-06-19T13:44:36Z
pr,182025,[MPS] Use MetalPerformancePrimitives for F.linear on macOS 26+,https://github.com/pytorch/pytorch/pull/182025,waiting on contributor,43,review decision requests changes; CI lane labels: ciflow/mps,wait for contributor update,2026-06-19T01:01:52Z
pr,187414,[Inductor][cond] Allow output-input aliasing in torch.cond branches (re-land D106106380 + dev-nosan test fix),https://github.com/pytorch/pytorch/pull/187414,waiting on contributor,43,"review decision requests changes; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-06-18T18:34:15Z
pr,179546,Add structured call_hierarchy metadata to FX nodes,https://github.com/pytorch/pytorch/pull/179546,waiting on contributor,43,review decision requests changes,wait for contributor update,2026-06-17T21:21:31Z
pr,179422,[inductor] Force contiguous inputs for custom ops in FallbackKernel,https://github.com/pytorch/pytorch/pull/179422,waiting on contributor,43,review decision requests changes,wait for contributor update,2026-06-17T21:19:45Z
pr,186104,[Inductor] Allow output-input aliasing in torch.cond branches (#186104),https://github.com/pytorch/pytorch/pull/186104,waiting on contributor,43,"review decision requests changes; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk",wait for contributor update,2026-06-17T21:07:05Z
pr,186950,Fixed issue with pickling of custom decomp dict,https://github.com/pytorch/pytorch/pull/186950,waiting on contributor,43,"review decision requests changes; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-06-17T19:18:49Z
pr,183673,[ROCm][Windows] torchgen: fix Windows DLL linkage for native and namespaced forward declarations,https://github.com/pytorch/pytorch/pull/183673,waiting on contributor,43,review decision requests changes,wait for contributor update,2026-06-17T14:34:13Z
pr,187492,Avoid std::views::reverse in CUDACachingAllocator,https://github.com/pytorch/pytorch/pull/187492,waiting on contributor,43,review decision requests changes,wait for contributor update,2026-06-17T01:16:14Z
pr,173519,Add on-demand ppc64le wheel build support,https://github.com/pytorch/pytorch/pull/173519,waiting on contributor,43,review decision requests changes,wait for contributor update,2026-06-16T13:06:41Z
pr,185741,[PyTorch][Profiler] Fix use-after-free in global RecordFunction callback teardown (#185741),https://github.com/pytorch/pytorch/pull/185741,waiting on contributor,43,review decision requests changes; CI lane labels: ciflow/trunk,wait for contributor update,2026-06-15T23:19:55Z
pr,180129,[scaled_grouped_mm] add support for out argument,https://github.com/pytorch/pytorch/pull/180129,waiting on contributor,43,"review decision requests changes; CI lane labels: ciflow/b200, ciflow/h100, ciflow/rocm-mi300",wait for contributor update,2026-06-15T22:07:32Z
pr,184654,Short-circuit pytree.tree_map_only when no leaves match,https://github.com/pytorch/pytorch/pull/184654,waiting on contributor,43,"review decision requests changes; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-06-15T16:34:26Z
pr,186642,[inductor][rocm] make AMD MM matrix_instr_nonkdim configurable,https://github.com/pytorch/pytorch/pull/186642,waiting on contributor,43,"review decision requests changes; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-06-15T13:09:55Z
pr,183669,[inductor] Make experimental benchmarker parameters configurable via env vars,https://github.com/pytorch/pytorch/pull/183669,waiting on contributor,43,"review decision requests changes; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-06-11T17:02:53Z
pr,183556,[inductor] Fix MemoryTracker assertion with shared storages in reordered schedules,https://github.com/pytorch/pytorch/pull/183556,waiting on contributor,43,"review decision requests changes; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-06-11T16:59:49Z
pr,178048,"[inductor] Add custom_pass_context for scoped, composable custom pass registration",https://github.com/pytorch/pytorch/pull/178048,waiting on contributor,43,review decision requests changes; CI lane labels: ciflow/inductor,wait for contributor update,2026-06-11T16:26:19Z
pr,182233,Fix torch.combinations dynamic shape support by replacing numel with sym_numel and propagating SymInt,https://github.com/pytorch/pytorch/pull/182233,waiting on contributor,43,"review decision requests changes; CI lane labels: ciflow/dtensor, ciflow/inductor, ciflow/trunk",wait for contributor update,2026-06-11T10:59:22Z
pr,184895,[Inductor] Index expression codegen for user Triton kernel fusion,https://github.com/pytorch/pytorch/pull/184895,waiting on contributor,43,"review decision requests changes; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-06-11T05:29:28Z
pr,186808,[inductor] Re-key range_tree_nodes on IterationRangesEntry rename,https://github.com/pytorch/pytorch/pull/186808,waiting on contributor,43,"review decision requests changes; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-06-10T14:44:27Z
pr,182712,[MPS] Enhance Linear tensor ops to avoid unnecessary .contiguous() calls on weight tensor,https://github.com/pytorch/pytorch/pull/182712,waiting on contributor,43,review decision requests changes; CI lane labels: ciflow/mps,wait for contributor update,2026-06-09T23:57:34Z
pr,186138,Support data-dependent 0-dim tensor indexing under tracing,https://github.com/pytorch/pytorch/pull/186138,waiting on contributor,43,review decision requests changes,wait for contributor update,2026-06-09T23:08:41Z
pr,186251,Wait for Docker images before starting builds,https://github.com/pytorch/pytorch/pull/186251,waiting on contributor,43,review decision requests changes; CI lane labels: ciflow/trunk,wait for contributor update,2026-06-09T00:18:59Z
pr,182208,Add missing stacklevel to 57 warnings.warn() calls in torch/,https://github.com/pytorch/pytorch/pull/182208,waiting on contributor,43,"review decision requests changes; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-06-03T21:42:10Z
pr,172945,Bf16 optimized heuristics,https://github.com/pytorch/pytorch/pull/172945,waiting on contributor,43,review decision requests changes,wait for contributor update,2026-06-03T15:44:33Z
pr,185475,[pytorch] Nested custom Python process groups (experimental) (#185475),https://github.com/pytorch/pytorch/pull/185475,waiting on contributor,43,review decision requests changes; CI lane labels: ciflow/torchtitan,wait for contributor update,2026-06-03T15:25:40Z
pr,181730,"Prevent DCE eliminate ""triton_kernel_wrapper_mutation""",https://github.com/pytorch/pytorch/pull/181730,waiting on contributor,43,review decision requests changes,wait for contributor update,2026-06-02T22:47:07Z
pr,169694,Increase precision for golden solution in transformer tests.,https://github.com/pytorch/pytorch/pull/169694,waiting on contributor,43,"review decision requests changes; CI lane labels: ciflow/rocm-mi300, ciflow/rocm-mi355",wait for contributor update,2026-06-02T20:35:18Z
pr,184533,Defer AOTAutograd version-check error for forward-only graph,https://github.com/pytorch/pytorch/pull/184533,waiting on contributor,43,review decision requests changes; CI lane labels: ciflow/inductor,wait for contributor update,2026-06-02T00:24:56Z
pr,184335,[spmd_types] pipeline parallel support for spmd_types-annotated tensors,https://github.com/pytorch/pytorch/pull/184335,waiting on contributor,43,"review decision requests changes; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-05-29T18:14:50Z
pr,185407,[RFC][dynamo] Narrow GB6297 to only graph-break on differentiable duplicate tensor inputs,https://github.com/pytorch/pytorch/pull/185407,waiting on contributor,43,"review decision requests changes; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-05-28T15:19:54Z
pr,184141,Remove pointer_to_optional_list,https://github.com/pytorch/pytorch/pull/184141,waiting on contributor,43,review decision requests changes; CI lane labels: ciflow/inductor,wait for contributor update,2026-05-28T03:00:31Z
pr,185418,[coor] Represent get_rank as a SymInt,https://github.com/pytorch/pytorch/pull/185418,waiting on contributor,43,"review decision requests changes; CI lane labels: ciflow/dtensor, ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-05-27T23:42:33Z
pr,178204,[PyTorch][OSS][AOTAutograd] Allow None values in saved tensors for vc_check split (#178204),https://github.com/pytorch/pytorch/pull/178204,waiting on contributor,43,"review decision requests changes; CI lane labels: ciflow/inductor, ciflow/trunk",wait for contributor update,2026-05-26T23:23:31Z
pr,182576,Improved static Type Hints for dataloader,https://github.com/pytorch/pytorch/pull/182576,waiting on contributor,43,review decision requests changes,wait for contributor update,2026-05-25T15:01:16Z
pr,185011,[inductor] Pair maxpool value and offset lowering,https://github.com/pytorch/pytorch/pull/185011,draft/noise,43,"PR is draft; failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-05-24T06:14:38Z
pr,184930,[inductor] Score broadcast reuse for expensive pointwise tiling,https://github.com/pytorch/pytorch/pull/184930,draft/noise,43,"PR is draft; failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-05-24T03:28:29Z
pr,184999,[inductor] Avoid pointwise cat for complex output branches,https://github.com/pytorch/pytorch/pull/184999,draft/noise,43,"PR is draft; failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-05-24T01:26:01Z
pr,185012,[inductor] Add FP64-heavy pointwise autotune candidate,https://github.com/pytorch/pytorch/pull/185012,draft/noise,43,"PR is draft; failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-05-24T01:25:36Z
pr,184949,Remove allow-untyped-defs from pyi files,https://github.com/pytorch/pytorch/pull/184949,draft/noise,43,PR is draft; failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/torchtitan,"identify whether block is CI, merge conflict, or review gate",2026-05-23T17:18:49Z
pr,184933,[inductor] Fold sparse one-hot sum reductions,https://github.com/pytorch/pytorch/pull/184933,draft/noise,43,"PR is draft; failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-05-23T16:46:11Z
pr,184962,[ROCm] Drop version cap on bf16/fp16 mm/bmm accuracy skip,https://github.com/pytorch/pytorch/pull/184962,draft/noise,43,"PR is draft; failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/b200, ciflow/h100, ciflow/rocm-mi300","identify whether block is CI, merge conflict, or review gate",2026-05-23T09:13:26Z
pr,184937,[inductor] Prototype channel-resident reduction suffix codegen,https://github.com/pytorch/pytorch/pull/184937,draft/noise,43,"PR is draft; failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-05-23T02:58:55Z
pr,184873,Scope third-party install rules out of the wheel via CMake components,https://github.com/pytorch/pytorch/pull/184873,draft/noise,43,PR is draft; failed checks or blocked/unstable merge state; PR has not updated for at least 45 days,"identify whether block is CI, merge conflict, or review gate",2026-05-22T18:52:43Z
pr,180904,Enable MyST Markdown docstrings in Sphinx docs,https://github.com/pytorch/pytorch/pull/180904,draft/noise,43,PR is draft; failed checks or blocked/unstable merge state; PR has not updated for at least 45 days,"identify whether block is CI, merge conflict, or review gate",2026-05-20T12:46:39Z
pr,183925,Covert notes/cuda from .rst to .md,https://github.com/pytorch/pytorch/pull/183925,draft/noise,43,PR is draft; failed checks or blocked/unstable merge state; PR has not updated for at least 45 days,"identify whether block is CI, merge conflict, or review gate",2026-05-20T12:46:29Z
pr,178628,[DRAFT]Add TorchInductor documentation to User Guide,https://github.com/pytorch/pytorch/pull/178628,draft/noise,43,PR is draft; failed checks or blocked/unstable merge state; PR has not updated for at least 45 days,"identify whether block is CI, merge conflict, or review gate",2026-05-20T12:16:49Z
pr,184464,Cache PyCompilerInterface method lookups during graph capture,https://github.com/pytorch/pytorch/pull/184464,draft/noise,43,PR is draft; failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2026-05-20T11:05:43Z
pr,184440,[dynamic_spec]Add optimizer to ease user's dynamic_spec adoption,https://github.com/pytorch/pytorch/pull/184440,draft/noise,43,"PR is draft; failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-05-20T03:01:16Z
pr,182921,[DO NOT MERGE][TESTING ONLY] Set MI300 ROCm build arch to SPIR-V,https://github.com/pytorch/pytorch/pull/182921,draft/noise,43,PR is draft; failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/rocm-mi300,"identify whether block is CI, merge conflict, or review gate",2026-05-19T20:45:28Z
pr,177605,Default to fused implementation when available,https://github.com/pytorch/pytorch/pull/177605,draft/noise,43,PR is draft; failed checks or blocked/unstable merge state; PR has not updated for at least 45 days,"identify whether block is CI, merge conflict, or review gate",2026-05-19T15:43:47Z
pr,178412,Move steps to CPU for fused optims,https://github.com/pytorch/pytorch/pull/178412,draft/noise,43,PR is draft; failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/mps,"identify whether block is CI, merge conflict, or review gate",2026-05-19T15:37:43Z
pr,184308,Route privateuse1-overrideable ops through AOTI dispatcher (,https://github.com/pytorch/pytorch/pull/184308,draft/noise,43,"PR is draft; failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-05-19T06:20:42Z
pr,183798,[ROCm] Fix HIP stream pool aliasing when GPU tracing is active,https://github.com/pytorch/pytorch/pull/183798,draft/noise,43,PR is draft; failed checks or blocked/unstable merge state; PR has not updated for at least 45 days,"identify whether block is CI, merge conflict, or review gate",2026-05-18T19:50:37Z
pr,183479,TMA scatter_add: add multi-row-per-warp path for small rows,https://github.com/pytorch/pytorch/pull/183479,draft/noise,43,PR is draft; failed checks or blocked/unstable merge state; PR has not updated for at least 45 days,"identify whether block is CI, merge conflict, or review gate",2026-05-16T21:13:13Z
pr,183966,[CUDA] Change argmax to int64_t for better numerics,https://github.com/pytorch/pytorch/pull/183966,draft/noise,43,PR is draft; failed checks or blocked/unstable merge state; PR has not updated for at least 45 days,"identify whether block is CI, merge conflict, or review gate",2026-05-16T19:37:12Z
pr,177705,[Do Not Land] Lookup Table Recorder,https://github.com/pytorch/pytorch/pull/177705,draft/noise,43,"PR is draft; failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-05-16T01:01:56Z
pr,179011,[Inductor][NVGEMM] Add fast path for GEMM dispatch,https://github.com/pytorch/pytorch/pull/179011,draft/noise,43,"PR is draft; failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-05-16T01:01:55Z
pr,179012,[Inductor][NVGEMM] Add supplement heuristic configs behind flag,https://github.com/pytorch/pytorch/pull/179012,draft/noise,43,"PR is draft; failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-05-16T01:01:54Z
pr,178840,Add parallel subprocess compilation and disk caching for NVGEMM backend,https://github.com/pytorch/pytorch/pull/178840,draft/noise,43,"PR is draft; failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-05-16T01:01:54Z
pr,179167,[Inductor][NVGEMM] Fix FP4 autotuning tensor creation,https://github.com/pytorch/pytorch/pull/179167,draft/noise,43,"PR is draft; failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-05-16T01:01:51Z
pr,179014,[Inductor][NVGEMM] Wrap GroupedGemmArguments in try/except,https://github.com/pytorch/pytorch/pull/179014,draft/noise,43,"PR is draft; failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-05-16T01:01:51Z
pr,179013,[Inductor][NVGEMM] Fix FP4 autotuning tensor creation,https://github.com/pytorch/pytorch/pull/179013,draft/noise,43,"PR is draft; failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-05-16T01:01:51Z
pr,179015,[Inductor][NVGEMM] Add fast path for GEMM dispatch,https://github.com/pytorch/pytorch/pull/179015,draft/noise,43,"PR is draft; failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-05-16T01:01:50Z
pr,179016,[Inductor][NVGEMM] Add supplement heuristic configs behind flag,https://github.com/pytorch/pytorch/pull/179016,draft/noise,43,"PR is draft; failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-05-16T01:01:49Z
pr,179049,[Inductor][NVGEMM] Add supplement heuristic configs behind flag,https://github.com/pytorch/pytorch/pull/179049,draft/noise,43,"PR is draft; failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-05-16T01:01:48Z
pr,179175,[Inductor][NVGEMM] Fix FP4 autotuning tensor creation,https://github.com/pytorch/pytorch/pull/179175,draft/noise,43,"PR is draft; failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-05-16T01:01:46Z
pr,183797,fix internal tests for [BE][Ez]: Make more dataclass slots=True (#183449),https://github.com/pytorch/pytorch/pull/183797,draft/noise,43,PR is draft; failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2026-05-15T04:28:17Z
pr,166063,Add Gluon based grouped MM kernel for Blackwell into Inductor,https://github.com/pytorch/pytorch/pull/166063,draft/noise,43,"PR is draft; failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/b200, ciflow/h100, ciflow/inductor, ciflow/rocm-mi300","identify whether block is CI, merge conflict, or review gate",2026-05-15T03:00:43Z
pr,180891,[ROCm] Migrate from rocm_smi to amd_smi,https://github.com/pytorch/pytorch/pull/180891,draft/noise,43,"PR is draft; failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/rocm-mi300, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-05-14T16:21:13Z
pr,178354,[XPU] Add XPU dispatch for _scaled_grouped_mm,https://github.com/pytorch/pytorch/pull/178354,draft/noise,43,"PR is draft; failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/trunk, ciflow/xpu","identify whether block is CI, merge conflict, or review gate",2026-05-14T12:16:43Z
pr,183612,"Update NVPL versions: blas 0.3.0 → 0.5.0.1, lapack 0.2.3.1 → 0.3.2",https://github.com/pytorch/pytorch/pull/183612,draft/noise,43,PR is draft; failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/binaries,"identify whether block is CI, merge conflict, or review gate",2026-05-14T05:59:26Z
pr,183374,"[ROCm] [ TEST , PLEASE DON""T MERGE ] TheRock 7.13 nightly deb install path",https://github.com/pytorch/pytorch/pull/183374,draft/noise,43,PR is draft; failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/op-benchmark,"identify whether block is CI, merge conflict, or review gate",2026-05-14T05:58:28Z
pr,177690,"DO NOT MERGE Default pre_grad_pass_timing to ""late"" in fbcode",https://github.com/pytorch/pytorch/pull/177690,draft/noise,43,"PR is draft; failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-05-13T22:28:58Z
pr,181306,Integrate profiler documentation into torch.profiler docs,https://github.com/pytorch/pytorch/pull/181306,draft/noise,43,PR is draft; failed checks or blocked/unstable merge state; PR has not updated for at least 45 days,"identify whether block is CI, merge conflict, or review gate",2026-05-13T21:16:39Z
pr,183456,[ROCm] [opbench] Temporary profiler workaround for rocprofiler crash (#182719),https://github.com/pytorch/pytorch/pull/183456,draft/noise,43,PR is draft; failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/op-benchmark,"identify whether block is CI, merge conflict, or review gate",2026-05-13T11:54:00Z
pr,183361,Prototype lazy module backward hook compile,https://github.com/pytorch/pytorch/pull/183361,draft/noise,43,"PR is draft; failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-05-13T02:15:55Z
pr,183452,[BE]: Update cutlass submodule to 4.5.0rc,https://github.com/pytorch/pytorch/pull/183452,draft/noise,43,PR is draft; failed checks or blocked/unstable merge state; PR has not updated for at least 45 days,"identify whether block is CI, merge conflict, or review gate",2026-05-12T21:53:53Z
pr,180874,Add CPU offload support for selective activation checkpointing,https://github.com/pytorch/pytorch/pull/180874,draft/noise,43,PR is draft; failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/torchtitan,"identify whether block is CI, merge conflict, or review gate",2026-05-11T20:06:23Z
pr,183264,[WIP][Dynamo] Add mark_batch_invariant API,https://github.com/pytorch/pytorch/pull/183264,draft/noise,43,"PR is draft; failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-05-11T14:19:47Z
pr,181089,[Inductor] Disable split-reduction lowering under batch_invariant,https://github.com/pytorch/pytorch/pull/181089,draft/noise,43,"PR is draft; failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-05-11T14:19:44Z
pr,182779,[WIP] Test playground,https://github.com/pytorch/pytorch/pull/182779,draft/noise,43,"PR is draft; failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/dtensor, ciflow/h100-symm-mem, ciflow/inductor, ciflow/inductor-pallas","identify whether block is CI, merge conflict, or review gate",2026-05-11T06:05:27Z
pr,183054,coverage-td plan,https://github.com/pytorch/pytorch/pull/183054,draft/noise,43,PR is draft; failed checks or blocked/unstable merge state; PR has not updated for at least 45 days,"identify whether block is CI, merge conflict, or review gate",2026-05-10T14:28:20Z
pr,130856,[pipelining] lazy shape inference for stage,https://github.com/pytorch/pytorch/pull/130856,draft/noise,43,PR is draft; failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/torchtitan,"identify whether block is CI, merge conflict, or review gate",2026-03-21T22:38:50Z
pr,172109,stage executor,https://github.com/pytorch/pytorch/pull/172109,draft/noise,43,PR is draft; failed checks or blocked/unstable merge state; PR has not updated for at least 45 days,"identify whether block is CI, merge conflict, or review gate",2026-03-20T22:32:07Z
pr,174864,[XPU] Fix accuracy issue in addmm for bf16/f16 dtypes,https://github.com/pytorch/pytorch/pull/174864,draft/noise,41,PR is draft; review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/xpu,wait for contributor update,2026-07-03T07:46:44Z
pr,133419,"Keep is_mkldnn, add is_onednn",https://github.com/pytorch/pytorch/pull/133419,draft/noise,41,"PR is draft; review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/linux-aarch64, ciflow/torchtitan, ciflow/trunk",wait for contributor update,2026-06-18T12:58:26Z
pr,133392,Update DispatchKey and namespace,https://github.com/pytorch/pytorch/pull/133392,draft/noise,41,"PR is draft; review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/linux-aarch64, ciflow/torchtitan, ciflow/trunk",wait for contributor update,2026-06-18T12:58:24Z
pr,133296,Update file name with onednn,https://github.com/pytorch/pytorch/pull/133296,draft/noise,41,"PR is draft; review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk",wait for contributor update,2026-06-18T12:58:20Z
pr,133315,Update non user APIs,https://github.com/pytorch/pytorch/pull/133315,draft/noise,41,"PR is draft; review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/linux-aarch64, ciflow/torchtitan, ciflow/trunk",wait for contributor update,2026-06-18T12:58:17Z
issue,189014,[RFC]: Generalize memory snapshot APIs under `torch.accelerator.memory`,https://github.com/pytorch/pytorch/issues/189014,needs design decision,38,needs design/research label present; component/oncall labels: module: accelerator,route to design/research decision before patch planning,2026-07-06T07:55:00Z
issue,114403,refactor TracingContext to take a more limited subset of ViewAndMutationMeta,https://github.com/pytorch/pytorch/issues/114403,good volunteer slice,38,"label says issue is suitable for a scoped external contribution; component/oncall labels: module: aotdispatch, module: pt2-dispatcher, oncall: pt2",scan when time allows,2026-07-03T09:11:23Z
issue,102207,skipIfTorchInductor Tracking Issue,https://github.com/pytorch/pytorch/issues/102207,good volunteer slice,38,"label says issue is suitable for a scoped external contribution; component/oncall labels: module: inductor, oncall: pt2",scan when time allows,2026-07-01T23:50:03Z
issue,99149,"Spectral Normalization can not be applied to Conv{1,2,3}d",https://github.com/pytorch/pytorch/issues/99149,needs design decision,38,"needs design/research label present; component/oncall labels: module: convolution, module: nn, module: norms and normalization",route to design/research decision before patch planning,2026-07-01T12:27:04Z
issue,106571,Enable more flake8-bugbear lints,https://github.com/pytorch/pytorch/issues/106571,good volunteer slice,38,label says issue is suitable for a scoped external contribution; component/oncall labels: module: lint,scan when time allows,2026-07-01T09:53:28Z
issue,9886,[doc] functionalities not documented,https://github.com/pytorch/pytorch/issues/9886,good volunteer slice,38,label says issue is suitable for a scoped external contribution; component/oncall labels: module: docs,scan when time allows,2026-06-30T08:53:16Z
issue,158968,foreach_map enhancements,https://github.com/pytorch/pytorch/issues/158968,good volunteer slice,38,"label says issue is suitable for a scoped external contribution; component/oncall labels: module: dynamo, module: inductor, oncall: pt2",scan when time allows,2026-06-27T21:11:51Z
issue,188216,Feature request: built-in loss extrapolator,https://github.com/pytorch/pytorch/issues/188216,needs design decision,38,needs design/research label present; component/oncall labels: module: loss,route to design/research decision before patch planning,2026-06-26T00:50:41Z
issue,122192,extra checks for reduce-overhead inconsistent when warmup/recording,https://github.com/pytorch/pytorch/issues/122192,needs design decision,38,"needs design/research label present; component/oncall labels: module: cuda graphs, module: inductor, oncall: pt2",route to design/research decision before patch planning,2026-06-23T14:21:46Z
issue,170839,Varlen attention API with custom varlen `attn_mask`,https://github.com/pytorch/pytorch/issues/170839,needs design decision,38,"needs design/research label present; component/oncall labels: module: flex attention, module: nn, module: sdpa, oncall: pt2",route to design/research decision before patch planning,2026-06-23T04:55:20Z
issue,176837,FlexAttention: extend AuxRequest to return min_scores and key indices for row-wise max/min,https://github.com/pytorch/pytorch/issues/176837,needs design decision,38,"needs design/research label present; component/oncall labels: module: flex attention, oncall: pt2",route to design/research decision before patch planning,2026-06-23T04:54:39Z
issue,178586,`make_graphed_callables` silently drops in-place mutations on user-provided input tensors,https://github.com/pytorch/pytorch/issues/178586,needs design decision,38,"needs design/research label present; component/oncall labels: module: correctness (silent), module: cuda, module: cuda graphs, oncall: pt2",route to design/research decision before patch planning,2026-06-18T16:13:13Z
issue,160230,Offer official Pytorch Vulkan backend on pytorch.org,https://github.com/pytorch/pytorch/issues/160230,needs design decision,38,needs design/research label present; component/oncall labels: module: vulkan,route to design/research decision before patch planning,2026-06-16T02:56:13Z
issue,50688,[feature request] `torch.scan` (also port `lax.fori_loop` / `lax.while_loop` / `lax.associative_scan` and hopefully parallelized associative scans),https://github.com/pytorch/pytorch/issues/50688,needs design decision,38,"needs design/research label present; component/oncall labels: module: functional ux, module: functorch, module: numpy, module: pt2-dispatcher",route to design/research decision before patch planning,2026-06-09T18:43:18Z
issue,136312,Fault tolerance with k8s,https://github.com/pytorch/pytorch/issues/136312,needs design decision,38,"needs design/research label present; component/oncall labels: oncall: distributed, oncall: distributed infra",route to design/research decision before patch planning,2026-06-05T17:02:45Z
issue,152323,compile generates inefficient code when mutating small slice of a graph input,https://github.com/pytorch/pytorch/issues/152323,needs design decision,38,"needs design/research label present; component/oncall labels: module: functionalization, module: inductor, module: pt2-dispatcher, oncall: pt2",route to design/research decision before patch planning,2026-06-04T20:21:56Z
issue,183434,Audit license-files glob in pyproject.toml for over-collection,https://github.com/pytorch/pytorch/issues/183434,needs design decision,38,"needs design/research label present; component/oncall labels: module: binaries, module: third_party",route to design/research decision before patch planning,2026-06-01T13:07:22Z
issue,184517,LRScheduler doesn't use _get_closed_form_lr when initalizing scheduler,https://github.com/pytorch/pytorch/issues/184517,needs design decision,38,needs design/research label present; component/oncall labels: module: lrscheduler,route to design/research decision before patch planning,2026-05-28T23:10:56Z
issue,158657,extend torch.util.checkpoint to support offload to cpu/nvme,https://github.com/pytorch/pytorch/issues/158657,needs design decision,38,"needs design/research label present; component/oncall labels: module: activation checkpointing, module: autograd",route to design/research decision before patch planning,2026-05-24T00:42:42Z
issue,141452,[Dynamo] Keep weak reference of parameters instead of accessing them by attributes,https://github.com/pytorch/pytorch/issues/141452,needs design decision,38,"needs design/research label present; component/oncall labels: module: dynamo, module: vllm, oncall: pt2",route to design/research decision before patch planning,2026-05-21T12:33:44Z
issue,180845,[Dynamo] Unsupported Tensor.item() call in list comprehension when capture_scalar_outputs=False,https://github.com/pytorch/pytorch/issues/180845,needs design decision,38,"needs design/research label present; component/oncall labels: module: compile ux, module: dynamo, oncall: pt2",route to design/research decision before patch planning,2026-05-19T15:48:14Z
issue,180107,Allow getting `AtenTensorHandle` from the tensor object via AOTI stable C API,https://github.com/pytorch/pytorch/issues/180107,needs design decision,38,"needs design/research label present; component/oncall labels: module: aotinductor, module: cpp, oncall: pt2",route to design/research decision before patch planning,2026-05-19T00:55:49Z
issue,151142,AOT Dispatcher converts a single `detach` call to multiple `aten.detach`,https://github.com/pytorch/pytorch/issues/151142,needs design decision,38,"needs design/research label present; component/oncall labels: module: aotdispatch, module: pt2-dispatcher, oncall: pt2",route to design/research decision before patch planning,2026-05-15T03:07:02Z
issue,146804,"`torch.library.register_fake` respects only positional order, but not kwargs order",https://github.com/pytorch/pytorch/issues/146804,needs design decision,38,"needs design/research label present; component/oncall labels: module: library, module: pt2-dispatcher, oncall: pt2",route to design/research decision before patch planning,2026-05-15T03:06:38Z
issue,149258,Auto-selective activation checkpointing is not optimal for speed (issue with min_cut_rematerialization_partition),https://github.com/pytorch/pytorch/issues/149258,needs design decision,38,"needs design/research label present; component/oncall labels: module: pt2-dispatcher, oncall: pt2",route to design/research decision before patch planning,2026-05-15T02:04:06Z
issue,140835,auto-grad graph replicate split_with_sizes(lengths) X times where X = len(lengths) effecting compile time,https://github.com/pytorch/pytorch/issues/140835,needs design decision,38,"needs design/research label present; component/oncall labels: module: functionalization, module: pt2-dispatcher, oncall: pt2",route to design/research decision before patch planning,2026-05-14T22:22:08Z
issue,132651,fix unnecessary re-trace in AOTDispatcher in subclass code path,https://github.com/pytorch/pytorch/issues/132651,needs design decision,38,"needs design/research label present; component/oncall labels: module: __torch_dispatch__, module: aotdispatch, module: compile-time, module: pt2-dispatcher",route to design/research decision before patch planning,2026-05-14T22:21:58Z
issue,113022,Operator with only `Tensor[][]` args unsupported by dispatcher,https://github.com/pytorch/pytorch/issues/113022,needs design decision,38,"needs design/research label present; component/oncall labels: module: custom-operators, module: dispatch, module: pt2-dispatcher, oncall: pt2",route to design/research decision before patch planning,2026-05-14T22:20:21Z
issue,149559,Improve handling for custom ops with no returns,https://github.com/pytorch/pytorch/issues/149559,needs design decision,38,"needs design/research label present; component/oncall labels: module: custom-operators, module: pt2-dispatcher, oncall: pt2",route to design/research decision before patch planning,2026-05-14T22:20:10Z
issue,139444,registering op through function does not work,https://github.com/pytorch/pytorch/issues/139444,needs design decision,38,"needs design/research label present; component/oncall labels: module: custom-operators, module: docs, module: pt2-dispatcher, oncall: pt2",route to design/research decision before patch planning,2026-05-14T22:19:57Z
issue,158827,[RFC] Graph generalization,https://github.com/pytorch/pytorch/issues/158827,needs design decision,38,"needs design/research label present; component/oncall labels: module: accelerator, module: privateuse1",route to design/research decision before patch planning,2026-05-12T02:21:44Z
issue,148077,Feature Request: CUDA-Optimized Queue Buffer for PyTorch,https://github.com/pytorch/pytorch/issues/148077,needs design decision,38,needs design/research label present; component/oncall labels: module: cuda,route to design/research decision before patch planning,2026-05-06T20:41:02Z
issue,164469,[RFC] Adopt Spin as developer cli,https://github.com/pytorch/pytorch/issues/164469,needs design decision,38,needs design/research label present; component/oncall labels: module: build,route to design/research decision before patch planning,2026-04-30T09:55:49Z
issue,146414,MX basic dtypes in pytorch/pytorch,https://github.com/pytorch/pytorch/issues/146414,needs design decision,38,needs design/research label present; component/oncall labels: module: python frontend,route to design/research decision before patch planning,2026-04-27T01:44:32Z
issue,180215,[RFC] Best practices for backend installations requiring +cpu / non-PyPI wheels,https://github.com/pytorch/pytorch/issues/180215,needs design decision,38,"needs design/research label present; component/oncall labels: module: binaries, module: privateuse1",route to design/research decision before patch planning,2026-04-13T17:33:41Z
issue,120376,Stochastic rounding in bfloat16,https://github.com/pytorch/pytorch/issues/120376,needs design decision,38,needs design/research label present; component/oncall labels: module: bfloat16,route to design/research decision before patch planning,2026-04-06T06:50:45Z
issue,77171,Allow users to express fused matmul/bias/relu,https://github.com/pytorch/pytorch/issues/77171,needs design decision,38,needs design/research label present; component/oncall labels: module: python frontend,route to design/research decision before patch planning,2026-03-28T19:02:17Z
issue,41625,[feature request] New function `torch.slice(...)` mirroring TorchScript op signature or add step argument to `torch.narrow`,https://github.com/pytorch/pytorch/issues/41625,needs design decision,38,needs design/research label present,route to design/research decision before patch planning,2026-03-28T08:51:48Z
issue,173382,Make torch.cuda.empty_cache() actually empty the GPU cache,https://github.com/pytorch/pytorch/issues/173382,needs design decision,38,"needs design/research label present; component/oncall labels: module: cuda, module: cudacachingallocator",route to design/research decision before patch planning,2026-03-25T16:54:02Z
issue,158346,Compatibility with C++20 `std::format`,https://github.com/pytorch/pytorch/issues/158346,needs design decision,38,needs design/research label present; component/oncall labels: module: build,route to design/research decision before patch planning,2026-03-24T18:05:16Z
issue,177623,add support for Predictive Coding as alternative to Backpropagation,https://github.com/pytorch/pytorch/issues/177623,needs design decision,38,"needs design/research label present; component/oncall labels: module: autograd, module: optimizer",route to design/research decision before patch planning,2026-03-17T07:06:03Z
issue,177537,[RFC] nn.BlockAttentionResidual: depth-wise softmax attention over layer outputs as a drop-in residual replacement,https://github.com/pytorch/pytorch/issues/177537,needs design decision,38,needs design/research label present; component/oncall labels: module: nn,route to design/research decision before patch planning,2026-03-16T19:22:08Z
issue,47117,ctx.save_for_backward doesn't save torch.Tensor subclasses fully,https://github.com/pytorch/pytorch/issues/47117,needs design decision,38,"needs design/research label present; component/oncall labels: module: __torch_function__, module: autograd, module: docs",route to design/research decision before patch planning,2026-03-08T04:23:36Z
issue,175862,Adding Laplacian smoothing to MultiHeadAttention,https://github.com/pytorch/pytorch/issues/175862,needs design decision,38,"needs design/research label present; component/oncall labels: module: nn, module: sdpa",route to design/research decision before patch planning,2026-02-27T20:46:50Z
issue,110080,Devices API,https://github.com/pytorch/pytorch/issues/110080,needs design decision,38,needs design/research label present; component/oncall labels: module: python frontend,route to design/research decision before patch planning,2026-02-25T17:33:38Z
issue,169803,[RFC] Add RGE256ctr: A Fast Counter-Mode ARX RNG for PyTorch GeneratorImpl,https://github.com/pytorch/pytorch/issues/169803,needs design decision,38,needs design/research label present; component/oncall labels: module: random,route to design/research decision before patch planning,2026-02-23T21:29:46Z
issue,62530,[feature request] `torch.switch` (to mirror `lax.switch`) and `torch.cond` - with jax arithmetic coder as a case study,https://github.com/pytorch/pytorch/issues/62530,needs design decision,38,needs design/research label present; component/oncall labels: module: nn,route to design/research decision before patch planning,2026-02-23T20:40:30Z
issue,113737,torch.compile + SAC: mutations in backward are not preserved,https://github.com/pytorch/pytorch/issues/113737,needs design decision,38,"needs design/research label present; component/oncall labels: module: activation checkpointing, module: aotdispatch, module: pt2-dispatcher, oncall: distributed",route to design/research decision before patch planning,2026-02-23T04:21:30Z
issue,115516,[FakeTensor] FakeTensorMode dispatch won't change generator state,https://github.com/pytorch/pytorch/issues/115516,needs design decision,38,"needs design/research label present; component/oncall labels: module: faketensor, module: pt2-dispatcher, module: random, oncall: pt2",route to design/research decision before patch planning,2026-02-22T18:20:05Z
issue,124296,Request for modifying native allocator to use nvidia unified memory `cudaMallocManage`,https://github.com/pytorch/pytorch/issues/124296,needs design decision,38,"needs design/research label present; component/oncall labels: module: cuda, module: cudacachingallocator",route to design/research decision before patch planning,2026-02-22T08:08:13Z
issue,170224,Way to ask Dynamo to hoist some warmup logic to the beginning of the compile region,https://github.com/pytorch/pytorch/issues/170224,needs design decision,38,"needs design/research label present; component/oncall labels: module: dynamo, oncall: pt2",route to design/research decision before patch planning,2026-02-21T01:04:53Z
issue,101082,Multiple Learning Rate Scheduler for Specific Parameters Groups,https://github.com/pytorch/pytorch/issues/101082,needs design decision,38,"needs design/research label present; component/oncall labels: module: lrscheduler, module: optimizer",route to design/research decision before patch planning,2026-02-21T00:30:01Z
issue,110485,[export] `torch.tensor(0)` should not get burned in as a constant,https://github.com/pytorch/pytorch/issues/110485,needs design decision,38,"needs design/research label present; component/oncall labels: module: aotdispatch, module: dynamo, module: pt2-dispatcher, oncall: export",route to design/research decision before patch planning,2026-02-20T16:47:30Z
issue,110175,Module-level bufferization for torch dynamo module spanning multiple `fx.GraphModule`,https://github.com/pytorch/pytorch/issues/110175,needs design decision,38,"needs design/research label present; component/oncall labels: module: aotdispatch, module: pt2-dispatcher, oncall: pt2",route to design/research decision before patch planning,2026-02-20T16:40:39Z
issue,105053,Multiple dimensions support for `torch.max`,https://github.com/pytorch/pytorch/issues/105053,needs design decision,38,"needs design/research label present; component/oncall labels: module: numpy, module: python frontend",route to design/research decision before patch planning,2026-01-30T07:32:41Z
issue,173559,Secondary Optimizer for Muon,https://github.com/pytorch/pytorch/issues/173559,needs design decision,38,needs design/research label present; component/oncall labels: module: optimizer,route to design/research decision before patch planning,2026-01-28T17:12:21Z
issue,117844,torch.fft.rfft: Unsupported dtype BFloat16,https://github.com/pytorch/pytorch/issues/117844,needs design decision,38,"needs design/research label present; component/oncall labels: module: bfloat16, module: cuda, module: fft",route to design/research decision before patch planning,2026-01-13T10:43:46Z
issue,171676,"PyTorch allows `tensor.storage().resize_(0)`, leading to segmentation fault on subsequent tensor operations",https://github.com/pytorch/pytorch/issues/171676,needs design decision,38,"needs design/research label present; component/oncall labels: module: edge cases, module: error checking, module: molly-guard",route to design/research decision before patch planning,2026-01-12T16:40:45Z
issue,143412,[Inductor test failure] torch:inductor/test_select_algorithm TestSelectAlgorithm.test_convolution1 with cuda 12.6.3,https://github.com/pytorch/pytorch/issues/143412,has linked PR,38,"open PR references this issue; needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-11-26T08:23:37Z
issue,113800,InternalTorchDynamoError rewrapping loses exception chaining,https://github.com/pytorch/pytorch/issues/113800,has linked PR,38,"open PR references this issue; needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: compile ux, module: dynamo, module: logging, oncall: pt2",confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-09-12T21:32:10Z
issue,115484,torch.compile() breaks when using DeepSpeed ZeRO Level 3 sharding,https://github.com/pytorch/pytorch/issues/115484,has linked PR,38,"open PR references this issue; needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-06-27T00:17:11Z
issue,97078,[compile] TypeError: __init__() missing 1 required positional argument: 'parent_module',https://github.com/pytorch/pytorch/issues/97078,has linked PR,38,open PR references this issue; needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: oncall: pt2,confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-03-13T18:41:08Z
pr,189310,Generalize GeneratorState input device type,https://github.com/pytorch/pytorch/pull/189310,has linked issue,35,"PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",review when ready,2026-07-08T22:01:33Z
pr,189288,[inductor] Fix compute_ancestors KeyError from self-referential mutation copy,https://github.com/pytorch/pytorch/pull/189288,has linked issue,35,"PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",review when ready,2026-07-08T21:22:14Z
pr,189119,Fix test-jsons artifact upload path,https://github.com/pytorch/pytorch/pull/189119,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-07-08T21:13:59Z
pr,189010,[Dynamo] Support relax frontend in tvm backend,https://github.com/pytorch/pytorch/pull/189010,has linked issue,35,"PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",review when ready,2026-07-08T21:12:14Z
pr,189158,Skip GHA stats artifacts from other workflow attempts,https://github.com/pytorch/pytorch/pull/189158,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-07-08T21:09:22Z
pr,189154,Bundle unversioned libnuma for ROCm wheels,https://github.com/pytorch/pytorch/pull/189154,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-07-08T21:08:02Z
pr,189125,[Dynamo] Raise clear error for dynamic shapes in tvm relay path,https://github.com/pytorch/pytorch/pull/189125,has linked issue,35,"PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",review when ready,2026-07-08T20:50:41Z
pr,189127,[inductor] Decompose broadcast-bias baddbmm before pointwise,https://github.com/pytorch/pytorch/pull/189127,has linked issue,35,"PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",review when ready,2026-07-08T20:38:12Z
pr,189124,[inductor] Cache DSR-expanded single-config reductions,https://github.com/pytorch/pytorch/pull/189124,has linked issue,35,"PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",review when ready,2026-07-08T20:36:12Z
pr,189285,Fix MIOpen RNN dropout: gradient stall and missing per-step mask variation,https://github.com/pytorch/pytorch/pull/189285,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-07-08T20:34:34Z
pr,185794,Fix torch.kron RuntimeError on non-contiguous tensors - noah,https://github.com/pytorch/pytorch/pull/185794,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-07-08T20:31:48Z
pr,189145,[Dynamo] Preserve Stream/Event subclass types in graph reconstruction,https://github.com/pytorch/pytorch/pull/189145,has linked issue,35,"PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",review when ready,2026-07-08T20:27:40Z
pr,189174,Fix dcp privateuse1 stream support,https://github.com/pytorch/pytorch/pull/189174,has linked issue,35,PR closes or references open issue numbers; CI lane labels: ciflow/torchtitan,review when ready,2026-07-08T20:19:18Z
pr,189234,"Fix 1+erf catastrophic cancellation in gelu, Normal.cdf, and ndtr",https://github.com/pytorch/pytorch/pull/189234,has linked issue,35,"PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/mps, ciflow/torchtitan",review when ready,2026-07-08T20:12:59Z
pr,188795,Fix `torch.asarray` to preserve an input's device when a default device is set,https://github.com/pytorch/pytorch/pull/188795,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-07-08T19:33:54Z
pr,182945,Add pattern matcher to deduplicate var and std calculation,https://github.com/pytorch/pytorch/pull/182945,has linked issue,35,"PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk",review when ready,2026-07-08T17:58:53Z
pr,189045,Fix NaN gradients in nn.MultiheadAttention for fully-masked rows (#41…,https://github.com/pytorch/pytorch/pull/189045,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-07-08T17:31:38Z
pr,184094,Clone aliased mm/addmm inputs on CPU to avoid in-place gemm,https://github.com/pytorch/pytorch/pull/184094,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-07-08T16:56:12Z
pr,181720,[MPS] Make pin_memory return CPU-aliased storage backed by a unified MTLBuffer,https://github.com/pytorch/pytorch/pull/181720,has linked issue,35,PR closes or references open issue numbers; CI lane labels: ciflow/mps,review when ready,2026-07-08T15:34:47Z
pr,181308,[CPU] Fix std_mean/var_mean all-reduce ~4x slower than std + mean (#122191),https://github.com/pytorch/pytorch/pull/181308,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-07-08T15:27:39Z
pr,188053,test: add svdvals NaN propagation test (fixes #187759),https://github.com/pytorch/pytorch/pull/188053,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-07-08T11:37:03Z
pr,185751,"[nn] Raise ValueError early for invalid (ndim, pad_size) in non-constant F.pad modes",https://github.com/pytorch/pytorch/pull/185751,has linked issue,35,PR closes or references open issue numbers; CI lane labels: ciflow/trunk,review when ready,2026-07-08T05:20:35Z
pr,185756,[clamp] Fix float16 scalar overflow check inconsistency between CPU and GPU,https://github.com/pytorch/pytorch/pull/185756,has linked issue,35,PR closes or references open issue numbers; CI lane labels: ciflow/trunk,review when ready,2026-07-08T05:20:15Z
pr,187104,Fix embedding_bag CPU segfault when offsets is empty and indices is not,https://github.com/pytorch/pytorch/pull/187104,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-07-08T03:34:58Z
pr,184830,[Dynamo] Guard source-backed identity comparisons,https://github.com/pytorch/pytorch/pull/184830,has linked issue,35,"PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",review when ready,2026-07-07T23:45:32Z
pr,181027,Factor nanmean dtype checks into a helper + add kwarg coverage,https://github.com/pytorch/pytorch/pull/181027,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-07-07T22:17:01Z
pr,181062,Validate grad_output shape in multi_margin_loss_backward,https://github.com/pytorch/pytorch/pull/181062,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-07-07T22:15:59Z
pr,188039,autograd: fix PyObject* refcount leak in setup_context path,https://github.com/pytorch/pytorch/pull/188039,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-07-07T22:06:39Z
pr,148674,Set /NODEFAULTLIB:vcomp for MSVC when linking caffe2::mkl with libiomp5md.lib,https://github.com/pytorch/pytorch/pull/148674,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-07-07T21:45:46Z
pr,184780,[MPS] Fast atomic-free path for flat torch.unique (fixes #97310 perf + #111173 correctness),https://github.com/pytorch/pytorch/pull/184780,has linked issue,35,PR closes or references open issue numbers; CI lane labels: ciflow/mps,review when ready,2026-07-07T21:14:05Z
pr,185054,[autograd] Validate forward-mode jacobian/hessian require vectorize=T…,https://github.com/pytorch/pytorch/pull/185054,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-07-07T21:10:35Z
pr,186039,Fix Dynamo verbose guard checks for frame locals,https://github.com/pytorch/pytorch/pull/186039,has linked issue,35,"PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",review when ready,2026-07-07T13:27:02Z
pr,182273,[c10d] Fix device_id not propagated through ProcessGroupWrapper,https://github.com/pytorch/pytorch/pull/182273,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-07-07T09:47:10Z
pr,188906,Add torch.utils.checkpoint.register_determinism_check,https://github.com/pytorch/pytorch/pull/188906,has linked issue,35,PR closes or references open issue numbers; CI lane labels: ciflow/torchtitan,review when ready,2026-07-07T09:13:44Z
pr,184809,Improve custom op Tensor subclass fake error,https://github.com/pytorch/pytorch/pull/184809,has linked issue,35,"PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",review when ready,2026-07-07T08:21:23Z
pr,178844,[c10d] Support stacked output tensors in Gloo all_gather_into_tensor …,https://github.com/pytorch/pytorch/pull/178844,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-07-07T05:58:57Z
pr,185782,fix: [PT2] [Hardening] Track recompiles alongside graph breaks in our actu...,https://github.com/pytorch/pytorch/pull/185782,has linked issue,35,"PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",review when ready,2026-07-07T05:44:08Z
pr,186375,[CUDA] Change nvmlDeviceGetGpuFabricInfoV assertion to warning,https://github.com/pytorch/pytorch/pull/186375,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-07-07T04:56:49Z
pr,180408,Fix torch.eye CPU performance cliff at d=182 (#48251),https://github.com/pytorch/pytorch/pull/180408,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-07-06T20:31:27Z
pr,188945,Fix assert_close to support comparisons involving Python and NumPy scalars,https://github.com/pytorch/pytorch/pull/188945,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-07-06T18:59:36Z
pr,188923,test: add gradient health validation tests (fixes #177116),https://github.com/pytorch/pytorch/pull/188923,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-07-06T18:59:05Z
pr,188969,[docs] document __torch_function__ behavior during autograd hooks,https://github.com/pytorch/pytorch/pull/188969,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-07-06T18:57:22Z
pr,188883,Handle empty autograd sections in backend YAML,https://github.com/pytorch/pytorch/pull/188883,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-07-06T18:51:38Z
pr,188831,Fix module re-export in tensor/experimental API,https://github.com/pytorch/pytorch/pull/188831,has linked issue,35,"PR closes or references open issue numbers; CI lane labels: ciflow/dtensor, ciflow/inductor, ciflow/torchtitan",review when ready,2026-07-06T18:50:44Z
pr,188893,[ONNX] Clarify that output_names labels outputs but does not reorder them,https://github.com/pytorch/pytorch/pull/188893,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-07-06T18:47:17Z
pr,188857,Fix type confusion in Stream comparison against non-Stream objects,https://github.com/pytorch/pytorch/pull/188857,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-07-06T18:46:58Z
pr,188843,Fix scatter/scatter_add/scatter_reduce src gradient when index is sma…,https://github.com/pytorch/pytorch/pull/188843,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-07-06T14:33:01Z
pr,188968,Fix record_stack_traces returning empty trace when no forward frame found,https://github.com/pytorch/pytorch/pull/188968,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-07-06T14:32:32Z
pr,188637,[inductor] Reindex transpose-contiguous clone to enable vertical fusion,https://github.com/pytorch/pytorch/pull/188637,has linked issue,35,"PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",review when ready,2026-07-06T07:13:02Z
pr,188642,Fix __qualname__ for OpOverloadPacket,https://github.com/pytorch/pytorch/pull/188642,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-07-05T23:21:14Z
pr,182541,add regression test for weight_norm meta initialization (#96409),https://github.com/pytorch/pytorch/pull/182541,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-07-05T20:49:12Z
pr,187325,[Metal][memory] add configurable memory budget API for MPS allocator (#105839),https://github.com/pytorch/pytorch/pull/187325,has linked issue,35,PR closes or references open issue numbers; CI lane labels: ciflow/mps,review when ready,2026-07-05T15:46:24Z
pr,188066,test: add F.normalize zero-input correctness tests (fixes #184575),https://github.com/pytorch/pytorch/pull/188066,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-07-05T15:20:50Z
pr,187453,[MPS] Add error handling for complex dtypes in max unpooling operations,https://github.com/pytorch/pytorch/pull/187453,has linked issue,35,PR closes or references open issue numbers; CI lane labels: ciflow/mps,review when ready,2026-07-05T12:44:29Z
pr,182296,fix(nn): add input dim validation to adaptive_avg_pool2d/3d,https://github.com/pytorch/pytorch/pull/182296,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-07-04T22:22:31Z
pr,182291,fix(nn): InstanceNorm always raises ValueError on channel mismatch,https://github.com/pytorch/pytorch/pull/182291,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-07-04T22:21:46Z
pr,188579,optimize split reduction heuristics on GB200,https://github.com/pytorch/pytorch/pull/188579,has linked issue,35,"PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",review when ready,2026-07-04T20:26:06Z
pr,188791,Fix MKL VML first-call dispatch race producing silent low-precision results,https://github.com/pytorch/pytorch/pull/188791,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-07-04T20:17:38Z
pr,182268,autograd: add `list_saved_tensors` utility to traverse saved tensors in autograd graph,https://github.com/pytorch/pytorch/pull/182268,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-07-04T18:50:12Z
pr,182262,Use TypeAliasType for type aliases and @exposed_in for function re-exports,https://github.com/pytorch/pytorch/pull/182262,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-07-04T18:50:11Z
pr,182204,torch.cat/concat/concatenate: support dim=None (flatten before concat),https://github.com/pytorch/pytorch/pull/182204,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-07-04T18:50:08Z
pr,182133,Allow torch.arange() to return empty tensor when bounds are inconsistent with step sign,https://github.com/pytorch/pytorch/pull/182133,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-07-04T18:50:05Z
pr,182076,[Fix] Add overflow and validation checks for pixel_shuffle in decomposition and meta,https://github.com/pytorch/pytorch/pull/182076,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-07-04T15:13:40Z
pr,181756,[Docs] Replace outdated flake8 and git-pre-commit references with lintrunner,https://github.com/pytorch/pytorch/pull/181756,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-07-04T15:13:39Z
pr,176265,Add backward batching rules for SDPA and fix attn_bias under vmap,https://github.com/pytorch/pytorch/pull/176265,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-07-04T10:38:32Z
pr,188719,Fix Inductor KeyError on view_as_complex of non-contiguous CUDA tensors,https://github.com/pytorch/pytorch/pull/188719,has linked issue,35,"PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",review when ready,2026-07-03T07:04:20Z
pr,188633,Draft: Update CUTLASS fence proxy API usage,https://github.com/pytorch/pytorch/pull/188633,has linked issue,35,"PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",review when ready,2026-07-03T02:40:12Z
pr,188528,cuda: reject negative caching allocator sizes,https://github.com/pytorch/pytorch/pull/188528,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-07-03T01:38:20Z
pr,166388,Fixing assertion fail by checking if no_sync is true before DDPsink,https://github.com/pytorch/pytorch/pull/166388,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-07-02T23:22:38Z
pr,188762,Fix multinomial without replacement silently sampling zero-probability categories,https://github.com/pytorch/pytorch/pull/188762,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-07-02T23:01:26Z
pr,188200,[MPS] Vectorized contiguous cat fast path for any dimension,https://github.com/pytorch/pytorch/pull/188200,has linked issue,35,PR closes or references open issue numbers; CI lane labels: ciflow/mps,review when ready,2026-07-02T20:29:24Z
pr,186524,Validate conv transpose output padding,https://github.com/pytorch/pytorch/pull/186524,has linked issue,35,PR closes or references open issue numbers; CI lane labels: ciflow/inductor,review when ready,2026-07-02T16:36:56Z
pr,188560,[MPS] Fix constant address space for scalar params in binary_dense_scalar kernels,https://github.com/pytorch/pytorch/pull/188560,has linked issue,35,PR closes or references open issue numbers; CI lane labels: ciflow/mps,review when ready,2026-07-02T12:56:21Z
pr,187327,[Metal][cache] dual-hash verification for MPSGraphCache and MPSKernelCache (#77176),https://github.com/pytorch/pytorch/pull/187327,has linked issue,35,PR closes or references open issue numbers; CI lane labels: ciflow/mps,review when ready,2026-07-02T12:49:24Z
pr,187330,[Metal][memory] clear graph/kernel compilation caches in empty_cache() (#154329),https://github.com/pytorch/pytorch/pull/187330,has linked issue,35,PR closes or references open issue numbers; CI lane labels: ciflow/mps,review when ready,2026-07-02T12:23:41Z
pr,188349,[HOP] Adding length parameter to scan,https://github.com/pytorch/pytorch/pull/188349,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-07-02T07:59:50Z
pr,172219,[XPU] Adapt test_basic_chrome_trace to XPU,https://github.com/pytorch/pytorch/pull/172219,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-07-02T05:15:03Z
pr,188647,torch/fx: fix _format_target to handle Python keywords and embedded quotes,https://github.com/pytorch/pytorch/pull/188647,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-07-02T05:07:39Z
pr,182627,Handle bool inputs in Scalar::log,https://github.com/pytorch/pytorch/pull/182627,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-07-02T03:43:44Z
pr,188531,pooling: guard max_pool1d integer parameters,https://github.com/pytorch/pytorch/pull/188531,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-07-02T01:32:30Z
pr,188732,[MPS] Enable ConvTranspose3d for fp16/bf16 via internal fp32 upcast,https://github.com/pytorch/pytorch/pull/188732,has linked issue,35,PR closes or references open issue numbers; CI lane labels: ciflow/mps,review when ready,2026-07-02T00:38:47Z
pr,187544,Fix torch.gcd returning negative results for INT_MIN inputs,https://github.com/pytorch/pytorch/pull/187544,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-07-01T18:10:34Z
pr,188260,[nn] Fix RReLU noise tensor not being set in eval mode,https://github.com/pytorch/pytorch/pull/188260,has linked issue,35,"PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",review when ready,2026-07-01T13:56:24Z
pr,187257,[DCP] Raise RuntimeError on load dtype mismatch to prevent precision …,https://github.com/pytorch/pytorch/pull/187257,has linked issue,35,PR closes or references open issue numbers; CI lane labels: ciflow/torchtitan,review when ready,2026-07-01T13:30:53Z
pr,179273,[optim] support of dynamic coefficients and normalization in Muon,https://github.com/pytorch/pytorch/pull/179273,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-07-01T11:36:49Z
pr,188543,Remove deprecated torch.utils.mobile_optimizer from API index,https://github.com/pytorch/pytorch/pull/188543,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-07-01T03:45:16Z
pr,187309,Add maximize parameter to LBFGS optimizer,https://github.com/pytorch/pytorch/pull/187309,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-07-01T01:54:07Z
pr,188526,error checking: validate _pdist_forward dimensions,https://github.com/pytorch/pytorch/pull/188526,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-30T22:49:04Z
pr,188524,test: centralize optimizer comparison tolerances,https://github.com/pytorch/pytorch/pull/188524,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-30T18:06:14Z
pr,188525,test: add Linear module error inputs,https://github.com/pytorch/pytorch/pull/188525,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-30T18:06:02Z
pr,176434,[export] Move schema deserialization to C++ for faster export.load(),https://github.com/pytorch/pytorch/pull/176434,has linked issue,35,PR closes or references open issue numbers; CI lane labels: ciflow/inductor,review when ready,2026-06-30T16:15:57Z
pr,188369,Add DTensor sharding strategy for FFT c2c,https://github.com/pytorch/pytorch/pull/188369,has linked issue,35,"PR closes or references open issue numbers; CI lane labels: ciflow/dtensor, ciflow/inductor, ciflow/torchtitan",review when ready,2026-06-30T04:26:57Z
pr,188399,Fix #171905: Use typing_extensions.TypeAliasType for better reexport of `__module__,https://github.com/pytorch/pytorch/pull/188399,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-29T21:15:17Z
pr,188373,Enforce 1D tensor validation for batch_norm parameters,https://github.com/pytorch/pytorch/pull/188373,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-29T20:46:06Z
pr,188363,Add kernel_size > 0 validation to FractionalMaxPool2d matching Fracti…,https://github.com/pytorch/pytorch/pull/188363,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-29T16:53:32Z
pr,188381,aten: add output_size kwarg to masked_select for meta/eager shape oracle support,https://github.com/pytorch/pytorch/pull/188381,has linked issue,35,PR closes or references open issue numbers; CI lane labels: ciflow/mps,review when ready,2026-06-29T16:50:49Z
pr,188347,Preserve tied parameter and buffer aliasing in load_state_dict(assign=True),https://github.com/pytorch/pytorch/pull/188347,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-29T16:28:37Z
pr,188334,Handle negative integer exponents in SymPy utilities,https://github.com/pytorch/pytorch/pull/188334,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-29T16:28:31Z
pr,188271,Allow PrivateUse1 backends to opt out of CEA decomposition,https://github.com/pytorch/pytorch/pull/188271,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-27T22:05:19Z
pr,181681,Fix UBSan null-pointer UB in vec256 partial-load Vec::loadu (#181510),https://github.com/pytorch/pytorch/pull/181681,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-27T19:56:28Z
pr,188183,[nn] Avoid mutating state_dict metadata in load_state_dict(assign=True),https://github.com/pytorch/pytorch/pull/188183,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-27T17:47:49Z
pr,179377,Fix gaussian_nll_loss rejecting valid multi-dim broadcastable var shapes,https://github.com/pytorch/pytorch/pull/179377,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-27T10:17:30Z
pr,185056,Raise error when class attributes shadow registered parameters or modules,https://github.com/pytorch/pytorch/pull/185056,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-27T08:37:47Z
pr,187795,Guard release_available_cached_blocks during CUDA graph capture,https://github.com/pytorch/pytorch/pull/187795,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-27T08:06:18Z
pr,183743,Add drift-based tolerance fallback to op tests,https://github.com/pytorch/pytorch/pull/183743,has linked issue,35,PR closes or references open issue numbers; CI lane labels: ciflow/mps,review when ready,2026-06-26T21:51:47Z
pr,188099,[JIT] Use native fill/zero ops for outplace tracing of fill_/zero_,https://github.com/pytorch/pytorch/pull/188099,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-26T20:26:35Z
pr,188258,"Fix _foreach_{add,mul,div}.Tensor autograd",https://github.com/pytorch/pytorch/pull/188258,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-26T19:29:57Z
pr,188149,Handle mixed Tensor/Scalar min/max bounds in torch.clamp,https://github.com/pytorch/pytorch/pull/188149,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-26T18:24:30Z
pr,188138,fix: Docs are little bit outdated for torch logs,https://github.com/pytorch/pytorch/pull/188138,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-26T17:22:45Z
pr,188270,Fix inplace modification error with softmax when combining forward and backward AD,https://github.com/pytorch/pytorch/pull/188270,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-26T15:54:53Z
pr,182601,docs: clarify tile dims argument,https://github.com/pytorch/pytorch/pull/182601,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-26T15:38:32Z
pr,182602,docs: cross-reference sigmoid APIs,https://github.com/pytorch/pytorch/pull/182602,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-26T15:38:24Z
pr,182603,docs: clarify RNN hidden state dimensions,https://github.com/pytorch/pytorch/pull/182603,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-26T15:38:15Z
pr,182604,docs: clarify multithreaded autograd guidance,https://github.com/pytorch/pytorch/pull/182604,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-26T15:38:03Z
pr,185398,docs: clean up multilabel soft margin loss args,https://github.com/pytorch/pytorch/pull/185398,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-26T15:37:51Z
pr,185744,docs: clarify LSTM equation notation,https://github.com/pytorch/pytorch/pull/185744,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-26T15:37:39Z
pr,185745,docs: clarify scalar fill values,https://github.com/pytorch/pytorch/pull/185745,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-26T15:37:31Z
pr,182605,test: strengthen optimizer step hook ordering checks,https://github.com/pytorch/pytorch/pull/182605,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-26T14:59:02Z
pr,183316,Fix/signbit cuda half nan,https://github.com/pytorch/pytorch/pull/183316,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-26T09:13:26Z
pr,187246,serialization: fix nondeterministic pickle bytes in _legacy_save (gh#39383),https://github.com/pytorch/pytorch/pull/187246,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-25T18:00:04Z
pr,187630,fx/passes: fix non-deterministic ordering in get_source_partitions(),https://github.com/pytorch/pytorch/pull/187630,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-24T19:05:24Z
pr,187803,fix(autograd): sinc_backward NaN for second derivative at x=0,https://github.com/pytorch/pytorch/pull/187803,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-24T18:38:42Z
pr,181051,Fix heap-buffer-overflow in quantized_max_pool2d for 1d args,https://github.com/pytorch/pytorch/pull/181051,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-24T02:31:02Z
pr,175525,vmap compatibility with non-tensor leaves,https://github.com/pytorch/pytorch/pull/175525,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-23T18:28:04Z
pr,186929,[fx] fix _patch_function using wrong function when forward is wrapped,https://github.com/pytorch/pytorch/pull/186929,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-23T14:29:39Z
pr,187764,[MPS] Add matrix_exp support,https://github.com/pytorch/pytorch/pull/187764,has linked issue,35,PR closes or references open issue numbers; CI lane labels: ciflow/mps,review when ready,2026-06-23T14:24:24Z
pr,187814,Reject factor=0 in ConstantLR to avoid later ZeroDivisionError,https://github.com/pytorch/pytorch/pull/187814,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-23T14:22:29Z
pr,187757,Narrow ModuleDict.__getattr__ return type to Module for type checkers,https://github.com/pytorch/pytorch/pull/187757,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-23T14:21:26Z
pr,187790,[ao][pruning] Add experimental topology-aware modules,https://github.com/pytorch/pytorch/pull/187790,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-23T14:00:31Z
pr,187783,fix(nn): raise ValueError instead of ZeroDivisionError for norm_type=0 in lp_pool1d/2d/3d and LPPool modules,https://github.com/pytorch/pytorch/pull/187783,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-23T13:59:01Z
pr,187748,[MPS] Build only the reduced Q in linalg.qr (fix tall-skinny perf cliff),https://github.com/pytorch/pytorch/pull/187748,has linked issue,35,PR closes or references open issue numbers; CI lane labels: ciflow/mps,review when ready,2026-06-23T13:58:56Z
pr,187153,jit: add bounds checking for textIndex in SourceRangeDeserializer,https://github.com/pytorch/pytorch/pull/187153,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-23T13:57:55Z
pr,187570,Fixed heaviside nan propagation issue,https://github.com/pytorch/pytorch/pull/187570,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-23T13:20:10Z
pr,187597,Add return type annotation to nn.Module.named_modules,https://github.com/pytorch/pytorch/pull/187597,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-23T03:43:47Z
pr,177985,Fix SDPA mem-efficient attention crash when num_heads >= 65536,https://github.com/pytorch/pytorch/pull/177985,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-22T19:18:16Z
pr,187771,Warn when convert_sync_batchnorm drops _BatchNorm subclass behavior,https://github.com/pytorch/pytorch/pull/187771,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-22T13:36:13Z
pr,187198,Add target range validation in binary_cross_entropy_with_logits,https://github.com/pytorch/pytorch/pull/187198,has linked issue,35,"PR closes or references open issue numbers; CI lane labels: ciflow/dtensor, ciflow/trunk",review when ready,2026-06-22T13:01:20Z
pr,180184,Export auto-generated unstructured inplace meta dispatch symbols,https://github.com/pytorch/pytorch/pull/180184,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-22T11:23:16Z
pr,184641,Fix GroupNorm validation for non-positive num-groups,https://github.com/pytorch/pytorch/pull/184641,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-22T07:34:41Z
pr,178099,Add TypeGuard[Tensor] return type to is_tensor_like,https://github.com/pytorch/pytorch/pull/178099,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-21T23:32:58Z
pr,179827,Replace BF16 matmul OpInfo checks with a feature query,https://github.com/pytorch/pytorch/pull/179827,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-21T21:58:37Z
pr,180781,Fix integer divide-by-zero UB in CUDA binary kernels,https://github.com/pytorch/pytorch/pull/180781,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-21T20:07:02Z
pr,186571,Fix #176298: support unsigned add and sub on CPU,https://github.com/pytorch/pytorch/pull/186571,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-20T22:51:02Z
pr,180601,"Fix pydocstyle errors in _guards.py, _ops.py, _jit_internal.py, functional.py, _tensor_str.py, library.py",https://github.com/pytorch/pytorch/pull/180601,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-20T20:02:57Z
pr,185319,Sort record names before hashing in writeSerializationId,https://github.com/pytorch/pytorch/pull/185319,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-20T09:15:33Z
pr,187595,Fix torch.gcd/lcm integer overflow on INT_MIN inputs,https://github.com/pytorch/pytorch/pull/187595,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-19T23:33:16Z
pr,180188,Validate opt_dtype in nanmean to reject non-floating-point types,https://github.com/pytorch/pytorch/pull/180188,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-19T17:18:04Z
pr,177455,Add nn.SwiGLU activation function,https://github.com/pytorch/pytorch/pull/177455,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-19T16:40:56Z
pr,180764,cpp_extension: add CUDA dlink flags to JIT load APIs,https://github.com/pytorch/pytorch/pull/180764,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-19T13:44:47Z
pr,187656,Reject complex scale/zero_point tensors in quantize_per_tensor with an explicit dtype error instead of value cannot be converted to type double without overflow.,https://github.com/pytorch/pytorch/pull/187656,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-19T06:01:06Z
pr,187662,Reject aliased Q/R out tensors in torch.linalg.qr,https://github.com/pytorch/pytorch/pull/187662,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-18T16:58:06Z
pr,187286,nn: add torch.nn.functional.rotary_embedding and torch.nn.RotaryEmbedding,https://github.com/pytorch/pytorch/pull/187286,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-17T14:54:27Z
pr,172539,[torchgen] Enable Structured Kernels and Out-as-Primary Wrappers for PrivateUse1,https://github.com/pytorch/pytorch/pull/172539,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-17T07:00:03Z
pr,186050,Output Shape Check added to transposed Conv,https://github.com/pytorch/pytorch/pull/186050,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-17T05:12:18Z
pr,185807,Stable sorting for consistent gradient for torch.quantile on tied values,https://github.com/pytorch/pytorch/pull/185807,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-17T05:11:36Z
pr,187253,Align dynamic qparams test helper with C++ overflow check,https://github.com/pytorch/pytorch/pull/187253,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-16T21:10:10Z
pr,187238,Add docs pages for torch.e and torch.pi,https://github.com/pytorch/pytorch/pull/187238,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-16T14:49:12Z
pr,186589,Fix serialization errors in SubTest calls,https://github.com/pytorch/pytorch/pull/186589,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-16T14:25:09Z
pr,187329,Raise ValueError for non-positive kernel_size/dilation/stride in unfold/fold,https://github.com/pytorch/pytorch/pull/187329,has linked issue,35,PR closes or references open issue numbers; CI lane labels: ciflow/mps,review when ready,2026-06-16T14:24:24Z
pr,187243,Reject p < 1 in triplet_margin_loss to prevent NaN from p-norm overflow,https://github.com/pytorch/pytorch/pull/187243,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-16T14:24:23Z
pr,187242,Validate LocalResponseNorm parameters to prevent silent inf output,https://github.com/pytorch/pytorch/pull/187242,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-16T14:24:23Z
pr,186930,Fix NaN propagation behavior in torch.sign for CPU and CUDA backends,https://github.com/pytorch/pytorch/pull/186930,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-16T10:13:22Z
pr,173894,[nn] Fix softplus numerical instability with large beta values,https://github.com/pytorch/pytorch/pull/173894,has linked issue,35,PR closes or references open issue numbers; CI lane labels: ciflow/mps,review when ready,2026-06-15T18:54:17Z
pr,179941,[inductor] Add reduction loop peeling for Triton codegen,https://github.com/pytorch/pytorch/pull/179941,has linked issue,35,"PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",review when ready,2026-06-15T16:51:54Z
pr,176746,Added 'gelsd' to CUDA torch.linalg.lstsq driver so that underdetermined least square systems can be solved on CUDA,https://github.com/pytorch/pytorch/pull/176746,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-15T12:42:30Z
pr,177866,Sparse semi structured setup/teardown latency,https://github.com/pytorch/pytorch/pull/177866,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-15T12:41:56Z
pr,178797,Fix meta conv shape checks for invalid groups vs out_channels,https://github.com/pytorch/pytorch/pull/178797,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-15T12:33:52Z
pr,179160,Fix uint32 overflow in CUDA scan _with_indices kernels,https://github.com/pytorch/pytorch/pull/179160,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-15T10:35:26Z
pr,179379,[DTensor] Convert type aliases to TypeAliasType in experimental/,https://github.com/pytorch/pytorch/pull/179379,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-14T16:59:28Z
pr,187101,Add missing dispatch for bf16 in tensor_cpu,https://github.com/pytorch/pytorch/pull/187101,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-14T16:22:04Z
pr,187187,[typing] Replace TypeAlias with TypeAliasType in torch/types.py,https://github.com/pytorch/pytorch/pull/187187,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-13T07:15:22Z
pr,187090,Fix nn.Module.to() failure on subclass parameters held by Python weakrefs,https://github.com/pytorch/pytorch/pull/187090,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-12T17:59:41Z
pr,178371,[Profiler] Fix profile_memory not working with profile_all_threads,https://github.com/pytorch/pytorch/pull/178371,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-12T03:48:16Z
pr,186940,[ShardedTensor] use device-agnostic fallback in tensor_device (fixes #186938),https://github.com/pytorch/pytorch/pull/186940,has linked issue,35,PR closes or references open issue numbers; CI lane labels: ciflow/torchtitan,review when ready,2026-06-11T20:01:15Z
pr,175509,Add fast path for min/max dim reductions with size 2 on CPU,https://github.com/pytorch/pytorch/pull/175509,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-11T19:24:58Z
pr,173239,Test openreg inference hooks,https://github.com/pytorch/pytorch/pull/173239,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-11T19:24:28Z
pr,174196,Add validation for lstm.data hidden state shapes to prevent segfaults,https://github.com/pytorch/pytorch/pull/174196,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-10T23:56:04Z
pr,186863,[MPS] Register unsigned binary Metal kernels,https://github.com/pytorch/pytorch/pull/186863,has linked issue,35,PR closes or references open issue numbers; CI lane labels: ciflow/mps,review when ready,2026-06-10T19:20:14Z
pr,174559,[lint] Enable Ruff PYI056 by replacing __all__.append/extend usage,https://github.com/pytorch/pytorch/pull/174559,has linked issue,35,"PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",review when ready,2026-06-09T15:34:39Z
pr,186177,[autograd] Defer device thread creation when multithreading is disabled,https://github.com/pytorch/pytorch/pull/186177,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-09T15:31:17Z
pr,186607,Raise a clear error for malformed allocator config instead of INTERNAL_ASSERT,https://github.com/pytorch/pytorch/pull/186607,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-09T03:28:06Z
pr,186207,Fix segfault and invalid cast in UntypedStorage,https://github.com/pytorch/pytorch/pull/186207,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-08T14:19:24Z
pr,172829,Fixes registration of custom sharding strategies when using register_sharding with TensorList,https://github.com/pytorch/pytorch/pull/172829,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-06T15:09:53Z
pr,185187,Validate forward-mode autograd.functional vectorize usage,https://github.com/pytorch/pytorch/pull/185187,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-05T21:45:21Z
pr,185864,Fix segfault in native_batch_norm with empty running_mean/running_var,https://github.com/pytorch/pytorch/pull/185864,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-05T18:37:05Z
pr,186073,Support weights_only=True in torch.export.load,https://github.com/pytorch/pytorch/pull/186073,has linked issue,35,PR closes or references open issue numbers; CI lane labels: ciflow/inductor,review when ready,2026-06-05T05:57:30Z
pr,186128,Reject dtype conversion of quantized TypedStorage with a clear error,https://github.com/pytorch/pytorch/pull/186128,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-04T22:18:57Z
pr,186159,Restore Any return type for Module.__getattr__ to fix typing false positives,https://github.com/pytorch/pytorch/pull/186159,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-04T13:43:08Z
pr,185959,Fix SIGFPE in torch.lcm with the integer type minimum,https://github.com/pytorch/pytorch/pull/185959,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-03T22:23:51Z
pr,185868,"Update clamp, clamp_min, clamp_max to use the minimum norm subgradient",https://github.com/pytorch/pytorch/pull/185868,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-03T22:01:26Z
pr,184577,nn: validate num_groups > 0 in GroupNorm.__init__,https://github.com/pytorch/pytorch/pull/184577,has linked issue,35,PR closes or references open issue numbers; CI lane labels: ciflow/dtensor,review when ready,2026-06-03T10:48:25Z
pr,185117,nn: add missing memory_format overload to Module.to(),https://github.com/pytorch/pytorch/pull/185117,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-03T10:48:12Z
pr,184898,autograd: raise ValueError for forward-mode without vectorize in jacobian/hessian,https://github.com/pytorch/pytorch/pull/184898,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-03T10:48:10Z
pr,185747,Use at::mps::OptionalMPSGuard for MPS dispatch keys in codegen,https://github.com/pytorch/pytorch/pull/185747,has linked issue,35,PR closes or references open issue numbers; CI lane labels: ciflow/mps,review when ready,2026-06-03T06:03:40Z
pr,175542,[DTensor] Register DTensorSpec as pytree constant at module level,https://github.com/pytorch/pytorch/pull/175542,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-03T03:46:21Z
pr,185792,test/distributed/pipelining: add end-to-end tests with HuggingFace GPT-2 and BERT,https://github.com/pytorch/pytorch/pull/185792,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-01T22:37:41Z
pr,180444,Add `validate` parameter to `torch.multinomial` to skip input validation,https://github.com/pytorch/pytorch/pull/180444,has linked issue,35,PR closes or references open issue numbers; CI lane labels: ciflow/mps,review when ready,2026-06-01T01:55:05Z
pr,180496,feat: Add OpenTelemetry trace export support for torch.profiler,https://github.com/pytorch/pytorch/pull/180496,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-01T01:53:04Z
pr,185454,"fix clamp and clamp_{min,max} gradient at scalar bound to be 0",https://github.com/pytorch/pytorch/pull/185454,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-05-29T15:35:41Z
pr,184698,Materialize lazy modules fully on load_state_dict instead of leaving them in a partially initialized state.,https://github.com/pytorch/pytorch/pull/184698,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-05-29T09:45:11Z
pr,183269,[DTensor] Switch test_embedding_ops to DTensorOpTestBase (MultiThreadedTestCase),https://github.com/pytorch/pytorch/pull/183269,has linked issue,35,PR closes or references open issue numbers; CI lane labels: ciflow/dtensor,review when ready,2026-05-29T05:58:52Z
pr,148023,torch.utils.checkpoint preserves torch function mode stack during recompute,https://github.com/pytorch/pytorch/pull/148023,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-05-28T12:47:26Z
pr,184064,Fix silent int64 precision loss in F.pad constant mode,https://github.com/pytorch/pytorch/pull/184064,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-05-27T20:32:14Z
pr,178031,[MPS] Add adaptive_avg_pool3d support for MPS,https://github.com/pytorch/pytorch/pull/178031,has linked issue,35,"PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/mps",review when ready,2026-05-27T05:44:35Z
pr,182384,[autograd] Refresh ctx.needs_input_grad for partial backward,https://github.com/pytorch/pytorch/pull/182384,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-05-26T19:44:04Z
pr,185026,[typing] Add Iterable base class to ModuleList,https://github.com/pytorch/pytorch/pull/185026,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-05-26T13:51:35Z
pr,184885,docs: correct torch.addcdiv prose to match implementation order,https://github.com/pytorch/pytorch/pull/184885,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-05-26T13:44:56Z
pr,184857,"Raise ValueError early when strategy=""forward-mode"" used without vectorize=True",https://github.com/pytorch/pytorch/pull/184857,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-05-26T13:44:24Z
pr,185193,Validate forward-mode functional autograd arguments,https://github.com/pytorch/pytorch/pull/185193,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-05-26T13:36:34Z
pr,184840,Add Module.to overload for memory_format,https://github.com/pytorch/pytorch/pull/184840,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-05-26T13:35:02Z
pr,181993,[MPS] fix device binary shaders,https://github.com/pytorch/pytorch/pull/181993,has linked issue,35,PR closes or references open issue numbers; CI lane labels: ciflow/mps,review when ready,2026-05-26T10:06:18Z
pr,177961,[ROCm] Enable native AsyncTP,https://github.com/pytorch/pytorch/pull/177961,waiting on contributor,35,"review decision requests changes; PR has not updated for at least 45 days; CI lane labels: ciflow/h100-symm-mem, ciflow/periodic-rocm-mi300, ciflow/torchtitan, ciflow/trunk",wait for contributor update,2026-05-21T14:38:28Z
pr,181352,Re-enable nightly tests on s390x,https://github.com/pytorch/pytorch/pull/181352,waiting on contributor,35,"review decision requests changes; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/s390",wait for contributor update,2026-05-20T06:57:57Z
pr,169107,Add failing test for PolynomialLR total_iters=0 closed-form crash,https://github.com/pytorch/pytorch/pull/169107,waiting on contributor,35,review decision requests changes; PR has not updated for at least 45 days,wait for contributor update,2026-05-19T19:11:12Z
pr,183749,Support auto_functionalized_v2 in can_fuse_into_auto_functionalized,https://github.com/pytorch/pytorch/pull/183749,waiting on contributor,35,"review decision requests changes; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-05-19T14:46:12Z
pr,181520,dataloader: add performance warning for in-memory datasets with num_workers>0,https://github.com/pytorch/pytorch/pull/181520,waiting on contributor,35,review decision requests changes; PR has not updated for at least 45 days,wait for contributor update,2026-05-16T01:07:15Z
pr,181501,docs: add DataLoader performance guidance for in-memory datasets,https://github.com/pytorch/pytorch/pull/181501,waiting on contributor,35,review decision requests changes; PR has not updated for at least 45 days,wait for contributor update,2026-05-16T01:07:10Z
pr,183433,[inductor] Fix backslash escaping order in user-defined Triton kernel…,https://github.com/pytorch/pytorch/pull/183433,waiting on contributor,35,"review decision requests changes; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-05-14T16:04:45Z
pr,181110,Workaround for nvrtcCompileProgram changing locale,https://github.com/pytorch/pytorch/pull/181110,waiting on contributor,35,review decision requests changes; PR has not updated for at least 45 days; CI lane labels: ciflow/binaries,wait for contributor update,2026-05-12T15:38:03Z
pr,182612,Ignore decoding errors when reading CMake cache files (errors='ignore'),https://github.com/pytorch/pytorch/pull/182612,waiting on contributor,35,review decision requests changes; PR has not updated for at least 45 days,wait for contributor update,2026-05-12T02:00:14Z
pr,178390,[inductor] fold half addmm+relu into _addmm_activation,https://github.com/pytorch/pytorch/pull/178390,waiting on contributor,35,"review decision requests changes; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-05-10T20:49:42Z
pr,134942,Enabling ATen Distribution kernels for AARCH64 using OpenRNG,https://github.com/pytorch/pytorch/pull/134942,waiting on contributor,35,review decision requests changes; PR has not updated for at least 45 days,wait for contributor update,2026-04-20T18:28:18Z
pr,138388,Extending SVE VEC Backend Support in PyTorch to SVE128 and SVE512.,https://github.com/pytorch/pytorch/pull/138388,waiting on contributor,35,"review decision requests changes; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/linux-aarch64",wait for contributor update,2025-08-11T18:47:31Z
issue,177952,Fix onnx tests running on GPU/CUDA,https://github.com/pytorch/pytorch/issues/177952,needs reproduction,32,needs reproduction label present; component/oncall labels: module: onnx,confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-07-08T14:55:43Z
issue,130253,"RuntimeError: CUDA error: CUBLAS_STATUS_EXECUTION_FAILED when calling `cublasGemmEx( handle, opa, opb, m, n, k, &falpha, a, CUDA_R_16F, lda, b, CUDA_R_16F, ldb, &fbeta, c, CUDA_R_16F, ldc, CUDA_R_32F, CUBLAS_GEMM_DEFAULT_TENSOR_OP)`",https://github.com/pytorch/pytorch/issues/130253,needs reproduction,32,"needs reproduction label present; component/oncall labels: module: cublas, module: cuda",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-07-05T07:13:33Z
issue,188411,RTX 5090 cudaErrorLaunchTimeout,https://github.com/pytorch/pytorch/issues/188411,needs reproduction,32,"needs reproduction label present; component/oncall labels: module: cuda, module: cudnn, module: reductions",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-07-02T08:35:28Z
issue,187824,libcudnn_engines_tensor_ir.so.9 not included in the PyTorch based on CUDA 13.0,https://github.com/pytorch/pytorch/issues/187824,needs reproduction,32,"needs reproduction label present; component/oncall labels: module: binaries, module: cudnn",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-06-30T18:59:19Z
issue,184030,torch.nn.PoissonNLLLoss returns NaN for negative input when log_input=False,https://github.com/pytorch/pytorch/issues/184030,needs reproduction,32,"needs reproduction label present; component/oncall labels: module: edge cases, module: nn",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-06-22T13:04:06Z
issue,170138,Performance regression of depthwise Conv1d with fp16/fp32 AMP from 2.7.1 -> 2.8.0 and the issue remains in 2.9.1,https://github.com/pytorch/pytorch/issues/170138,needs reproduction,32,"needs reproduction label present; component/oncall labels: module: convolution, module: cuda, module: performance",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-06-10T08:20:07Z
issue,164342,Official support for sm_120 (RTX 50-series / Blackwell) in stable PyTorch builds,https://github.com/pytorch/pytorch/issues/164342,needs reproduction,32,"needs reproduction label present; component/oncall labels: module: cuda, module: windows",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-06-05T18:39:38Z
issue,88251,Weird random SIGTERM occurance,https://github.com/pytorch/pytorch/issues/88251,needs reproduction,32,"needs reproduction label present; component/oncall labels: module: ddp, module: elastic, oncall: distributed, oncall: distributed parallelisms",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-06-05T17:10:51Z
issue,98542,Training runs 50% slower when using 2 GPUs comparing to 1,https://github.com/pytorch/pytorch/issues/98542,needs reproduction,32,"needs reproduction label present; component/oncall labels: module: ddp, oncall: distributed, oncall: distributed parallelisms",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-06-05T17:09:47Z
issue,114861,Torchrun threads with different computing speed.,https://github.com/pytorch/pytorch/issues/114861,needs reproduction,32,"needs reproduction label present; component/oncall labels: module: ddp, oncall: distributed, oncall: distributed parallelisms",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-06-05T17:06:33Z
issue,120383,Numeric stability issue with full parameter fine tuning of Llama 2 7B model with FSDP,https://github.com/pytorch/pytorch/issues/120383,needs reproduction,32,"needs reproduction label present; component/oncall labels: module: fsdp, oncall: distributed, oncall: distributed parallelisms",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-06-05T17:05:47Z
issue,121606,"RuntimeError: Rank 2 successfully reached monitoredBarrier, but received errors while waiting for send/recv from rank 0. Please check rank 0 logs for faulty rank.",https://github.com/pytorch/pytorch/issues/121606,needs reproduction,32,"needs reproduction label present; component/oncall labels: oncall: distributed, oncall: distributed infra",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-06-05T17:05:06Z
issue,121944,raise RendezvousConnectionError( torch.distributed.elastic.rendezvous.api.RendezvousConnectionError: The connection to the C10d store has failed. See inner exception for details.,https://github.com/pytorch/pytorch/issues/121944,needs reproduction,32,"needs reproduction label present; component/oncall labels: oncall: distributed, oncall: distributed infra",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-06-05T17:05:01Z
issue,139579,There is a bug when training SwinIR model in pychram and running it,https://github.com/pytorch/pytorch/issues/139579,needs reproduction,32,"needs reproduction label present; component/oncall labels: oncall: distributed, oncall: distributed infra",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-06-05T17:02:04Z
issue,158724,Problem with RPC on dual stack networks,https://github.com/pytorch/pytorch/issues/158724,needs reproduction,32,"needs reproduction label present; component/oncall labels: module: rpc, oncall: distributed, oncall: distributed infra",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-06-05T16:57:10Z
issue,165959,"torch.distributed.DistStoreError: use_libuv was requested but PyTorch was built without libuv support, run with USE_LIBUV=0 to disable it.",https://github.com/pytorch/pytorch/issues/165959,needs reproduction,32,"needs reproduction label present; component/oncall labels: oncall: distributed, oncall: distributed infra",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-06-05T16:54:11Z
issue,171499,bug in torch.stack a long 5d video and yields a misordered video output,https://github.com/pytorch/pytorch/issues/171499,needs reproduction,32,needs reproduction label present; component/oncall labels: module: cuda,confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-05-19T15:37:45Z
issue,138422,vmap with torch.autograd.grad does not work on output of compiled function,https://github.com/pytorch/pytorch/issues/138422,needs reproduction,32,"needs reproduction label present; component/oncall labels: module: functorch, module: pt2-dispatcher, module: vmap, oncall: pt2",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-05-15T22:49:16Z
issue,148811,Custom Triton operator not registered when loading AOT-compiled .so in C++ environment,https://github.com/pytorch/pytorch/issues/148811,needs reproduction,32,"needs reproduction label present; component/oncall labels: module: aotdispatch, module: custom-operators, module: pt2-dispatcher, oncall: pt2",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-05-14T22:00:31Z
issue,147115,Unable to print in a branch run by torch.cond,https://github.com/pytorch/pytorch/issues/147115,needs reproduction,32,"needs reproduction label present; component/oncall labels: module: higher order operators, module: pt2-dispatcher, oncall: pt2",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-05-14T19:06:16Z
issue,181901,[Inductor][NVGEMM] tile_M<128 EFC kernels broadcast incorrectly,https://github.com/pytorch/pytorch/issues/181901,needs reproduction,32,"needs reproduction label present; component/oncall labels: module: correctness (silent), module: dependency bug, module: inductor, oncall: pt2",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-05-14T17:14:36Z
issue,141545,Missing Fake/Meta/Abstract Kernels for Operators,https://github.com/pytorch/pytorch/issues/141545,needs reproduction,32,"needs reproduction label present; component/oncall labels: module: faketensor, module: pt2-dispatcher, oncall: pt2",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-05-14T15:46:36Z
issue,146994,subclasses + HOPs fail with `Attempting to use FunctionalTensor on its own`,https://github.com/pytorch/pytorch/issues/146994,needs reproduction,32,"needs reproduction label present; component/oncall labels: module: aotdispatch, module: pt2-dispatcher, oncall: pt2",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-05-14T03:49:00Z
issue,145451,[Dynamo]while_loop raise an exception,https://github.com/pytorch/pytorch/issues/145451,needs reproduction,32,"needs reproduction label present; component/oncall labels: module: higher order operators, module: pt2-dispatcher, oncall: pt2",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-05-14T03:46:44Z
issue,143579,`assert_size_stride` failing in `_inductor/utils.py` `return model(new_inputs)`,https://github.com/pytorch/pytorch/issues/143579,needs reproduction,32,"needs reproduction label present; component/oncall labels: module: custom-operators, module: pt2-dispatcher, oncall: pt2",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-05-13T23:41:36Z
issue,145782,Behavior Difference with Flex Attention + Sequence Packing,https://github.com/pytorch/pytorch/issues/145782,needs reproduction,32,"needs reproduction label present; component/oncall labels: module: flex attention, module: higher order operators, module: pt2-dispatcher, oncall: pt2",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-05-13T23:37:38Z
issue,144803,Something is fishy with discard_graph_changes,https://github.com/pytorch/pytorch/issues/144803,needs reproduction,32,"needs reproduction label present; component/oncall labels: module: higher order operators, module: pt2-dispatcher, oncall: pt2",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-05-13T04:03:54Z
issue,182760,"[Build] UnicodeDecodeError in cmake_utils.py due to non-standard library version suffixes (e.g., OpenSSL 1.1.1wa)",https://github.com/pytorch/pytorch/issues/182760,needs reproduction,32,needs reproduction label present; component/oncall labels: module: build,confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-05-11T20:21:35Z
issue,173389,Inference latency increase on ARM after upgrading from PyTorch 1.13.1 to 2.x,https://github.com/pytorch/pytorch/issues/173389,needs reproduction,32,"needs reproduction label present; component/oncall labels: module: arm, module: performance",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-05-05T12:40:22Z
issue,159207,Add official support for CUDA sm_120 (RTX 5090 / Blackwell architecture),https://github.com/pytorch/pytorch/issues/159207,needs reproduction,32,needs reproduction label present; component/oncall labels: module: binaries,confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-05-01T11:40:06Z
issue,136344,Composite Compliance Tensor ut check fails in privateuse1 device,https://github.com/pytorch/pytorch/issues/136344,needs reproduction,32,needs reproduction label present; component/oncall labels: module: privateuse1,confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-04-28T04:15:35Z
issue,177020,tensor cuda memory is not released when using nested_tensor_from_jagged,https://github.com/pytorch/pytorch/issues/177020,needs reproduction,32,"needs reproduction label present; component/oncall labels: module: cuda, module: memory usage, module: python frontend, module: python version",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-04-22T14:42:29Z
issue,162957,torch.linalg.eigh uses a large amount of memory in pytorch 2.8.0,https://github.com/pytorch/pytorch/issues/162957,needs reproduction,32,"needs reproduction label present; component/oncall labels: module: cuda, module: linear algebra, module: memory usage",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-04-17T07:28:54Z
issue,177819,"MPS unavailable on macOS 26.3.1 arm64, and `device=""mps""` reports incorrect macOS requirement",https://github.com/pytorch/pytorch/issues/177819,needs reproduction,32,"needs reproduction label present; component/oncall labels: module: collect_env.py, module: macos",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-04-16T06:26:56Z
issue,175057,torch.compile / Inductor generates invalid C++ (undeclared zuf0/zuf1) during compile (CppCompileError) on RTX 3090,https://github.com/pytorch/pytorch/issues/175057,needs reproduction,32,"needs reproduction label present; component/oncall labels: module: dynamic shapes, module: inductor, oncall: pt2",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-04-16T05:31:36Z
issue,179904,PyTorch 2.6 weights_only=True breaks loading of existing checkpoints,https://github.com/pytorch/pytorch/issues/179904,needs reproduction,32,needs reproduction label present; component/oncall labels: module: serialization,confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-04-13T16:40:16Z
issue,177028,tensor cuda memory cannot be release by gc when a tensor hold ref of another,https://github.com/pytorch/pytorch/issues/177028,needs reproduction,32,"needs reproduction label present; component/oncall labels: module: cuda, module: memory usage",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-03-31T08:01:55Z
issue,178145,"Large fp32 discrepancy in LeNet5Fashion conv2 output on CUDA eager baseline, with similar error against both CUDA compile and CPU eager",https://github.com/pytorch/pytorch/issues/178145,needs reproduction,32,needs reproduction label present,confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-03-25T03:36:08Z
issue,178251,RuntimeError: Function 'ScaledDotProductEfficientAttentionBackward0' returned nan values in its 0th output when using attn_mask in SDPA backward,https://github.com/pytorch/pytorch/issues/178251,needs reproduction,32,"needs reproduction label present; component/oncall labels: module: autograd, module: cuda, module: sdpa",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-03-24T19:23:26Z
issue,178049,print is repeated sometimes when the calc is done on gpu,https://github.com/pytorch/pytorch/issues/178049,needs reproduction,32,needs reproduction label present,confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-03-24T19:15:51Z
issue,177738,C++ Can't capture Transformer CUDA Graph,https://github.com/pytorch/pytorch/issues/177738,needs reproduction,32,"needs reproduction label present; component/oncall labels: module: cpp, module: cuda graphs, module: nn",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-03-24T19:10:21Z
issue,126654,torch.nn.functional.scaled_dot_product_attention returns NaN values after backward pass.,https://github.com/pytorch/pytorch/issues/126654,needs reproduction,32,"needs reproduction label present; component/oncall labels: module: numerical-stability, module: sdpa",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-03-24T07:12:40Z
issue,133548,Floating point exception on H20,https://github.com/pytorch/pytorch/issues/133548,needs reproduction,32,needs reproduction label present; component/oncall labels: module: cuda,confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-03-19T08:51:28Z
issue,135196,ModuleNotFoundError: No module named 'torch.fx.experimental.shape_inference',https://github.com/pytorch/pytorch/issues/135196,needs reproduction,32,needs reproduction label present; component/oncall labels: module: fx,confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-03-18T23:59:48Z
issue,97635,JAX + PyTorch produces `OMP: Error #13: Assertion failure at kmp_affinity.cpp(532)`,https://github.com/pytorch/pytorch/issues/97635,needs reproduction,32,needs reproduction label present,confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-03-11T14:54:54Z
issue,171672,torch.compile max-autotune and reduce-overhead much slower than default due to repeated CUDA graph instantiation,https://github.com/pytorch/pytorch/issues/171672,needs reproduction,32,"needs reproduction label present; component/oncall labels: module: cuda graphs, oncall: pt2",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-03-10T22:10:29Z
issue,165532,PyTorch CPU / DataLoader: “Too many open files” (`EMFILE`),https://github.com/pytorch/pytorch/issues/165532,needs reproduction,32,needs reproduction label present; component/oncall labels: module: dataloader,confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-03-09T18:36:57Z
issue,166633,"Command '['ninja', '-v']' returned non-zero exit status 255.",https://github.com/pytorch/pytorch/issues/166633,needs reproduction,32,"needs reproduction label present; component/oncall labels: module: cpp-extensions, module: cuda",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-03-07T13:16:23Z
issue,175637,Conv2d output inconsistency across different execution modes (device/dtype/optimization),https://github.com/pytorch/pytorch/issues/175637,needs reproduction,32,"needs reproduction label present; component/oncall labels: module: convolution, module: nn",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-03-07T01:41:30Z
issue,118092,torch.load with weights_only=True show error for pickle protocol 4,https://github.com/pytorch/pytorch/issues/118092,needs reproduction,32,needs reproduction label present; component/oncall labels: module: serialization,confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-03-05T06:49:24Z
issue,173315,[cuda] Numerical Inconsistency: NaN in Eager CUDA vs Valid Output in Compiled CPU for nn.LSTM,https://github.com/pytorch/pytorch/issues/173315,needs reproduction,32,"needs reproduction label present; component/oncall labels: module: cuda, module: edge cases, module: nans and infs, module: rnn",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-03-03T16:15:33Z
issue,173476,[Hang] Unrecoverable CUDA deadlock in torch.ops.aten.lstm - process becomes unresponsive,https://github.com/pytorch/pytorch/issues/173476,needs reproduction,32,"needs reproduction label present; component/oncall labels: module: cuda, module: rnn",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-03-03T06:51:02Z
issue,125954,Error: command buffer exited with error status.,https://github.com/pytorch/pytorch/issues/125954,needs reproduction,32,"needs reproduction label present; component/oncall labels: module: intel, module: macos, module: mps",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-03-02T08:37:19Z
issue,173922,[Numerical Accuracy] 100% relative error in LSTM output between CUDA and CPU backends in eval mode,https://github.com/pytorch/pytorch/issues/173922,needs reproduction,32,"needs reproduction label present; component/oncall labels: module: cpu, module: cuda, module: rnn",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-02-28T06:40:08Z
issue,174010,Numerical divergence in GPT2 wpe layer: 2.19e-4 relative error between float32 and float16 on CUDA,https://github.com/pytorch/pytorch/issues/174010,needs reproduction,32,"needs reproduction label present; component/oncall labels: module: cuda, module: embedding, module: half, module: numerical-stability",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-02-28T06:38:47Z
issue,153172,triggered internal assert in matmul,https://github.com/pytorch/pytorch/issues/153172,needs reproduction,32,"needs reproduction label present; component/oncall labels: module: cuda, module: cudacachingallocator",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-02-23T06:54:04Z
issue,167414,C10_CUDA_KERNEL_LAUNCH_CHECK uses getLastError without clearing the error.,https://github.com/pytorch/pytorch/issues/167414,needs reproduction,32,"needs reproduction label present; component/oncall labels: module: error checking, module: rocm",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-02-12T17:27:28Z
issue,108041,"Undefined Symobl: pybind11::detail::type_caster<at::Tensor, void>::load(pybind11::handle, bool)",https://github.com/pytorch/pytorch/issues/108041,needs reproduction,32,needs reproduction label present; component/oncall labels: module: pybind,confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-02-10T15:20:27Z
issue,170676,"CUBLAS_STATUS_INVALID_VALUE at F.linear((14,4096), (4096,4096)) device='cuda:0', dtype=torch.float16 on 4090 and B300",https://github.com/pytorch/pytorch/issues/170676,needs reproduction,32,"needs reproduction label present; component/oncall labels: module: cuda, module: linear algebra",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-02-05T20:15:09Z
issue,172217,test_numba_integration consistently fails with SIGSEGV on CUDA-enabled environments (Ubuntu 22.04 & RHEL 9),https://github.com/pytorch/pytorch/issues/172217,needs reproduction,32,"needs reproduction label present; component/oncall labels: module: ci, module: third_party",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-02-02T03:15:46Z
issue,137001,BCE loss mps device,https://github.com/pytorch/pytorch/issues/137001,needs reproduction,32,"needs reproduction label present; component/oncall labels: module: correctness (silent), module: mps",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-01-30T23:34:08Z
issue,173066,Training accuracy may suddenly fall to near 0 in the early training stage using pytorch 2.9 and AMP (bfloat16),https://github.com/pytorch/pytorch/issues/173066,needs reproduction,32,needs reproduction label present; component/oncall labels: module: optimizer,confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-01-22T23:56:18Z
issue,172944,Tried to access the schema for which doesn't have a schema registered yet,https://github.com/pytorch/pytorch/issues/172944,needs reproduction,32,needs reproduction label present; component/oncall labels: module: dispatch,confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-01-22T04:02:54Z
issue,172547,Cold start models with compile-cache have significant overhead,https://github.com/pytorch/pytorch/issues/172547,needs reproduction,32,needs reproduction label present; component/oncall labels: oncall: pt2,confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-01-18T09:34:56Z
issue,172628,pip download torch report nvidia_nccl_cu11 checksum mismatch,https://github.com/pytorch/pytorch/issues/172628,needs reproduction,32,needs reproduction label present; component/oncall labels: module: binaries,confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-01-17T01:39:47Z
issue,168897,Can't install on python 3.13,https://github.com/pytorch/pytorch/issues/168897,needs reproduction,32,needs reproduction label present; component/oncall labels: module: binaries,confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-01-15T18:19:34Z
issue,111182,Training iresnet with torch.compile is slower than eager mode for torch 2.1.0,https://github.com/pytorch/pytorch/issues/111182,needs reproduction,32,"needs reproduction label present; needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: inductor, module: performance, oncall: pt2",route to design/research decision before patch planning,2024-11-09T02:17:39Z
pr,181926,Enable XPU for sparse_sampled_addmm,https://github.com/pytorch/pytorch/pull/181926,draft/noise,31,PR is draft; review decision requests changes,wait for contributor update,2026-07-08T20:47:00Z
pr,187165,[ROCm] Enable FMA inductor lowering on ROCm,https://github.com/pytorch/pytorch/pull/187165,draft/noise,31,"PR is draft; review decision requests changes; CI lane labels: ciflow/inductor, ciflow/inductor-perf-test-nightly-rocm-mi350, ciflow/rocm-mi200, ciflow/rocm-mi300",wait for contributor update,2026-07-07T17:14:35Z
pr,176726,Fix Muon momentum/update numerics for XPU state_dict parity,https://github.com/pytorch/pytorch/pull/176726,draft/noise,31,PR is draft; review decision requests changes,wait for contributor update,2026-07-06T15:46:35Z
pr,184372,[Draft] Remove ACL,https://github.com/pytorch/pytorch/pull/184372,draft/noise,31,"PR is draft; review decision requests changes; CI lane labels: ciflow/docker, ciflow/inductor, ciflow/torchtitan, ciflow/trunk",wait for contributor update,2026-07-03T11:46:42Z
pr,188242,[ROCm] Support USE_ASAN=1 build,https://github.com/pytorch/pytorch/pull/188242,draft/noise,31,PR is draft; review decision requests changes; CI lane labels: ciflow/rocm-mi300,wait for contributor update,2026-07-01T20:52:52Z
pr,182109,[MPS] Replace MPSGraph LayerNorm backward with native Metal kernels,https://github.com/pytorch/pytorch/pull/182109,draft/noise,31,PR is draft; review decision requests changes; CI lane labels: ciflow/mps,wait for contributor update,2026-06-21T22:49:50Z
pr,133753,"Keep mkldnn_rnn_layer, add onednn_rnn_layer",https://github.com/pytorch/pytorch/pull/133753,draft/noise,31,"PR is draft; review decision requests changes; CI lane labels: ciflow/inductor, ciflow/linux-aarch64, ciflow/torchtitan, ciflow/trunk",wait for contributor update,2026-06-18T15:22:00Z
pr,133667,Update non user APIs - 2,https://github.com/pytorch/pytorch/pull/133667,draft/noise,31,"PR is draft; review decision requests changes; CI lane labels: ciflow/inductor, ciflow/linux-aarch64, ciflow/mps, ciflow/torchtitan",wait for contributor update,2026-06-18T15:21:56Z
pr,133423,"Keep torch.backends.mkldnn, add torch.backends.onednn",https://github.com/pytorch/pytorch/pull/133423,draft/noise,31,"PR is draft; review decision requests changes; CI lane labels: ciflow/inductor, ciflow/linux-aarch64, ciflow/mps, ciflow/torchtitan",wait for contributor update,2026-06-18T15:21:52Z
issue,180948,DISABLED test_emulate_precision_casts_preserves_explicit_precision_cast_float16 (__main__.CudaReproTests),https://github.com/pytorch/pytorch/issues/180948,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: inductor, module: tests, module: xpu, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-07-09T01:40:51Z
issue,180194,DISABLED test_vjp_grad_cpu (__main__.TestComposabilityCPU),https://github.com/pytorch/pytorch/issues/180194,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, module: functorch, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-07-09T01:16:58Z
issue,174507,PyTorch Stable ABI support,https://github.com/pytorch/pytorch/issues/174507,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: abi,"decide whether this is bug, use case, docs, or support",2026-07-09T00:24:27Z
issue,189164,[CI][B200] B200 Runner Maintenance Notification,https://github.com/pytorch/pytorch/issues/189164,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: ci, module: infra","decide whether this is bug, use case, docs, or support",2026-07-09T00:23:26Z
issue,188898,DISABLED test_vmap_over_vmap_two_inputs (__main__.FuncTorchHigherOrderOpTestsWithCompiledAutograd),https://github.com/pytorch/pytorch/issues/188898,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-07-08T19:02:54Z
issue,189279,[CPU] operator_benchmark: eager transcendental unary (cosh/sinh/atan/log2/log10) ~3-5x slower since ~Nov 2025,https://github.com/pytorch/pytorch/issues/189279,needs maintainer decision,28,open issue has no stronger automated bucket,"decide whether this is bug, use case, docs, or support",2026-07-08T17:30:21Z
issue,189262,DISABLED rocm-nightly / linux-noble-rocm-nightly-py3.12-gfx942 / test,https://github.com/pytorch/pytorch/issues/189262,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: ci, module: rocm","decide whether this is bug, use case, docs, or support",2026-07-08T17:29:46Z
issue,167822,[RFD] Composable Kernel (CK) moving to C++20 — request for feedback/concerns,https://github.com/pytorch/pytorch/issues/167822,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: build,"decide whether this is bug, use case, docs, or support",2026-07-08T17:25:42Z
issue,188756,"[MPS] index_select / F.embedding silently return all-zero rows when the source tensor exceeds 2**31 elements (fixed on main, still broken in release/2.13)",https://github.com/pytorch/pytorch/issues/188756,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: 64-bit, module: correctness (silent), module: embedding, module: mps","decide whether this is bug, use case, docs, or support",2026-07-08T17:18:22Z
issue,189270,[inductor] Emit per-kernel physical op structure (with rematerialization) in provenance,https://github.com/pytorch/pytorch/issues/189270,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: inductor, module: trace, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-07-08T15:16:17Z
issue,188496,torch.fft.irfft2 ONNX export produces graph that onnxruntime cannot execute.,https://github.com/pytorch/pytorch/issues/188496,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: fft, module: onnx","decide whether this is bug, use case, docs, or support",2026-07-08T14:48:17Z
issue,189261,DISABLED periodic-rocm-mi300 / linux-noble-rocm-py3.12-mi300 / test,https://github.com/pytorch/pytorch/issues/189261,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: ci, module: rocm","decide whether this is bug, use case, docs, or support",2026-07-08T14:28:01Z
issue,189255,DISABLED test_delayed_optim_step_offload_true_no_shard_cuda (__main__.TestParityWithDDPCUDA),https://github.com/pytorch/pytorch/issues/189255,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, module: rocm, oncall: distributed","decide whether this is bug, use case, docs, or support",2026-07-08T13:12:54Z
issue,179799,DISABLED test_grad_has_aux (__main__.FuncTorchHigherOrderOpTestsWithCompiledAutograd),https://github.com/pytorch/pytorch/issues/179799,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-07-08T13:12:35Z
issue,160740,MPS torch.argmax/torch.argmin fail for non-contiguous input,https://github.com/pytorch/pytorch/issues/160740,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: correctness (silent), module: mps","decide whether this is bug, use case, docs, or support",2026-07-08T11:16:06Z
issue,143954,"No ONNX function found for <OpOverload(op='quantized_decomposed.dequantize_per_channel', overload='default')>",https://github.com/pytorch/pytorch/issues/143954,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: onnx,"decide whether this is bug, use case, docs, or support",2026-07-08T10:40:54Z
issue,187455,[MPS] Migrate remaining MPSGraph-backed ops to native Metal to bound graph-cache memory,https://github.com/pytorch/pytorch/issues/187455,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: memory usage, module: mps, module: performance","decide whether this is bug, use case, docs, or support",2026-07-08T09:10:44Z
issue,168995,MXFP8 + torch.compile error: NotImplementedError: make_reindexer NYI on DtypeView,https://github.com/pytorch/pytorch/issues/168995,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-07-08T08:04:22Z
issue,189244,DISABLED test_cat_pointwise (__main__.NumBytesMetricTests),https://github.com/pytorch/pytorch/issues/189244,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, module: inductor, module: rocm, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-07-08T07:30:24Z
issue,189223,DISABLED test_vmap_free_tensor (__main__.FuncTorchHigherOrderOpTestsWithCompiledAutograd),https://github.com/pytorch/pytorch/issues/189223,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-07-08T07:30:08Z
issue,187289,DISABLED test_col2im_uint32_overflow (__main__.TestMPS),https://github.com/pytorch/pytorch/issues/187289,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, module: macos, module: mps","decide whether this is bug, use case, docs, or support",2026-07-08T07:30:03Z
issue,152632,DISABLED test_torchvision_models_efficientnet_v2_l (__main__.TestVisionTracing),https://github.com/pytorch/pytorch/issues/152632,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, module: fx","decide whether this is bug, use case, docs, or support",2026-07-08T07:30:01Z
issue,179879,DISABLED test_vmap_vmap_cpu (__main__.TestComposabilityCPU),https://github.com/pytorch/pytorch/issues/179879,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, module: functorch, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-07-08T07:29:55Z
issue,189236,[Feature Request] Add 'symmetric' padding_mode (repeats boundary sample),https://github.com/pytorch/pytorch/issues/189236,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: nn, module: padding","decide whether this is bug, use case, docs, or support",2026-07-08T06:22:05Z
issue,187175,[feature request] Autotune TMA and non-TMA path for pointwise and reduction kernels,https://github.com/pytorch/pytorch/issues/187175,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: cuda, module: inductor, module: performance, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-07-08T05:08:55Z
issue,163650,[DO NOT CLOSE] Autorevert actions shadow mode stream,https://github.com/pytorch/pytorch/issues/163650,needs maintainer decision,28,open issue has no stronger automated bucket,"decide whether this is bug, use case, docs, or support",2026-07-08T01:30:49Z
issue,189221,DISABLED test_nested_dict_partial_load (__main__.TestCheckpointer),https://github.com/pytorch/pytorch/issues/189221,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, module: rocm, oncall: distributed checkpointing","decide whether this is bug, use case, docs, or support",2026-07-08T01:09:24Z
issue,179742,DISABLED test_grad_freevar_python_scalar (__main__.FuncTorchHigherOrderOpTestsWithCompiledAutograd),https://github.com/pytorch/pytorch/issues/179742,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-07-08T01:09:11Z
issue,180213,DISABLED test_vmap_vjp_cpu (__main__.TestComposabilityCPU),https://github.com/pytorch/pytorch/issues/180213,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, module: functorch, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-07-08T01:09:09Z
issue,76185,Ensure custom Function are correct in double backward setting,https://github.com/pytorch/pytorch/issues/76185,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: autograd,"decide whether this is bug, use case, docs, or support",2026-07-07T23:37:07Z
issue,187073,torch.compile recompiles when input alternates between AsyncCollectiveTensor and Tensor,https://github.com/pytorch/pytorch/issues/187073,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dynamo, module: fsdp, module: guards, module: performance","decide whether this is bug, use case, docs, or support",2026-07-07T22:55:54Z
issue,189189,Subtraction between two `torch.complex64` tensors fails on modules compiled with `aoti_compile_and_package`,https://github.com/pytorch/pytorch/issues/189189,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: aotinductor, module: complex, oncall: export, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-07-07T22:04:29Z
issue,50282,Add MultiScheduler,https://github.com/pytorch/pytorch/issues/50282,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: optimizer,"decide whether this is bug, use case, docs, or support",2026-07-07T19:02:00Z
issue,189161,[Inductor] Remove Redundant NaN Propagation in `max` op in Softmax,https://github.com/pytorch/pytorch/issues/189161,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: inductor, module: performance, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-07-07T15:30:19Z
issue,189151,DISABLED test_pending_fusion_pro_and_epi_use_async_compile_False (__main__.TestPrologueFusion),https://github.com/pytorch/pytorch/issues/189151,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, module: inductor, module: rocm, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-07-07T13:25:32Z
issue,152944,DISABLED test_compiler_collectives_automatic_dynamic_tensor (__main__.TestMultiProc),https://github.com/pytorch/pytorch/issues/152944,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dynamic shapes, module: dynamo, module: flaky-tests, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-07-07T13:25:25Z
issue,186865,DISABLED test_dynamic_shape_backend_eager_device_cpu (__main__.TestPackage),https://github.com/pytorch/pytorch/issues/186865,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dynamo, module: flaky-tests, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-07-07T13:25:18Z
issue,180195,DISABLED test_vjp_vmap_cpu (__main__.TestComposabilityCPU),https://github.com/pytorch/pytorch/issues/180195,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, module: functorch, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-07-07T13:25:13Z
issue,189137,[Sub-RFC] PrivateUse1 Inductor codegen registration interfaces (RFC 3/3),https://github.com/pytorch/pytorch/issues/189137,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: aotinductor, module: backend, module: inductor, module: privateuse1","decide whether this is bug, use case, docs, or support",2026-07-07T12:01:59Z
issue,189136,[Sub-RFC] PrivateUse1 Dynamo tracing via registry splicing (RFC 2/3),https://github.com/pytorch/pytorch/issues/189136,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dynamo, module: privateuse1, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-07-07T12:01:30Z
issue,189135,[Sub-RFC] PrivateUse1 device identity/capability in torch.compile (RFC 1/3),https://github.com/pytorch/pytorch/issues/189135,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: backend, module: dynamo, module: inductor, module: privateuse1","decide whether this is bug, use case, docs, or support",2026-07-07T12:00:59Z
issue,189138,[RFC] DeviceInterface: Registry Support for PrivateUse1 Backends in torch.compile,https://github.com/pytorch/pytorch/issues/189138,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: backend, module: dynamo, module: inductor, module: privateuse1","decide whether this is bug, use case, docs, or support",2026-07-07T11:20:54Z
issue,188798,DISABLED test_linearize_jvp_fn (__main__.FuncTorchHigherOrderOpTestsWithCompiledAutograd),https://github.com/pytorch/pytorch/issues/188798,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-07-07T07:52:39Z
issue,176966,DISABLED test_trace_return_namedtuple (__main__.TestFX),https://github.com/pytorch/pytorch/issues/176966,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, module: fx, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-07-07T07:52:34Z
issue,124152,DISABLED test_torchvision_models_efficientnet_v2_m (__main__.TestVisionTracing),https://github.com/pytorch/pytorch/issues/124152,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, module: fx","decide whether this is bug, use case, docs, or support",2026-07-07T07:52:32Z
issue,188987,DISABLED test_optimizer_non_static_param (__main__.LoggingTests),https://github.com/pytorch/pytorch/issues/188987,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: logging, module: tests, module: xpu, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-07-07T06:22:00Z
issue,189046,Treat PrivateUse1 as CUDA-like under env override to reuse CUDA optimization paths in shared `at::native` operators,https://github.com/pytorch/pytorch/issues/189046,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: advanced indexing, module: autograd, module: backend, module: dispatch","decide whether this is bug, use case, docs, or support",2026-07-07T02:58:57Z
issue,188602,[RFC] Optimizing CUDA+AArch64 builds using LLVM-BOLT,https://github.com/pytorch/pytorch/issues/188602,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: arm, module: build, module: ci, module: cuda","decide whether this is bug, use case, docs, or support",2026-07-07T02:04:59Z
issue,189067,DISABLED test_pyobject_redispatch_requires_keyset (__main__.TestCustomOpWithCompiledAutograd),https://github.com/pytorch/pytorch/issues/189067,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-07-07T01:15:59Z
issue,180337,DISABLED test_vmap_grad_cpu (__main__.TestComposabilityCPU),https://github.com/pytorch/pytorch/issues/180337,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, module: functorch, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-07-07T01:15:46Z
issue,188239,DISABLED test_comprehensive_linalg_polar_xpu_float32 (__main__.TestInductorOpInfoXPU),https://github.com/pytorch/pytorch/issues/188239,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: inductor, module: linear algebra, module: xpu, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-07-07T01:05:29Z
issue,189078,[RFC] Build profiling for PyTorch,https://github.com/pytorch/pytorch/issues/189078,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: build, module: performance","decide whether this is bug, use case, docs, or support",2026-07-06T21:57:17Z
issue,189093,Misleading full-backward-hook warning when only `register_full_backward_pre_hook` is used,https://github.com/pytorch/pytorch/issues/189093,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: autograd, module: nn","decide whether this is bug, use case, docs, or support",2026-07-06T21:25:42Z
issue,188669,fx_graph_cse skipping duplicate element-wise CUDA optimization.,https://github.com/pytorch/pytorch/issues/188669,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: fx.passes, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-07-06T19:44:16Z
issue,189077,operator_microbenchmark + attention_op_microbenchmark: b200 jobs never complete -- run ~8h and are cancelled at the workflow timeout,https://github.com/pytorch/pytorch/issues/189077,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: benchmark, module: ci, module: compile-time, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-07-06T19:39:16Z
issue,189075,periodic cuda13.0 slow-gradcheck: default test shards time out (~270 min) before finishing -- most shards killed,https://github.com/pytorch/pytorch/issues/189075,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: ci,"decide whether this is bug, use case, docs, or support",2026-07-06T19:27:49Z
issue,189066,DISABLED test_jacrev (__main__.FuncTorchHigherOrderOpTestsWithCompiledAutograd),https://github.com/pytorch/pytorch/issues/189066,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-07-06T19:15:07Z
issue,140556,DISABLED test_backward_prod_cuda_float32 (__main__.TestNestedTensorOpInfoCUDA),https://github.com/pytorch/pytorch/issues/140556,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, module: nestedtensor","decide whether this is bug, use case, docs, or support",2026-07-06T19:11:19Z
issue,188564,rocm-navi31 gfx1100 test (default) jobs time out after 270 min on inductor/test_torchinductor (red since 2026-05-20),https://github.com/pytorch/pytorch/issues/188564,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: ci, module: compile-time, module: convolution, module: inductor","decide whether this is bug, use case, docs, or support",2026-07-06T17:19:11Z
issue,188382,torch.special.logit with eps > 0.5 has tiny vectorized-vs-scalar exact parity mismatch on non-contiguous float32 input,https://github.com/pytorch/pytorch/issues/188382,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: edge cases, module: special, module: vectorization, module: windows","decide whether this is bug, use case, docs, or support",2026-07-06T16:59:45Z
issue,189032,nightly / update-metamates-rule failing since ~2026-06-29 — UPDATEBOT_TOKEN can't read the pytorch/metamates team,https://github.com/pytorch/pytorch/issues/189032,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: ci,"decide whether this is bug, use case, docs, or support",2026-07-06T15:34:13Z
issue,120308,ROCm: foreach is substantially slower than for loop MI250x,https://github.com/pytorch/pytorch/issues/120308,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: performance, module: rocm","decide whether this is bug, use case, docs, or support",2026-07-06T15:13:23Z
issue,188983,DISABLED test_comprehensive_combinations_cuda_float16 (__main__.TestInductorOpInfoCUDA),https://github.com/pytorch/pytorch/issues/188983,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-07-06T13:57:24Z
issue,189015,DISABLED test_jacfwd_two_tensors_argnums (__main__.FuncTorchHigherOrderOpTestsWithCompiledAutograd),https://github.com/pytorch/pytorch/issues/189015,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-07-06T13:57:21Z
issue,188705,DISABLED test_reduction_not_fused (__main__.TestNVUniversalGemmEpilogueFusion),https://github.com/pytorch/pytorch/issues/188705,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: cuda, module: inductor, module: tests, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-07-06T10:24:27Z
issue,188677,DISABLED test_graph_partition_view_fallback (__main__.CudaGraphTreeTests),https://github.com/pytorch/pytorch/issues/188677,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: cuda graphs, module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-07-06T10:19:00Z
issue,89431,The problem caused by the parameter dim of torch.norm,https://github.com/pytorch/pytorch/issues/89431,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: edge cases, module: error checking","decide whether this is bug, use case, docs, or support",2026-07-06T09:25:16Z
issue,140718,'lobcpg' gives an output on CPU but fails on GPU('mps'),https://github.com/pytorch/pytorch/issues/140718,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: linear algebra, module: mps","decide whether this is bug, use case, docs, or support",2026-07-06T07:21:35Z
issue,171659,[RFC] RISC-V Architecture Support Roadmap for PyTorch,https://github.com/pytorch/pytorch/issues/171659,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: build, module: risc-v","decide whether this is bug, use case, docs, or support",2026-07-06T05:41:43Z
issue,188953,"Intel Arc B580 (Battlemage): ""could not create an engine"" on XPU — torch 2.6.0+xpu",https://github.com/pytorch/pytorch/issues/188953,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dependency bug, module: linear algebra, module: xpu","decide whether this is bug, use case, docs, or support",2026-07-06T05:08:42Z
issue,188985,DISABLED test_output_grad_match_combinations_mps_float16 (__main__.TestConsistencyMPS),https://github.com/pytorch/pytorch/issues/188985,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, module: macos, module: mps","decide whether this is bug, use case, docs, or support",2026-07-06T01:15:06Z
issue,188041,DISABLED test_jacfwd_randomness (__main__.FuncTorchHigherOrderOpTestsWithCompiledAutograd),https://github.com/pytorch/pytorch/issues/188041,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-07-06T01:14:57Z
issue,175891,[RFC] torch.switch,https://github.com/pytorch/pytorch/issues/175891,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: cuda graphs, module: higher order operators, module: pt2-dispatcher, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-07-05T21:09:29Z
issue,188926,[CPU] torch.unravel_index raises ZeroDivisionError when shape has a zero dimension,https://github.com/pytorch/pytorch/issues/188926,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: cpu, module: edge cases, module: empty tensor, module: error checking","decide whether this is bug, use case, docs, or support",2026-07-05T05:26:05Z
issue,188319,[RFC] Publish official ROCm Docker images on pytorch/pytorch,https://github.com/pytorch/pytorch/issues/188319,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: binaries, module: ci, module: docker, module: rocm","decide whether this is bug, use case, docs, or support",2026-07-04T16:30:49Z
issue,68506,cudnn_convolution_relu is 17x slower for 7x7 convolutions and channels last,https://github.com/pytorch/pytorch/issues/68506,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: cuda, module: cudnn","decide whether this is bug, use case, docs, or support",2026-07-04T08:23:51Z
issue,188935,find_library(ROCM_ROCTX_LIB roctx64 ...) should be gated behind USE_ROCTRACER,https://github.com/pytorch/pytorch/issues/188935,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: build, module: rocm","decide whether this is bug, use case, docs, or support",2026-07-04T00:51:38Z
issue,187738,DISABLED test_grad_two_tensor_has_aux (__main__.FuncTorchHigherOrderOpTestsWithCompiledAutograd),https://github.com/pytorch/pytorch/issues/187738,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-07-03T20:01:08Z
issue,188899,DISABLED test_cache_hot_load_caching_precompile_dynamic_True_device_cuda_bfloat16 (__main__.TestFxGraphCache),https://github.com/pytorch/pytorch/issues/188899,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, module: inductor, module: rocm, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-07-03T13:18:10Z
issue,188897,DISABLED test_addmm_relu_tunableop_cuda_float32 (__main__.TestLinalgCudaOnlyCUDA),https://github.com/pytorch/pytorch/issues/188897,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, module: linear algebra, module: rocm","decide whether this is bug, use case, docs, or support",2026-07-03T13:18:03Z
issue,188895,DISABLED test_conv2d_backward_parametrized_dilation_2_stride_1_padding_1_kernel_3_channels_groups4_nhwc_True_cuda (__main__.GPUTests),https://github.com/pytorch/pytorch/issues/188895,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, module: inductor, module: rocm, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-07-03T13:17:59Z
issue,187718,DISABLED test_grad_two_tensor_all_grad_has_aux (__main__.FuncTorchHigherOrderOpTestsWithCompiledAutograd),https://github.com/pytorch/pytorch/issues/187718,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-07-03T13:17:52Z
issue,159113,DISABLED test_graph_two_successive (__main__.TestCuda),https://github.com/pytorch/pytorch/issues/159113,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: cuda, module: flaky-tests","decide whether this is bug, use case, docs, or support",2026-07-03T13:17:49Z
issue,188670,[RFC] Inductor / AOTI tests: decouple “correct codegen path” from tight eager-fp16 equality,https://github.com/pytorch/pytorch/issues/188670,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: aotinductor, module: inductor, module: tests, oncall: export","decide whether this is bug, use case, docs, or support",2026-07-03T12:31:22Z
issue,187658,DISABLED test_grad_non_tensor_input (__main__.FuncTorchHigherOrderOpTestsWithCompiledAutograd),https://github.com/pytorch/pytorch/issues/187658,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-07-03T07:45:47Z
issue,180414,DISABLED test_torchvision_models_efficientnet_b2 (__main__.TestVisionTracing),https://github.com/pytorch/pytorch/issues/180414,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, module: fx","decide whether this is bug, use case, docs, or support",2026-07-03T07:45:40Z
issue,181629,DISABLED test_torchvision_models_efficientnet_b5 (__main__.TestVisionTracing),https://github.com/pytorch/pytorch/issues/181629,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, module: fx","decide whether this is bug, use case, docs, or support",2026-07-03T07:45:39Z
issue,188871,[FlexAttention] FLASH backend: cute.where Int64/Int32 dtype mismatch in captured-buffer negative-index wrap under dynamic shapes,https://github.com/pytorch/pytorch/issues/188871,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dynamic shapes, module: flex attention, module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-07-03T03:47:46Z
issue,188851,[CI][ROCm] DebertaV2ForMaskedLM fails to import from transformers on gfx950 (dynamic_inductor_huggingface eager_fail_to_run),https://github.com/pytorch/pytorch/issues/188851,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: benchmark, module: ci, module: dependency bug, module: rocm","decide whether this is bug, use case, docs, or support",2026-07-03T00:20:40Z
issue,188648,"treespec_dumps does not preserve tuple keys in dict/OrderedDict/defaultdict pytrees, so the round-trip fails with `unhashable type: 'list'`",https://github.com/pytorch/pytorch/issues/188648,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: pytree, module: serialization","decide whether this is bug, use case, docs, or support",2026-07-02T21:43:19Z
issue,188840,DISABLED test_eager_equivalence_nn_functional_gelu_backend_inductor_default_cuda_float32 (__main__.TestOpInfoPropertiesCUDA),https://github.com/pytorch/pytorch/issues/188840,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: inductor, module: nn, module: tests, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-07-02T19:43:18Z
issue,188837,DISABLED test_dropout_parity_cudagraphs_reduce_overhead (__main__.TestDropoutAlignRandomEager),https://github.com/pytorch/pytorch/issues/188837,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: cuda graphs, module: inductor, module: tests, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-07-02T19:24:35Z
issue,179979,DISABLED test_sparse_tensor_multiprocessing_context_spawn_cuda (__main__.TestDataLoaderDeviceTypeCUDA),https://github.com/pytorch/pytorch/issues/179979,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dataloader, module: multiprocessing, module: sparse","decide whether this is bug, use case, docs, or support",2026-07-02T19:05:11Z
issue,187425,DISABLED test_grad_closure_scalar (__main__.FuncTorchHigherOrderOpTestsWithCompiledAutograd),https://github.com/pytorch/pytorch/issues/187425,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-07-02T19:02:55Z
issue,179798,DISABLED test_memory_plots_free_segment_stack (__main__.TestCudaAllocator),https://github.com/pytorch/pytorch/issues/179798,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: cuda, module: flaky-tests","decide whether this is bug, use case, docs, or support",2026-07-02T19:02:53Z
issue,188676,DISABLED test_float8_e4m3fn_uint8_decode_codegen_cuda (__main__.TestFP8TypesCUDA),https://github.com/pytorch/pytorch/issues/188676,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: floatx (formerly float8), module: inductor, module: tests, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-07-02T18:48:43Z
issue,153236,DISABLED test_clip_grad_norm_2d (__main__.TestClipGradNormWorldSize4),https://github.com/pytorch/pytorch/issues/153236,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, module: fsdp, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-07-02T16:41:08Z
issue,168168,[CD][ARM][CUDA] PYTORCH Released Docker Images Should Ideally Support Multi-ARCH But Currently It is only x86_64,https://github.com/pytorch/pytorch/issues/168168,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: arm, module: cuda","decide whether this is bug, use case, docs, or support",2026-07-02T15:22:10Z
issue,188727,torch.distributed.pipelining _split_block_mask hides batch offset from non-strict tracing,https://github.com/pytorch/pytorch/issues/188727,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: pipelining, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-07-02T06:30:59Z
issue,179152,`linalg_cholesky_ex_out` contains some dubious logic,https://github.com/pytorch/pytorch/issues/179152,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: cuda, module: linear algebra","decide whether this is bug, use case, docs, or support",2026-07-02T03:17:36Z
issue,188767,DISABLED test_reconfigure_scale_down_up (__main__.GlooFaultToleranceTest),https://github.com/pytorch/pytorch/issues/188767,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, module: rocm, oncall: distributed","decide whether this is bug, use case, docs, or support",2026-07-02T01:24:57Z
issue,188707,DISABLED test_main_loop_scaling_shape0_use_fast_accum_False_scaling_block_sizes1_cuda (__main__.TestFP8LoweringCUDA),https://github.com/pytorch/pytorch/issues/188707,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: cuda,"decide whether this is bug, use case, docs, or support",2026-07-02T00:34:16Z
issue,188706,DISABLED test_main_loop_scaling_shape0_use_fast_accum_False_scaling_block_sizes0_cuda (__main__.TestFP8LoweringCUDA),https://github.com/pytorch/pytorch/issues/188706,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: cuda,"decide whether this is bug, use case, docs, or support",2026-07-02T00:34:16Z
issue,188704,DISABLED test_main_loop_scaling_shape2_use_fast_accum_True_scaling_block_sizes1_cuda (__main__.TestFP8LoweringCUDA),https://github.com/pytorch/pytorch/issues/188704,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: cuda,"decide whether this is bug, use case, docs, or support",2026-07-02T00:34:15Z
issue,188740,DISABLED test_linear_block_compile_parity_backward (__main__.TestDropoutAlignRandomEager),https://github.com/pytorch/pytorch/issues/188740,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: inductor, module: tests, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-07-01T22:06:26Z
issue,188721,UNSTABLE B200 Smoke Tests / linux-jammy-cuda13.0-py3.12-gcc11-sm100 / test-osdc (smoke_b200),https://github.com/pytorch/pytorch/issues/188721,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: ci,"decide whether this is bug, use case, docs, or support",2026-07-01T19:21:42Z
issue,180542,DISABLED test_vmap_two_inputs_tuple_in_dims (__main__.FuncTorchHigherOrderOpTestsWithCompiledAutograd),https://github.com/pytorch/pytorch/issues/180542,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-07-01T19:16:03Z
issue,145344,DISABLED test_cat (__main__.NumBytesMetricTests),https://github.com/pytorch/pytorch/issues/145344,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, module: inductor, module: rocm, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-07-01T19:16:02Z
issue,188714,DISABLED test_main_loop_scaling_shape1_use_fast_accum_True_scaling_block_sizes1_cuda (__main__.TestFP8LoweringCUDA),https://github.com/pytorch/pytorch/issues/188714,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: cuda, module: floatx (formerly float8), module: inductor, module: tests","decide whether this is bug, use case, docs, or support",2026-07-01T19:04:21Z
issue,188715,DISABLED test_main_loop_scaling_shape1_use_fast_accum_True_scaling_block_sizes2_cuda (__main__.TestFP8LoweringCUDA),https://github.com/pytorch/pytorch/issues/188715,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: inductor, module: tests, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-07-01T19:04:19Z
issue,188712,DISABLED test_main_loop_scaling_shape1_use_fast_accum_False_scaling_block_sizes1_cuda (__main__.TestFP8LoweringCUDA),https://github.com/pytorch/pytorch/issues/188712,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: cuda, module: floatx (formerly float8), module: inductor, module: tests","decide whether this is bug, use case, docs, or support",2026-07-01T19:04:17Z
issue,188709,DISABLED test_main_loop_scaling_shape0_use_fast_accum_True_scaling_block_sizes0_cuda (__main__.TestFP8LoweringCUDA),https://github.com/pytorch/pytorch/issues/188709,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: floatx (formerly float8), module: inductor, module: tests, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-07-01T19:04:15Z
issue,188713,DISABLED test_main_loop_scaling_shape1_use_fast_accum_False_scaling_block_sizes2_cuda (__main__.TestFP8LoweringCUDA),https://github.com/pytorch/pytorch/issues/188713,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: cuda, module: floatx (formerly float8), module: inductor, module: linear algebra","decide whether this is bug, use case, docs, or support",2026-07-01T19:04:13Z
issue,188717,DISABLED test_main_loop_scaling_shape2_use_fast_accum_False_scaling_block_sizes2_cuda (__main__.TestFP8LoweringCUDA),https://github.com/pytorch/pytorch/issues/188717,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: floatx (formerly float8), module: inductor, module: tests, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-07-01T19:04:11Z
issue,188708,DISABLED test_main_loop_scaling_shape0_use_fast_accum_False_scaling_block_sizes2_cuda (__main__.TestFP8LoweringCUDA),https://github.com/pytorch/pytorch/issues/188708,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: floatx (formerly float8), module: inductor, module: tests, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-07-01T19:03:40Z
issue,188716,DISABLED test_main_loop_scaling_shape2_use_fast_accum_False_scaling_block_sizes1_cuda (__main__.TestFP8LoweringCUDA),https://github.com/pytorch/pytorch/issues/188716,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: floatx (formerly float8), module: inductor, module: tests, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-07-01T19:03:27Z
issue,188710,DISABLED test_main_loop_scaling_shape0_use_fast_accum_True_scaling_block_sizes1_cuda (__main__.TestFP8LoweringCUDA),https://github.com/pytorch/pytorch/issues/188710,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: floatx (formerly float8), module: inductor, module: tests, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-07-01T19:03:19Z
issue,188693,DISABLED test_vjp_scalar_logging_tensor (__main__.TestAutogradFunctional),https://github.com/pytorch/pytorch/issues/188693,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: autograd, module: tests","decide whether this is bug, use case, docs, or support",2026-07-01T16:19:38Z
issue,188692,DISABLED test_mkldnn_rnn_backward_gru_bias_shape (__main__.TestMetaKernelRegistrations),https://github.com/pytorch/pytorch/issues/188692,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: meta tensors, module: mkldnn, module: rnn, module: tests","decide whether this is bug, use case, docs, or support",2026-07-01T16:16:22Z
issue,188688,DISABLED test_backend_simple (__main__.PrivateUse1BackendTest),https://github.com/pytorch/pytorch/issues/188688,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: privateuse1, module: tests","decide whether this is bug, use case, docs, or support",2026-07-01T16:14:20Z
issue,188687,DISABLED test_linear_block_compile_parity_forward (__main__.TestDropoutAlignRandomEager),https://github.com/pytorch/pytorch/issues/188687,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: inductor, module: random, module: tests, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-07-01T16:05:15Z
issue,188686,DISABLED test_redispatch_nn_functional_conv_transpose3d_cuda_complex32 (__main__.TestTorchFunctionRedispatchOpsCUDA),https://github.com/pytorch/pytorch/issues/188686,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: __torch_function__, module: complex, module: convolution, module: cuda","decide whether this is bug, use case, docs, or support",2026-07-01T16:04:51Z
issue,188684,DISABLED test_redispatch_nn_functional_conv_transpose3d_cuda_bfloat16 (__main__.TestTorchFunctionRedispatchOpsCUDA),https://github.com/pytorch/pytorch/issues/188684,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: __torch_function__, module: bfloat16, module: convolution, module: cuda","decide whether this is bug, use case, docs, or support",2026-07-01T16:04:00Z
issue,188675,DISABLED test_main_loop_scaling_shape2_use_fast_accum_True_scaling_block_sizes2_cuda (__main__.TestFP8LoweringCUDA),https://github.com/pytorch/pytorch/issues/188675,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: floatx (formerly float8), module: inductor, module: tests, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-07-01T15:21:50Z
issue,188672,[Inductor] Prefer CUDA Toolkit libdevice for Triton-backed Inductor kernels,https://github.com/pytorch/pytorch/issues/188672,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: inductor, module: performance, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-07-01T15:04:50Z
issue,188652,Pytorch can't be built in an air-gapped environment,https://github.com/pytorch/pytorch/issues/188652,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: build, module: nnpack","decide whether this is bug, use case, docs, or support",2026-07-01T14:12:10Z
issue,185534,Sub-optimal heuristics in MixOrderReduction prevent fusion,https://github.com/pytorch/pytorch/issues/185534,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: inductor, module: performance, module: reductions, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-07-01T14:11:37Z
issue,188668,DISABLED test_extern_kernel_benchmark_valid_timing (__main__.TestExternKernelCaller),https://github.com/pytorch/pytorch/issues/188668,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, module: inductor, module: rocm, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-07-01T13:32:58Z
issue,156818,Tighter invariants for the inductor graph,https://github.com/pytorch/pytorch/issues/156818,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-07-01T08:06:35Z
issue,187378,DISABLED test_grad_capture_tensor (__main__.FuncTorchHigherOrderOpTestsWithCompiledAutograd),https://github.com/pytorch/pytorch/issues/187378,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-07-01T08:03:48Z
issue,180490,DISABLED test_vmap_two_inputs (__main__.FuncTorchHigherOrderOpTestsWithCompiledAutograd),https://github.com/pytorch/pytorch/issues/180490,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-07-01T08:03:41Z
issue,182868,[RFC] [Inductor] L2 cache aware welford/two-pass selction,https://github.com/pytorch/pytorch/issues/182868,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-07-01T03:27:25Z
issue,188420,torch.compile produces inconsistent outputs for MultiHeadAttentionModel with scaled_dot_product_attention,https://github.com/pytorch/pytorch/issues/188420,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: determinism, module: random, module: sdpa, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-07-01T03:02:28Z
issue,117974,RuntimeError: derivative for aten::_scaled_dot_product_efficient_attention_backward is not implemented,https://github.com/pytorch/pytorch/issues/117974,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: sdpa,"decide whether this is bug, use case, docs, or support",2026-07-01T01:35:30Z
issue,188628,DISABLED test_rank0_broadcast_meta_device_init (__main__.TestFullyShardMetaDeviceInit),https://github.com/pytorch/pytorch/issues/188628,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, module: rocm, oncall: distributed","decide whether this is bug, use case, docs, or support",2026-07-01T01:33:04Z
issue,188627,DISABLED test_control_deps_passthrough_creates_ordering_barrier (__main__.TestControlDeps),https://github.com/pytorch/pytorch/issues/188627,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-07-01T01:33:03Z
issue,187000,DISABLED test_upper_bound_i64_cuda (__main__.AOTInductorTestDualWrapper),https://github.com/pytorch/pytorch/issues/187000,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, module: inductor, oncall: export, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-07-01T01:32:55Z
issue,187176,DISABLED test_functional_call (__main__.FuncTorchHigherOrderOpTestsWithCompiledAutograd),https://github.com/pytorch/pytorch/issues/187176,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-07-01T01:32:49Z
issue,149080,DISABLED test_aoti_debug_printer_codegen_cuda (__main__.AOTInductorTestABICompatibleGpu),https://github.com/pytorch/pytorch/issues/149080,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-07-01T01:32:46Z
issue,182260,DISABLED test_assert_size_stride_guarded_by_env_check_cpu_with_stack_allocation (__main__.AOTInductorTestABICompatibleCpuWithStackAllocation),https://github.com/pytorch/pytorch/issues/182260,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-07-01T01:32:43Z
issue,187046,DISABLED test_tensor_with_unbacked_shape_closure (__main__.HigherOrderOpTestsWithCompiledAutograd),https://github.com/pytorch/pytorch/issues/187046,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-06-30T19:16:09Z
issue,180439,DISABLED test_vmap_over_vmap_captured (__main__.FuncTorchHigherOrderOpTestsWithCompiledAutograd),https://github.com/pytorch/pytorch/issues/180439,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-06-30T19:16:06Z
issue,48628,Add docs on PyTorch - NumPy interaction,https://github.com/pytorch/pytorch/issues/48628,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: docs, module: numpy","decide whether this is bug, use case, docs, or support",2026-06-30T12:15:56Z
issue,188542,PyProcessGroup is missing several `ProcessGroup` collective overrides,https://github.com/pytorch/pytorch/issues/188542,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-30T10:53:52Z
issue,186872,DISABLED test_call_count_tunableop_cuda_float32 (__main__.TestLinalgCudaOnlyCUDA),https://github.com/pytorch/pytorch/issues/186872,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, module: linear algebra","decide whether this is bug, use case, docs, or support",2026-06-30T07:59:52Z
issue,186909,DISABLED test_trace_only_export_matches_default_use_cuda_True (__main__.TestProfiler),https://github.com/pytorch/pytorch/issues/186909,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, module: rocm, oncall: profiler","decide whether this is bug, use case, docs, or support",2026-06-30T07:59:46Z
issue,180366,DISABLED test_vmap_free_const (__main__.FuncTorchHigherOrderOpTestsWithCompiledAutograd),https://github.com/pytorch/pytorch/issues/180366,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-06-30T07:59:45Z
issue,187996,TestDTensorStateDictStager.test_async_save_no_memory_leak fails on GB200,https://github.com/pytorch/pytorch/issues/187996,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed checkpointing","decide whether this is bug, use case, docs, or support",2026-06-30T07:51:56Z
issue,106665,Inconsistency between CPU and GPU for `Linear()` layer with input size 0,https://github.com/pytorch/pytorch/issues/106665,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: edge cases, module: nn","decide whether this is bug, use case, docs, or support",2026-06-30T07:37:01Z
issue,188360,Inconsistent and unclear error messages for `offsets` validation when `input.dim() == 1` in `torch/nn/functional.py`,https://github.com/pytorch/pytorch/issues/188360,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: error checking, module: nn","decide whether this is bug, use case, docs, or support",2026-06-30T06:29:08Z
issue,187290,DISABLED test_Conv2d_depthwise_kernel_flag_cuda_float16 (__main__.TestConvolutionNNCUDACUDA),https://github.com/pytorch/pytorch/issues/187290,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, module: nn","decide whether this is bug, use case, docs, or support",2026-06-30T01:25:58Z
issue,180324,DISABLED test_jvp_two_tensors_has_aux (__main__.FuncTorchHigherOrderOpTestsWithCompiledAutograd),https://github.com/pytorch/pytorch/issues/180324,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-06-29T19:17:43Z
issue,187463,PyTorch has Inconsistent Negative Indexing Semantics,https://github.com/pytorch/pytorch/issues/187463,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: advanced indexing, module: python frontend, module: scatter & gather ops","decide whether this is bug, use case, docs, or support",2026-06-29T18:07:23Z
issue,188446,Add issue de-duplication system,https://github.com/pytorch/pytorch/issues/188446,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: devx,"decide whether this is bug, use case, docs, or support",2026-06-29T17:50:26Z
issue,132845,[PTD BE DAY]Burn Down Distributed Disabled Tests!!,https://github.com/pytorch/pytorch/issues/132845,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: oncall: distributed,"decide whether this is bug, use case, docs, or support",2026-06-29T17:39:14Z
issue,187473,[torch 2.13] umbrella issue - vLLM CI failures,https://github.com/pytorch/pytorch/issues/187473,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: vllm, oncall: pt2, oncall: releng","decide whether this is bug, use case, docs, or support",2026-06-29T17:21:49Z
issue,188430,"torch.compile produces large absolute errors for model with LSTM, MultiheadAttention, and torch.reciprocal",https://github.com/pytorch/pytorch/issues/188430,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: correctness (silent), oncall: pt2","decide whether this is bug, use case, docs, or support",2026-06-29T17:18:16Z
issue,188432,torch.compile produces inconsistent outputs for Multi-Query Attention model with manually implemented attention,https://github.com/pytorch/pytorch/issues/188432,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: correctness (silent), oncall: pt2","decide whether this is bug, use case, docs, or support",2026-06-29T17:17:55Z
issue,188375,"CPU LayerNorm normalized_shape=(1,) gives nonzero weight.grad although output is independent of weight",https://github.com/pytorch/pytorch/issues/188375,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: autograd, module: bfloat16, module: correctness (silent), module: cpu","decide whether this is bug, use case, docs, or support",2026-06-29T17:15:40Z
issue,187974,reserving issue for myself,https://github.com/pytorch/pytorch/issues/187974,needs maintainer decision,28,open issue has no stronger automated bucket,"decide whether this is bug, use case, docs, or support",2026-06-29T17:08:44Z
issue,188074,torch.compile produces inconsistent outputs for stateful SNN model with custom LIF neurons,https://github.com/pytorch/pytorch/issues/188074,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: correctness (silent), module: determinism, module: dynamo, module: nn","decide whether this is bug, use case, docs, or support",2026-06-29T17:08:27Z
issue,176662,C++20 Rollout Tracking,https://github.com/pytorch/pytorch/issues/176662,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: cpp,"decide whether this is bug, use case, docs, or support",2026-06-29T17:05:49Z
issue,154449,16KB pagination support for PyTorch Torch Vision and PyTorch Lite,https://github.com/pytorch/pytorch/issues/154449,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: android,"decide whether this is bug, use case, docs, or support",2026-06-29T16:05:30Z
issue,167684,[RFC] Access to ncclDevComm in kernel,https://github.com/pytorch/pytorch/issues/167684,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: symm_mem, oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-29T12:32:44Z
issue,188222,[aot_autograd] `RuntimeError: view size is not compatible ... use .reshape()` under `torch.compile`,https://github.com/pytorch/pytorch/issues/188222,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: aotdispatch, module: functionalization, module: pt2-dispatcher, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-06-29T10:36:07Z
issue,188005,[RFC] Protect SymmetricMemory signal pads against cross-stream races,https://github.com/pytorch/pytorch/issues/188005,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: symm_mem, oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-29T09:38:16Z
issue,186030,[DTensor][local_map]Support `None` placements in `local_map` placement specs,https://github.com/pytorch/pytorch/issues/186030,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dtensor, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-29T08:30:15Z
issue,180204,DISABLED test_jacrev_has_aux (__main__.FuncTorchHigherOrderOpTestsWithCompiledAutograd),https://github.com/pytorch/pytorch/issues/180204,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-06-29T08:25:25Z
issue,188393,DISABLED test_comprehensive_linalg_matrix_sqrth_xpu_float32 (__main__.TestInductorOpInfoXPU),https://github.com/pytorch/pytorch/issues/188393,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: inductor, module: linear algebra, module: xpu, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-06-29T06:56:35Z
issue,188392,DISABLED test_comprehensive_linalg_matrix_sqrth_xpu_float64 (__main__.TestInductorOpInfoXPU),https://github.com/pytorch/pytorch/issues/188392,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: inductor, module: linear algebra, module: xpu, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-06-29T06:53:33Z
issue,188391,DISABLED test_compare_cpu_linalg_matrix_sqrth_xpu_float32 (__main__.TestCommonXPU),https://github.com/pytorch/pytorch/issues/188391,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: linear algebra, module: tests, module: xpu","decide whether this is bug, use case, docs, or support",2026-06-29T06:51:21Z
issue,188238,DISABLED test_comprehensive_linalg_polar_xpu_float64 (__main__.TestInductorOpInfoXPU),https://github.com/pytorch/pytorch/issues/188238,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: inductor, module: linear algebra, module: xpu, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-06-29T06:50:23Z
issue,188389,[Docs] torch.nn.Module.parameters() does not specify whether the ordering is deterministic,https://github.com/pytorch/pytorch/issues/188389,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: docs,"decide whether this is bug, use case, docs, or support",2026-06-29T05:58:42Z
issue,147971,Fp8 scaled-mm row-wise is substantially slower than tensor-wise,https://github.com/pytorch/pytorch/issues/147971,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: cuda, module: floatx (formerly float8), module: performance","decide whether this is bug, use case, docs, or support",2026-06-28T06:22:30Z
issue,175229,Idiomatic way of matching against flex attention regions in SAC policy with wrap_inductor_compiled_regions,https://github.com/pytorch/pytorch/issues/175229,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: compile ux, module: flex attention, module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-06-28T03:27:23Z
issue,77764,General MPS op coverage tracking issue,https://github.com/pytorch/pytorch/issues/77764,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: mps,"decide whether this is bug, use case, docs, or support",2026-06-27T11:10:58Z
issue,139792,Speed up `DataLoader` worker start time for large datasets,https://github.com/pytorch/pytorch/issues/139792,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: dataloader,"decide whether this is bug, use case, docs, or support",2026-06-27T02:12:17Z
issue,188317,[RFC] Add process group revoke in torch.distributed,https://github.com/pytorch/pytorch/issues/188317,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-27T00:27:05Z
issue,166166,MultiheadAttention fast_path is substantially less efficient than the alternative,https://github.com/pytorch/pytorch/issues/166166,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: nn,"decide whether this is bug, use case, docs, or support",2026-06-26T23:55:53Z
issue,179897,DISABLED test_jacfwd (__main__.FuncTorchHigherOrderOpTestsWithCompiledAutograd),https://github.com/pytorch/pytorch/issues/179897,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-06-26T19:15:43Z
issue,179853,DISABLED test_hessian_argnums (__main__.FuncTorchHigherOrderOpTestsWithCompiledAutograd),https://github.com/pytorch/pytorch/issues/179853,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-06-26T19:15:36Z
issue,188077,torch._dynamo.disable seems expensive,https://github.com/pytorch/pytorch/issues/188077,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dynamo, module: performance, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-06-26T14:23:21Z
issue,179811,DISABLED test_hessian (__main__.FuncTorchHigherOrderOpTestsWithCompiledAutograd),https://github.com/pytorch/pytorch/issues/179811,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-06-26T13:21:40Z
issue,188205,Warn if torch.Tensor.backward() or torch.autograd.grad() is called underneath torch.cuda.use_mem_pool() with torch.autograd.grad_mode.set_multithreading_enabled(True),https://github.com/pytorch/pytorch/issues/188205,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: autograd, module: cuda","decide whether this is bug, use case, docs, or support",2026-06-26T12:15:36Z
issue,179990,DISABLED test_jacfwd_has_aux (__main__.FuncTorchHigherOrderOpTestsWithCompiledAutograd),https://github.com/pytorch/pytorch/issues/179990,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-06-26T07:59:07Z
issue,187019,DISABLED test_combo_kernel_no_bench_persistent_reduction (__main__.ComboKernelTests),https://github.com/pytorch/pytorch/issues/187019,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: inductor, module: tests, module: xpu, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-06-26T01:52:28Z
issue,188234,DISABLED test_del_attr_multi_stream_sync_dealloc (__main__.TestStreams),https://github.com/pytorch/pytorch/issues/188234,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dynamo, module: flaky-tests, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-06-26T01:30:04Z
issue,188232,[aot_autograd] `linalg.eigh` fails to converge (ill-conditioned) under `torch.compile` but not eager,https://github.com/pytorch/pytorch/issues/188232,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: aotdispatch, module: edge cases, module: linear algebra, module: pt2-dispatcher","decide whether this is bug, use case, docs, or support",2026-06-26T01:30:01Z
issue,181322,DISABLED test_trace_return_dataclass_nested (__main__.TestFX),https://github.com/pytorch/pytorch/issues/181322,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, module: fx, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-06-26T01:29:52Z
issue,179874,DISABLED test_lazy_init_with_view (__main__.LazyFuncionalizationTest),https://github.com/pytorch/pytorch/issues/179874,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, oncall: jit, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-06-26T01:29:48Z
issue,188212,[RFC] Add Gluon Backend To Inductor Backend,https://github.com/pytorch/pytorch/issues/188212,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flex attention, module: inductor, module: rocm, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-06-26T00:13:14Z
issue,162588,scaled_dot_product_attention produces incorrect results with attn_mask on large sequence lengths,https://github.com/pytorch/pytorch/issues/162588,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: sdpa,"decide whether this is bug, use case, docs, or support",2026-06-25T22:34:26Z
issue,136275,Please offer `<version>+cpu`-versioned packages for MacOS and linux ARM CPU-only builds,https://github.com/pytorch/pytorch/issues/136275,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: oncall: releng,"decide whether this is bug, use case, docs, or support",2026-06-25T22:10:13Z
issue,178368,On the future of triton-windows,https://github.com/pytorch/pytorch/issues/178368,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: binaries, module: ci, module: windows, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-06-25T21:47:02Z
issue,187862,`OpOverload.__call__` is roughly 100ns overhead,https://github.com/pytorch/pytorch/issues/187862,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: custom-operators, module: performance, module: pt2-dispatcher, module: python frontend","decide whether this is bug, use case, docs, or support",2026-06-25T19:29:14Z
issue,188151,torch.cond fails in torch.compile with simpliest symbolic shapes (torch 2.12+),https://github.com/pytorch/pytorch/issues/188151,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dynamic shapes, module: dynamo, module: higher order operators, module: pt2-dispatcher","decide whether this is bug, use case, docs, or support",2026-06-25T17:40:51Z
issue,158032,extension-cpp claims ROCM support with no additional changes; we should test this.,https://github.com/pytorch/pytorch/issues/158032,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: cpp-extensions, module: rocm","decide whether this is bug, use case, docs, or support",2026-06-25T16:41:24Z
issue,181499,RocM: Assumed realtime clock rate,https://github.com/pytorch/pytorch/issues/181499,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: build, module: rocm, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-06-25T16:34:28Z
issue,181939,HIP related errors compiling against PyTorch,https://github.com/pytorch/pytorch/issues/181939,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: build, module: cpp, module: rocm","decide whether this is bug, use case, docs, or support",2026-06-25T16:21:47Z
issue,186350,[Feature Request] Distribute triton with device-local version identifiers instead of separate package names,https://github.com/pytorch/pytorch/issues/186350,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: binaries, module: rocm, module: xpu","decide whether this is bug, use case, docs, or support",2026-06-25T15:52:40Z
issue,187715,`torch.compile(mode='reduce-overhead')` silently degrades training on ROCm — `cudagraph_trees` has no HIP guard,https://github.com/pytorch/pytorch/issues/187715,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: correctness (silent), module: cuda graphs, module: inductor, module: rocm","decide whether this is bug, use case, docs, or support",2026-06-25T15:44:55Z
issue,187907,AOTInductor: reverse-mode autograd.grad through sqrt/reciprocal bakes a symbolic-shaped constant under a dynamic batch (strict export),https://github.com/pytorch/pytorch/issues/187907,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: aotinductor, module: dynamic shapes, module: inductor, oncall: export","decide whether this is bug, use case, docs, or support",2026-06-25T15:30:04Z
issue,188153,"torch.export.export Fails with RuntimeError and ""Please file an issue on github""",https://github.com/pytorch/pytorch/issues/188153,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: oncall: pt2,"decide whether this is bug, use case, docs, or support",2026-06-25T14:23:09Z
issue,130330,why not enable expandable_segments by default？,https://github.com/pytorch/pytorch/issues/130330,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: cuda, module: performance","decide whether this is bug, use case, docs, or support",2026-06-25T12:57:50Z
issue,188144,DISABLED test_custom_op_dce_dynamic_shapes_cuda (__main__.DynamicShapesCodegenGPUTests),https://github.com/pytorch/pytorch/issues/188144,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-06-25T07:55:18Z
issue,188123,DISABLED test_index_cuda_with_cpu_int64_propagate_real_tensors (__main__.PropagateRealTensorsFakeTensorTest),https://github.com/pytorch/pytorch/issues/188123,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, module: meta tensors, module: rocm","decide whether this is bug, use case, docs, or support",2026-06-25T07:55:02Z
issue,153460,DISABLED test_mempool_ctx_multithread (__main__.TestMemPool),https://github.com/pytorch/pytorch/issues/153460,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: cuda, module: flaky-tests","decide whether this is bug, use case, docs, or support",2026-06-25T07:55:00Z
issue,188135,DISABLED test_compare_cpu_linalg_polar_xpu_float32 (__main__.TestCommonXPU),https://github.com/pytorch/pytorch/issues/188135,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: linear algebra, module: tests, module: xpu","decide whether this is bug, use case, docs, or support",2026-06-25T06:03:25Z
issue,186871,DISABLED test_addmm_relu_tunableop_rocm_cuda_float32 (__main__.TestLinalgCudaOnlyCUDA),https://github.com/pytorch/pytorch/issues/186871,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, module: linear algebra, module: rocm","decide whether this is bug, use case, docs, or support",2026-06-25T01:27:39Z
issue,180322,DISABLED test_vjp_vjp_cpu (__main__.TestComposabilityCPU),https://github.com/pytorch/pytorch/issues/180322,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, module: functorch, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-06-25T01:27:23Z
issue,187615,[FSDP2] Fractional CPU offload policy (PartialOffloadPolicy),https://github.com/pytorch/pytorch/issues/187615,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: fsdp, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-25T00:23:11Z
issue,188001,[RFC] Unify rendezvous caching across SymmetricMemory backends,https://github.com/pytorch/pytorch/issues/188001,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: symm_mem, oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-24T18:56:32Z
issue,188087,"torch.nanmedian(..., dim=None) raises a named-dimension RuntimeError while omitting dim works",https://github.com/pytorch/pytorch/issues/188087,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: error checking, module: reductions","decide whether this is bug, use case, docs, or support",2026-06-24T18:13:41Z
issue,187963,Pallas CPU max_unpool2d does not preserve channels-last output layout,https://github.com/pytorch/pytorch/issues/187963,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: memory format, module: pallas, module: pooling, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-06-24T17:31:33Z
issue,9983,[feature request] Add matrix functions,https://github.com/pytorch/pytorch/issues/9983,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: linear algebra, module: numpy","decide whether this is bug, use case, docs, or support",2026-06-24T11:33:27Z
issue,187649,[Inductor] Support fx wrapper with non-triton fused kernel.,https://github.com/pytorch/pytorch/issues/187649,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: fx, module: higher order operators, module: inductor, module: pt2-dispatcher","decide whether this is bug, use case, docs, or support",2026-06-24T08:58:29Z
issue,186825,Bernoulli(probs=0).log_prob(1) returns dtype-dependent finite values instead of -inf (Geometric special-cases this and tests it),https://github.com/pytorch/pytorch/issues/186825,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: distributions, module: edge cases","decide whether this is bug, use case, docs, or support",2026-06-24T03:49:18Z
issue,188020,Allow platform context in DataLoader worker fatal signal messages,https://github.com/pytorch/pytorch/issues/188020,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: dataloader,"decide whether this is bug, use case, docs, or support",2026-06-24T03:00:14Z
issue,117662,dynamo <> autograd.Function assumes that grad_outputs have the same subclass-ness as forward outputs,https://github.com/pytorch/pytorch/issues/117662,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: autograd, module: dynamo, module: pt2-dispatcher, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-06-23T20:27:09Z
issue,185587,[inductor] Numerical inconsistency: polygamma + cumulative_trapezoid under torch.compile(dynamic=True) produces wrong results after Conv2d + GroupNorm,https://github.com/pytorch/pytorch/issues/185587,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: correctness (silent), module: dynamic shapes, module: inductor, module: special","decide whether this is bug, use case, docs, or support",2026-06-23T17:30:32Z
issue,185295,[Inductor][feature] Block FP8 quantization: fuse scale layout transformation with quant math,https://github.com/pytorch/pytorch/issues/185295,needs maintainer decision,28,open issue has no stronger automated bucket,"decide whether this is bug, use case, docs, or support",2026-06-23T17:27:07Z
issue,186381,[export] No-op contiguous() elided during tracing bakes trace-time layout into the graph; downstream view becomes layout-specific,https://github.com/pytorch/pytorch/issues/186381,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: export, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-06-23T16:30:11Z
issue,186594,torch.ops.higher_order.associative_scan cannot be torch.export'ed,https://github.com/pytorch/pytorch/issues/186594,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: export, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-06-23T16:29:54Z
issue,186999,DISABLED test_bmm_large_batch_dynamic_cuda (__main__.AOTInductorTestDualWrapper),https://github.com/pytorch/pytorch/issues/186999,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, module: inductor, oncall: export, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-06-23T16:29:45Z
issue,148819,addition of muon optimizer to torch.optim,https://github.com/pytorch/pytorch/issues/148819,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: optimizer,"decide whether this is bug, use case, docs, or support",2026-06-23T16:25:57Z
issue,187183,Update slow tests job removes tests that may still be slow,https://github.com/pytorch/pytorch/issues/187183,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: ci, module: tests","decide whether this is bug, use case, docs, or support",2026-06-23T14:31:00Z
issue,187807,[PyTorch 2.13][API-Unstable] Intel XPU Enhancements,https://github.com/pytorch/pytorch/issues/187807,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: floatx (formerly float8), module: windows, module: xpu","decide whether this is bug, use case, docs, or support",2026-06-23T07:36:47Z
issue,180974,[RFC][inductor] Support pre/post kernel execution hooks in compiled graphs,https://github.com/pytorch/pytorch/issues/180974,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-06-23T06:40:17Z
issue,93531,TorchInductor CPU Performance Dashboard,https://github.com/pytorch/pytorch/issues/93531,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: cpu inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-06-23T06:13:49Z
issue,173071,Allow privateuse1 devices in FlexAttention device validation,https://github.com/pytorch/pytorch/issues/173071,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flex attention, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-06-23T04:49:48Z
issue,159547,grouped_mm for float16,https://github.com/pytorch/pytorch/issues/159547,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: cuda, module: half, module: linear algebra","decide whether this is bug, use case, docs, or support",2026-06-23T03:39:52Z
issue,186260,[Accelerator] Maybe host memory should be included in torch.accelerator.mem_stats,https://github.com/pytorch/pytorch/issues/186260,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: accelerator,"decide whether this is bug, use case, docs, or support",2026-06-22T22:54:41Z
issue,187770,torch.nn.attention.varlen: make backend selection configurable,https://github.com/pytorch/pytorch/issues/187770,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: sdpa,"decide whether this is bug, use case, docs, or support",2026-06-22T21:14:09Z
issue,178196,libvnrtc error in torch nightly on cuda 13.0,https://github.com/pytorch/pytorch/issues/178196,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: binaries, module: cuda","decide whether this is bug, use case, docs, or support",2026-06-22T16:20:54Z
issue,187721,[Inductor] Silent correctness : F.kl_div reduction over -0.0 yields different signbits in Eager vs Compiled,https://github.com/pytorch/pytorch/issues/187721,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: correctness (silent), module: half, module: inductor, module: loss","decide whether this is bug, use case, docs, or support",2026-06-22T15:23:49Z
issue,187834,reminder to check cu129 binaries aren't released with pytorch 2.14,https://github.com/pytorch/pytorch/issues/187834,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: binaries, module: ci","decide whether this is bug, use case, docs, or support",2026-06-22T14:11:00Z
issue,186939,pitch: fast development mode where we use pre-build nightly .so files,https://github.com/pytorch/pytorch/issues/186939,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: devx,"decide whether this is bug, use case, docs, or support",2026-06-22T13:43:56Z
issue,187827,[pr-review skill] Add MPS/Metal kernel review checks (tracking),https://github.com/pytorch/pytorch/issues/187827,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: mps,"decide whether this is bug, use case, docs, or support",2026-06-22T13:41:29Z
issue,105632,torch.nn.TransformerDecoderLayer lacks parameter validation check,https://github.com/pytorch/pytorch/issues/105632,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: nn,"decide whether this is bug, use case, docs, or support",2026-06-22T12:49:53Z
issue,174131,turn on autograd_cache_normalize_inputs in fbcode,https://github.com/pytorch/pytorch/issues/174131,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: pt2-dispatcher, module: vllm, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-06-22T11:05:17Z
issue,186465,[RFC] Fix Control flow operators consistency,https://github.com/pytorch/pytorch/issues/186465,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dynamo, module: higher order operators, module: inductor, module: pt2-dispatcher","decide whether this is bug, use case, docs, or support",2026-06-22T07:05:47Z
issue,145679,Gamma cdf function,https://github.com/pytorch/pytorch/issues/145679,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: distributions,"decide whether this is bug, use case, docs, or support",2026-06-21T21:14:19Z
issue,121135,[torch.compile] `randperm_index_add_pattern` doesn't check the shape mismatch between self and source tensor,https://github.com/pytorch/pytorch/issues/121135,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-06-20T13:54:37Z
issue,153809,Optimize rowwise scaled mm CUTLASS kernel for large values of K,https://github.com/pytorch/pytorch/issues/153809,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: floatx (formerly float8),"decide whether this is bug, use case, docs, or support",2026-06-20T08:26:31Z
issue,143281,torch.linalg.qr is significantly slower on GPU compared to CPU and SVD for batched small matrices,https://github.com/pytorch/pytorch/issues/143281,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: cuda, module: linear algebra","decide whether this is bug, use case, docs, or support",2026-06-20T05:18:44Z
issue,181861,nn.Sequential type hints too general for nested lists,https://github.com/pytorch/pytorch/issues/181861,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: typing,"decide whether this is bug, use case, docs, or support",2026-06-20T02:12:37Z
issue,187180,torch.nn.functional.softplus returns +Inf for finite float16 input when beta is very small,https://github.com/pytorch/pytorch/issues/187180,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: correctness (silent), module: edge cases, module: half, module: nn","decide whether this is bug, use case, docs, or support",2026-06-19T18:52:25Z
issue,187470,MemberType LLM readability,https://github.com/pytorch/pytorch/issues/187470,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: custom-operators, module: pt2-dispatcher, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-06-19T18:31:39Z
issue,187561,DISABLED test_view_ops (__main__.TestViewOpsWithLocalTensor),https://github.com/pytorch/pytorch/issues/187561,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dtensor, module: flaky-tests, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-19T08:40:27Z
issue,155512,Unable to find caffe2 when building libtorch from the source and then trying to use it in a cmake project,https://github.com/pytorch/pytorch/issues/155512,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: build, module: cpp","decide whether this is bug, use case, docs, or support",2026-06-19T07:59:54Z
issue,186826,"Distribution.icdf performs no argument validation even with validate_args=True, unlike cdf/log_prob",https://github.com/pytorch/pytorch/issues/186826,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: distributions, module: error checking","decide whether this is bug, use case, docs, or support",2026-06-19T06:35:43Z
issue,187698,"Vulkan nn.Linear unavailable with requires_grad=True on Windows (inference works, training path blocked)",https://github.com/pytorch/pytorch/issues/187698,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: autograd, module: nn, module: vulkan, module: windows","decide whether this is bug, use case, docs, or support",2026-06-18T21:26:43Z
issue,187241,"[inductor] TypeError: Too few arguments for CSE in cutedsl_kernel.py (torch 2.10.0, Python 3.11)",https://github.com/pytorch/pytorch/issues/187241,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: inductor, module: python version, module: typing, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-06-18T20:45:58Z
issue,181428,torch.compile + torch.func.grad: superlinear compile time with MaxPool2d → log/clamp → pow → AdaptiveAvgPool2d,https://github.com/pytorch/pytorch/issues/181428,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: compile-time, module: dynamic shapes, module: functorch, module: inductor","decide whether this is bug, use case, docs, or support",2026-06-18T19:27:18Z
issue,110700,More informative variable names in AOTAutograd,https://github.com/pytorch/pytorch/issues/110700,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: aotdispatch, module: fx, module: pt2-dispatcher, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-06-18T18:16:25Z
issue,164666,sparse.mm(CSR x dense) backward fails on CUDA for bfloat16 but CSR @ dense backward works,https://github.com/pytorch/pytorch/issues/164666,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: bfloat16, module: sparse","decide whether this is bug, use case, docs, or support",2026-06-18T17:38:04Z
issue,187239,[NCCL][FSDP2] Support parameter-contiguous all-gather output to skip copy-out,https://github.com/pytorch/pytorch/issues/187239,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: c10d, module: fsdp, module: performance, oncall: distributed","decide whether this is bug, use case, docs, or support",2026-06-18T16:39:08Z
issue,187645,DISABLED test_async_future_results (__main__.TestAsyncCheckpointerSpecific),https://github.com/pytorch/pytorch/issues/187645,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, module: rocm, oncall: distributed checkpointing","decide whether this is bug, use case, docs, or support",2026-06-18T08:23:18Z
issue,179745,DISABLED test_memory_snapshot (__main__.TestCudaAllocator),https://github.com/pytorch/pytorch/issues/179745,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: cuda, module: flaky-tests","decide whether this is bug, use case, docs, or support",2026-06-18T08:23:08Z
issue,185452,DISABLED test_torchvision_models_convnext_large (__main__.TestVisionTracing),https://github.com/pytorch/pytorch/issues/185452,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, module: fx","decide whether this is bug, use case, docs, or support",2026-06-18T08:22:58Z
issue,177010,[RFC] Introduce the unified RNG-related API to torch.accelerator,https://github.com/pytorch/pytorch/issues/177010,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: accelerator, module: random","decide whether this is bug, use case, docs, or support",2026-06-18T07:55:22Z
issue,167365,PyTorch is no longer importable from `build` folder,https://github.com/pytorch/pytorch/issues/167365,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: build,"decide whether this is bug, use case, docs, or support",2026-06-18T05:41:19Z
issue,184466,Implementing Self-Regulated Kinetic Learning (SRKL) Optimizer using Second-Order Curvature,https://github.com/pytorch/pytorch/issues/184466,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: autograd, module: optimizer","decide whether this is bug, use case, docs, or support",2026-06-18T03:32:00Z
issue,187179,"torch.cdist returns +Inf for small p (p-norm overflow, docstring admits p in [0, inf])",https://github.com/pytorch/pytorch/issues/187179,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: distance functions, module: edge cases, module: numerical-stability","decide whether this is bug, use case, docs, or support",2026-06-17T20:51:55Z
issue,187591,[Dynamo] Symbolic Strings,https://github.com/pytorch/pytorch/issues/187591,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dynamo, module: guards, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-06-17T18:47:23Z
issue,187590,[Dynamo] LazyConstants -- Avoid over-guarding on constant values,https://github.com/pytorch/pytorch/issues/187590,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dynamo, module: guards, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-06-17T18:35:04Z
issue,118107,[Tracker] Nested tensor op coverage requests,https://github.com/pytorch/pytorch/issues/118107,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: nestedtensor,"decide whether this is bug, use case, docs, or support",2026-06-17T18:01:16Z
issue,187582,torch.nested NJTs with identical layout fail on binary operation,https://github.com/pytorch/pytorch/issues/187582,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: nestedtensor,"decide whether this is bug, use case, docs, or support",2026-06-17T17:48:25Z
issue,104053,Tracking issue for optimizer graph not being an inference graph,https://github.com/pytorch/pytorch/issues/104053,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: aotdispatch, module: pt2-dispatcher, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-06-17T15:17:58Z
issue,186601,[RFC]SDMA AllGather and Zero-Copy Output for PyTorch FSDP2,https://github.com/pytorch/pytorch/issues/186601,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: fsdp, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-17T14:00:54Z
issue,187148,[Proposal] Fix cpu_fallback returning tensors on wrong device for reverse binary ops,https://github.com/pytorch/pytorch/issues/187148,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: backend, module: dispatch, module: privateuse1","decide whether this is bug, use case, docs, or support",2026-06-17T07:59:05Z
issue,185639,"torch.func.jacfwd gives incorrect Jacobian for torch.pow(torch.abs(complex), complex exponent)",https://github.com/pytorch/pytorch/issues/185639,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: autograd, module: complex, module: functorch","decide whether this is bug, use case, docs, or support",2026-06-17T02:04:28Z
issue,187505,OpaqueObject Fuzzing,https://github.com/pytorch/pytorch/issues/187505,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: custom-operators, module: faketensor, module: fx, module: pt2-dispatcher","decide whether this is bug, use case, docs, or support",2026-06-16T21:53:49Z
issue,163429,[CI][CUDA][B200] test_max_autotune.py failures on B200 runner,https://github.com/pytorch/pytorch/issues/163429,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: ci, module: cuda, module: tests","decide whether this is bug, use case, docs, or support",2026-06-16T18:56:31Z
issue,187293,"torch.nn.functional.gelu(approximate=""none"") returns NaN for +Inf input",https://github.com/pytorch/pytorch/issues/187293,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: edge cases, module: nans and infs, module: nn","decide whether this is bug, use case, docs, or support",2026-06-16T14:08:06Z
issue,187276,Two-node DDP ResNet-50/Adam run ends with matching rank-1 loss but gradient and parameter-fingerprint divergence across CPU/Gloo and CUDA/NCCL,https://github.com/pytorch/pytorch/issues/187276,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: ddp, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-16T10:49:58Z
issue,187279,Two-node DDP shows reproducible FP32 loss divergence between CPU/Gloo and CUDA/NCCL with large lr and resnet18,https://github.com/pytorch/pytorch/issues/187279,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: ddp, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-16T09:22:50Z
issue,187424,DISABLED test_range_based_static_shape_no_cond_dispatch (__main__.TestCustomOpAutoTune),https://github.com/pytorch/pytorch/issues/187424,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-06-16T08:36:49Z
issue,104701,System memory leak when using different input size of torch.nn.Conv3d,https://github.com/pytorch/pytorch/issues/104701,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: cuda, module: cudnn, module: memory usage, module: nn","decide whether this is bug, use case, docs, or support",2026-06-16T06:36:35Z
issue,162859,[RFC] support symmetric memory in torch.compile,https://github.com/pytorch/pytorch/issues/162859,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: inductor, module: symm_mem, module: vllm, oncall: distributed","decide whether this is bug, use case, docs, or support",2026-06-15T23:55:21Z
issue,187158,[Inductor] AssertionError: Unregistered range symbol 'yindex' in TMA template epilogue fusion,https://github.com/pytorch/pytorch/issues/187158,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-06-15T23:03:03Z
issue,187227,[SDPA] Fix 32-bit offset overflow in mem-efficient attention forward with attn_bias,https://github.com/pytorch/pytorch/issues/187227,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: correctness (silent), module: cuda, module: int overflow, module: sdpa","decide whether this is bug, use case, docs, or support",2026-06-15T22:17:26Z
issue,156771,Support fp16 and fp32 outputs for rowwise scaled_mm,https://github.com/pytorch/pytorch/issues/156771,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: floatx (formerly float8),"decide whether this is bug, use case, docs, or support",2026-06-15T13:32:37Z
issue,180168,[RFC] XPUGraph Trees,https://github.com/pytorch/pytorch/issues/180168,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: inductor, module: xpu, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-06-15T03:49:07Z
issue,186449,[RFC]: Contextvars Support in TorchDynamo,https://github.com/pytorch/pytorch/issues/186449,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: compile ux, module: dynamo, module: graph breaks, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-06-14T17:41:28Z
issue,182001,CUDA linalg: use auxiliary streams for independent looped cuSOLVER batch calls,https://github.com/pytorch/pytorch/issues/182001,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: cuda, module: linear algebra, module: performance","decide whether this is bug, use case, docs, or support",2026-06-14T09:58:26Z
issue,187266,scaled_dot_product_attention falls back to math for unbatched (3D) inputs under vmap,https://github.com/pytorch/pytorch/issues/187266,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: functorch, module: performance, module: sdpa, module: vmap","decide whether this is bug, use case, docs, or support",2026-06-14T06:07:48Z
issue,187020,DISABLED test_redistribute_cost_latency (__main__.TestCostModel),https://github.com/pytorch/pytorch/issues/187020,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dtensor, module: flaky-tests, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-13T16:39:20Z
issue,173123,"[DTensor] Gradient of unused output becomes plain torch.Tensor in backward, causing type mismatch with torch.compile",https://github.com/pytorch/pytorch/issues/173123,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dtensor, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-13T16:38:34Z
issue,172491,Direct access to DTensor._local_tensor breaks _dynamo_graph_capture_for_export,https://github.com/pytorch/pytorch/issues/172491,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dtensor, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-13T16:38:02Z
issue,172485,[DTensor] Summary of issues related to `Partial` placement,https://github.com/pytorch/pytorch/issues/172485,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dtensor, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-13T16:37:52Z
issue,179374,"[Infra] Missing PEP 700 upload-time metadata breaks supply-chain security tools (uv, pip)",https://github.com/pytorch/pytorch/issues/179374,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: infra,"decide whether this is bug, use case, docs, or support",2026-06-13T03:02:17Z
issue,186143,AOTInductor Lowering Backward Compatibility,https://github.com/pytorch/pytorch/issues/186143,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: aotinductor, module: docs, oncall: export, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-06-12T21:25:07Z
issue,187203,DISABLED test_disable_external_correlation (__main__.TestProfilerCUDA),https://github.com/pytorch/pytorch/issues/187203,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, module: rocm, oncall: profiler","decide whether this is bug, use case, docs, or support",2026-06-12T19:22:08Z
issue,187202,DISABLED test_combo_kernel_fusion_within_same_stream (__main__.TestUserStreamCompile),https://github.com/pytorch/pytorch/issues/187202,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-06-12T19:22:06Z
issue,170479,nvshmem [GPU 0] Peer GPU 8 is not accessible,https://github.com/pytorch/pytorch/issues/170479,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: symm_mem, oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-12T17:10:47Z
issue,177245,AArch64 Unit Test Failures - Multiple failures in test/nn/test_convolution.py TestConvolutionNNDeviceTypeCPU for oneDNN,https://github.com/pytorch/pytorch/issues/177245,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: convolution, module: cpu, module: mkldnn","decide whether this is bug, use case, docs, or support",2026-06-12T17:10:32Z
issue,169814,DTensor.detach() retains stale DeviceMesh after destroying and reinitializing process groups,https://github.com/pytorch/pytorch/issues/169814,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dtensor, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-12T17:09:18Z
issue,162257,Optimize FSDP1 sharded_state_dict saving performance,https://github.com/pytorch/pytorch/issues/162257,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: fsdp, oncall: distributed, oncall: distributed checkpointing","decide whether this is bug, use case, docs, or support",2026-06-12T17:09:12Z
issue,169320,[DTensor] A New Placement -- RaggedShard,https://github.com/pytorch/pytorch/issues/169320,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dtensor, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-12T17:08:21Z
issue,169035,"[Question] Why torch.ops.symm_mem.multimem_all_reduce_() don't support e4m3, e5m2, fp16?",https://github.com/pytorch/pytorch/issues/169035,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: symm_mem, oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-12T17:08:20Z
issue,167917,Operator aten.detach_.default does not have a sharding strategy registered,https://github.com/pytorch/pytorch/issues/167917,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dtensor, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-12T17:07:32Z
issue,167897,Hooks as extension points for torch.pipelining,https://github.com/pytorch/pytorch/issues/167897,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: pipelining, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-12T17:07:31Z
issue,167657,make_fx trick for tracing DTensor compute doesn't work with redistribute,https://github.com/pytorch/pytorch/issues/167657,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dtensor, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-12T17:06:57Z
issue,186618,Inconsistent rounding between CUDA and CPU for division operator,https://github.com/pytorch/pytorch/issues/186618,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: cpu, module: cuda, module: edge cases","decide whether this is bug, use case, docs, or support",2026-06-12T08:20:55Z
issue,187156,DISABLED test_load_with_map_location (__main__.TestCheckpointer),https://github.com/pytorch/pytorch/issues/187156,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, module: rocm, oncall: distributed checkpointing","decide whether this is bug, use case, docs, or support",2026-06-12T08:10:37Z
issue,186837,[RFC]: Auto-select and run Operator Performance Benchmarks from PR Diffs,https://github.com/pytorch/pytorch/issues/186837,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: ci, module: performance","decide whether this is bug, use case, docs, or support",2026-06-12T04:23:15Z
issue,186905,download.pytorch.org packages doesn't have release timestamp,https://github.com/pytorch/pytorch/issues/186905,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: binaries,"decide whether this is bug, use case, docs, or support",2026-06-11T20:08:59Z
issue,187045,DISABLED test_out_of_memory_retry (__main__.TestCuda),https://github.com/pytorch/pytorch/issues/187045,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: cuda, module: flaky-tests","decide whether this is bug, use case, docs, or support",2026-06-11T19:31:29Z
issue,167074,DTensor with `torch.compile` fails with `aten.as_strided.default` error,https://github.com/pytorch/pytorch/issues/167074,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dtensor, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-11T17:25:46Z
issue,167219,Are there limitations to dtensor's registration strategy?,https://github.com/pytorch/pytorch/issues/167219,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dtensor, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-11T17:25:14Z
issue,167090,DTensor `F.conv2d` fails with `bias=None` - AssertionError in sharding propagation,https://github.com/pytorch/pytorch/issues/167090,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dtensor, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-11T17:25:12Z
issue,163692,[feature][dtensor] support nn.Module inputs containing DTensor parameters in local_map,https://github.com/pytorch/pytorch/issues/163692,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dtensor, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-11T17:24:00Z
issue,166638,[DTensor] calling item on dtensor should perform full tensor first,https://github.com/pytorch/pytorch/issues/166638,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dtensor, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-11T17:23:56Z
issue,163688,[torch.distributed.pipelining] Gradients are None in first training step with ScheduleGPipe,https://github.com/pytorch/pytorch/issues/163688,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: amp (automated mixed precision), module: pipelining, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-11T17:23:15Z
issue,161974,DTensor communications cost model is way off for H100,https://github.com/pytorch/pytorch/issues/161974,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dtensor, module: performance, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-11T17:23:10Z
issue,162215,consider allowing more dtype in the dtensor opinfo test,https://github.com/pytorch/pytorch/issues/162215,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dtensor, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-11T17:22:33Z
issue,161847,Feedback about FSDP Notes,https://github.com/pytorch/pytorch/issues/161847,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: docs, module: fsdp, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-11T17:22:06Z
issue,160968,[DTensor] broadcast_to fails with KeyError,https://github.com/pytorch/pytorch/issues/160968,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dtensor, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-11T17:21:46Z
issue,161218,[DTensor] Losing TupleStrategy in OpSchema.args_strategy,https://github.com/pytorch/pytorch/issues/161218,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dtensor, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-11T17:21:37Z
issue,155657,"[DTensor] DTensor is not well supported on older versions of GPUs, such as A10",https://github.com/pytorch/pytorch/issues/155657,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dtensor, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-11T17:21:09Z
issue,185142,RFC Test class classification,https://github.com/pytorch/pytorch/issues/185142,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: ci, module: tests","decide whether this is bug, use case, docs, or support",2026-06-11T16:38:14Z
issue,175869,[dynamo] Tracking issue for itertools polyfills,https://github.com/pytorch/pytorch/issues/175869,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dynamo, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-06-11T13:43:34Z
issue,34130,ExceptionWrapper cannot handle exceptions with more than one argument,https://github.com/pytorch/pytorch/issues/34130,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dataloader, module: error checking","decide whether this is bug, use case, docs, or support",2026-06-11T09:25:14Z
issue,158911,[RFC] Refactor test_c10d_nccl.py to Support Backend-Agnostic Testing for PyTorch,https://github.com/pytorch/pytorch/issues/158911,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: privateuse1, module: testing, oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-11T07:02:48Z
issue,180345,Allow registering multiple out-of-tree backends and setting current_accelerator,https://github.com/pytorch/pytorch/issues/180345,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: accelerator, module: backend","decide whether this is bug, use case, docs, or support",2026-06-11T04:44:45Z
issue,186911,DISABLED test_launch_elastic_multiple_agents (__main__.ElasticLaunchTest),https://github.com/pytorch/pytorch/issues/186911,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, oncall: r2p","decide whether this is bug, use case, docs, or support",2026-06-10T22:26:30Z
issue,184352,Python 3.15 support for PyTorch,https://github.com/pytorch/pytorch/issues/184352,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: ci, module: python frontend, module: python version","decide whether this is bug, use case, docs, or support",2026-06-10T21:31:03Z
issue,160340,Meta device initialization on HF models + FSDP leads to different numerical behaviour,https://github.com/pytorch/pytorch/issues/160340,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: fsdp, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-10T17:16:37Z
issue,160911,Severe differences between outputs of Linear with DTensor and Tensor in half precision,https://github.com/pytorch/pytorch/issues/160911,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dtensor, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-10T17:16:24Z
issue,160169,"When using DP + TP, DP only parameters diverge across TP ranks if using operations with non-deterministic implementations",https://github.com/pytorch/pytorch/issues/160169,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: fsdp, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-10T17:16:15Z
issue,160068,FSDP2 not compatible with transformers >= 4.54.0 GenericForTokenClassification,https://github.com/pytorch/pytorch/issues/160068,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: fsdp, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-10T17:15:47Z
issue,160272,Add Green Context Support for fsdp2,https://github.com/pytorch/pytorch/issues/160272,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: fsdp, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-10T17:15:39Z
issue,160648,How to Use Pipeline Parallelism in Multi-input Models,https://github.com/pytorch/pytorch/issues/160648,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: pipelining, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-10T17:15:38Z
issue,159836,Missing documentation for device mesh on DDP,https://github.com/pytorch/pytorch/issues/159836,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: ddp, module: docs, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-10T17:15:04Z
issue,159959,TensorMetadata not retained during DTensor convolution double backpropagation,https://github.com/pytorch/pytorch/issues/159959,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dtensor, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-10T17:15:03Z
issue,156217,[Tracker] AutoParallel's feature request to DTensor,https://github.com/pytorch/pytorch/issues/156217,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dtensor, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-10T17:14:25Z
issue,154657,[FSDP2] offer public API to share communication context aross fsdp roots,https://github.com/pytorch/pytorch/issues/154657,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: fsdp, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-10T17:14:09Z
issue,145218,getting different results when adding `torch.Tensor` or python number to a DTensor - Is that expected?,https://github.com/pytorch/pytorch/issues/145218,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dtensor, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-10T17:13:32Z
issue,146884,[DTensor]Operator aten.mT.default/aten.mv.default/aten.dot.default does not have a sharding strategy registered,https://github.com/pytorch/pytorch/issues/146884,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dtensor, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-10T17:13:27Z
issue,143748,[DTensor]`Linear` fails on 3D DTensor with `batch size > 1` and `Replicate` input redistributed from `Shard`,https://github.com/pytorch/pytorch/issues/143748,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dtensor, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-10T17:13:20Z
issue,143823,Possible race condition found in TailLogTest.test_tail,https://github.com/pytorch/pytorch/issues/143823,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: elastic, module: tests, oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-10T17:13:07Z
issue,143985,[DTensor] Allow multiple dimensions to be sharded together (as if flattened),https://github.com/pytorch/pytorch/issues/143985,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dtensor, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-10T17:12:51Z
issue,143647,A very weird bug involving ddp,https://github.com/pytorch/pytorch/issues/143647,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: ddp, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-10T17:12:16Z
issue,185770,"torch.nn.functional.gelu returns NaN for +inf input with approximate=""none""",https://github.com/pytorch/pytorch/issues/185770,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: edge cases, module: nans and infs, module: nn","decide whether this is bug, use case, docs, or support",2026-06-10T16:18:18Z
issue,186912,DISABLED test_checkpoint_write_future_state_dict (__main__.TestCheckpointProcess),https://github.com/pytorch/pytorch/issues/186912,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, module: rocm, oncall: distributed checkpointing","decide whether this is bug, use case, docs, or support",2026-06-10T08:02:47Z
issue,186910,DISABLED test_Embedding_discontiguous_cuda (__main__.TestNN),https://github.com/pytorch/pytorch/issues/186910,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, module: nn, module: rocm","decide whether this is bug, use case, docs, or support",2026-06-10T08:02:41Z
issue,186908,DISABLED test_nested_tensor_multiprocessing_context_forkserver_cuda (__main__.TestDataLoaderDeviceTypeCUDA),https://github.com/pytorch/pytorch/issues/186908,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dataloader, module: flaky-tests, module: rocm","decide whether this is bug, use case, docs, or support",2026-06-10T08:02:35Z
issue,186873,DISABLED test_mm_submatrix_offline_tunableop_cuda_float32 (__main__.TestLinalgCudaOnlyCUDA),https://github.com/pytorch/pytorch/issues/186873,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, module: linear algebra, module: rocm","decide whether this is bug, use case, docs, or support",2026-06-10T08:02:10Z
issue,86735,DISABLED test_reentrant_parent_error_on_cpu_cuda (__main__.TestAutogradDeviceTypeCUDA),https://github.com/pytorch/pytorch/issues/86735,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: autograd, module: flaky-tests","decide whether this is bug, use case, docs, or support",2026-06-10T08:02:00Z
issue,186868,DISABLED test_LSTM_dropout_per_call_randomness_dropout_p_0_5_training_True_cuda (__main__.TestNNDeviceTypeCUDA),https://github.com/pytorch/pytorch/issues/186868,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, module: nn, module: rocm","decide whether this is bug, use case, docs, or support",2026-06-10T08:01:58Z
issue,104695,DISABLED test_cuda_memory_leak_detection (__main__.TestCudaMultiGPU),https://github.com/pytorch/pytorch/issues/104695,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: cuda, module: flaky-tests","decide whether this is bug, use case, docs, or support",2026-06-10T08:01:58Z
issue,181318,DISABLED test_codegen_structure_three_stream_pipeline (__main__.TestUserStreamCompile),https://github.com/pytorch/pytorch/issues/181318,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-06-10T08:01:55Z
issue,181224,DISABLED test_cross_stream_stride_copy (__main__.TestUserStreamCompile),https://github.com/pytorch/pytorch/issues/181224,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-06-10T08:01:54Z
issue,186864,DISABLED test_triggers (__main__.CallbackTests),https://github.com/pytorch/pytorch/issues/186864,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dynamo, module: flaky-tests, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-06-10T08:01:40Z
issue,183603,torch.compile fails on second-order gradient after autograd.grad(create_graph=True),https://github.com/pytorch/pytorch/issues/183603,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: double backwards, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-06-10T07:27:02Z
issue,106692,torch.polygamma inconsistent with scipy.special.polygamma for n >= 1,https://github.com/pytorch/pytorch/issues/106692,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: numpy, module: special","decide whether this is bug, use case, docs, or support",2026-06-10T02:55:54Z
issue,182174,DISABLED test_torchvision_models_efficientnet_b0 (__main__.TestVisionTracing),https://github.com/pytorch/pytorch/issues/182174,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, module: fx","decide whether this is bug, use case, docs, or support",2026-06-10T01:29:39Z
issue,100074,DISABLED test_checkpointing_resets_persistent_refs (__main__.CudaGraphTreeTests),https://github.com/pytorch/pytorch/issues/100074,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, module: inductor","decide whether this is bug, use case, docs, or support",2026-06-10T01:29:33Z
issue,138464,`torch.special.zeta` ignores `nan` input when `other=-inf`.,https://github.com/pytorch/pytorch/issues/138464,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: edge cases, module: nans and infs, module: python frontend, module: special","decide whether this is bug, use case, docs, or support",2026-06-09T18:51:05Z
issue,164563,[Blackwell][Inductor] Numerical mismatches in test_max_autotune.py::TestMaxAutotune::test_non_contiguous_input_mm_plus_mm,https://github.com/pytorch/pytorch/issues/164563,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: correctness (silent), module: cuda, module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-06-09T17:29:44Z
issue,156523,[Dtensor] handle dtensor ops that only need to operate on certain shard without all_gather first.,https://github.com/pytorch/pytorch/issues/156523,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dtensor, module: performance, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-09T17:05:00Z
issue,156533,`<<` and `>>` operators seem silently broken for DTensor operand 1 and scalar operand 2,https://github.com/pytorch/pytorch/issues/156533,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dtensor, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-09T17:04:57Z
issue,156784,[FSDP2] allow different dtypes for the model params with gradients,https://github.com/pytorch/pytorch/issues/156784,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: fsdp, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-09T17:04:44Z
issue,155589,"pipeline() fails when a sub-module uses ""no_grad()""; impacts RoPE implementation on HF models",https://github.com/pytorch/pytorch/issues/155589,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: pipelining, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-09T17:03:54Z
issue,153484,Pytorch PP requires all parameters to have grad in backward,https://github.com/pytorch/pytorch/issues/153484,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: pipelining, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-09T17:02:46Z
issue,143721,"add ""enabled=True"" to DistributedDataParallel.no_sync()",https://github.com/pytorch/pytorch/issues/143721,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: ddp, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-09T17:00:45Z
issue,144608,Build breaks on FreeBSD on arm platforms: Unrecognized CMAKE_SYSTEM_NAME = FreeBSD,https://github.com/pytorch/pytorch/issues/144608,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: build,"decide whether this is bug, use case, docs, or support",2026-06-09T10:15:41Z
issue,125604,Compiling GPT2 with tensorrt backend and dynamic shapes results in guard failures.,https://github.com/pytorch/pytorch/issues/125604,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dynamic shapes, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-06-09T02:05:42Z
issue,182259,DISABLED test_torchvision_models_convnext_base (__main__.TestVisionTracing),https://github.com/pytorch/pytorch/issues/182259,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, module: fx","decide whether this is bug, use case, docs, or support",2026-06-09T01:24:27Z
issue,178153,[RFC] Refactor Triton Heuristics via Unified Device-Scoped Class Hierarchy,https://github.com/pytorch/pytorch/issues/178153,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: inductor, module: performance, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-06-08T23:43:29Z
issue,167885,Ideas for Simple CUDA Dynamic Analysis Tools for Common Bugs,https://github.com/pytorch/pytorch/issues/167885,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: cuda, module: performance","decide whether this is bug, use case, docs, or support",2026-06-08T17:49:37Z
issue,142356,AllReduce hangs due to no device_id into init_process_group,https://github.com/pytorch/pytorch/issues/142356,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: c10d, module: nccl, oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-08T17:16:53Z
issue,142095,torch.distributed.new_group got stuck when use backend=mpi,https://github.com/pytorch/pytorch/issues/142095,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: mpi, oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-08T17:16:30Z
issue,140069,DTensor support for fused qkv matmul,https://github.com/pytorch/pytorch/issues/140069,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dtensor, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-08T17:16:28Z
issue,139129,[pgnccl] unstable restarting of PGs,https://github.com/pytorch/pytorch/issues/139129,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: c10d, oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-08T17:16:24Z
issue,137014,[RFC] Allow user to abort group (followed by custom recovery),https://github.com/pytorch/pytorch/issues/137014,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: c10d, oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-08T17:15:24Z
issue,137010,[RFC] Investigate multi-thread support of c10d,https://github.com/pytorch/pytorch/issues/137010,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: c10d, oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-08T17:15:18Z
issue,136950,Incomplete Parameter Gathering on Rank 0 with FSDP Model Saving,https://github.com/pytorch/pytorch/issues/136950,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: fsdp, oncall: distributed, oncall: distributed checkpointing","decide whether this is bug, use case, docs, or support",2026-06-08T17:15:08Z
issue,135358,"UserWarning: Grad strides do not match bucket view strides. This may indicate grad was not created according to the gradient layout contract, or that the param's strides changed since DDP was constructed. This is not an error, but may impair performance. grad.sizes() = [1, 64, 64, 64], strides() = [1, 4096, 64, 1] bucket_view.sizes() = [1, 64, 64, 64], strides() = [262144, 4096, 64, 1]",https://github.com/pytorch/pytorch/issues/135358,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: ddp, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-08T17:14:44Z
issue,136390,[c10d][nccl][cuda] Regression (unspecific cuda launch error) with test_c10d_nncl,https://github.com/pytorch/pytorch/issues/136390,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: nccl, oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-08T17:14:20Z
issue,129862,The call to ncclCommSplit does not have the correct config parameters set,https://github.com/pytorch/pytorch/issues/129862,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: nccl, module: performance, oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-08T17:14:00Z
issue,124058,[RFC] Support Torchrun for Out-of-the-tree Devices,https://github.com/pytorch/pytorch/issues/124058,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: elastic, oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-08T17:13:08Z
issue,186004,git fetch error: not our ref a56dcd79c699366e7ac6466792c3025883ff7704,https://github.com/pytorch/pytorch/issues/186004,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: build, module: third_party","decide whether this is bug, use case, docs, or support",2026-06-08T14:27:58Z
issue,186521,Feature Request: Add CPU-only PyTorch wheels for Termux (Android ARM64) + Python 3.13 (cp313),https://github.com/pytorch/pytorch/issues/186521,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: arm, module: binaries","decide whether this is bug, use case, docs, or support",2026-06-08T05:00:29Z
issue,186548,"`ProcessGroupXCCL` never overrides `Backend::supportsSplitting()`, breaking `DeviceMesh._unflatten` (and any `dist.split_group` caller) on XPU",https://github.com/pytorch/pytorch/issues/186548,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: xpu, oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-08T03:52:30Z
issue,168044,RFC: Internal Unification of LRScheduler Semantics for Correct Resuming and Composition,https://github.com/pytorch/pytorch/issues/168044,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: lrscheduler, module: optimizer","decide whether this is bug, use case, docs, or support",2026-06-07T20:34:06Z
issue,120158,"The forward accuracy cannot be aligned when using DTensor and ColwiseParallel, RowwiseParallel for TP training on the Llama2",https://github.com/pytorch/pytorch/issues/120158,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dtensor, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-07T16:39:34Z
issue,116683,DTensor tensor subwrapper not be handle with apex,https://github.com/pytorch/pytorch/issues/116683,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dtensor, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-07T16:39:11Z
issue,121594,[DDP] Gradient Synchronization Failure Induced by model.gradient_checkpointing_enable() in multi-task setting,https://github.com/pytorch/pytorch/issues/121594,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: activation checkpointing, module: ddp, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-07T16:38:58Z
issue,120945,DISABLED distributed/_composable/fsdp/test_fully_shard_training.py::TestFullyShard1DTrainingCompose::test_train_parity_with_activation_checkpointing,https://github.com/pytorch/pytorch/issues/120945,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: activation checkpointing, module: fsdp, module: rocm, oncall: distributed","decide whether this is bug, use case, docs, or support",2026-06-07T16:38:42Z
issue,120741,[C10D] coalesced op hang does not respect timeout,https://github.com/pytorch/pytorch/issues/120741,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: c10d, oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-07T16:38:32Z
issue,114884,Undefined reference to ncclCommSplit,https://github.com/pytorch/pytorch/issues/114884,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: nccl, oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-07T16:38:28Z
issue,115861,gather_object does not work with ThreadPoolExecutor,https://github.com/pytorch/pytorch/issues/115861,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: c10d, oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-07T16:37:49Z
issue,113281,NCCL p2p ops hung after communicator aborts,https://github.com/pytorch/pytorch/issues/113281,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: c10d, oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-07T16:37:47Z
issue,109477,[DTensor] optimizer step performance is still too bad,https://github.com/pytorch/pytorch/issues/109477,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dtensor, module: performance, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-07T16:37:14Z
issue,106851,RPC all_gather doesn't work with dynamic world size (world_size=None),https://github.com/pytorch/pytorch/issues/106851,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: rpc, oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-07T16:37:14Z
issue,114437,[Help Needed] [DataParallel/DataLoader/AMP] Severe slowdown on large dataset with fp16 precision and DataParallel wrapper,https://github.com/pytorch/pytorch/issues/114437,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: data parallel, module: dataloader, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-07T16:37:08Z
issue,90052,DDP overlapped optimizer: set grads to None enhancements,https://github.com/pytorch/pytorch/issues/90052,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: data parallel, module: ddp, module: performance, oncall: distributed","decide whether this is bug, use case, docs, or support",2026-06-07T16:36:06Z
issue,171938,Device id in distributed process group with multiple backends,https://github.com/pytorch/pytorch/issues/171938,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: oncall: distributed,"decide whether this is bug, use case, docs, or support",2026-06-07T15:17:56Z
issue,186496,"Typo in PyTorch 2.12 quantization docs: ""cetralizing"" should be ""centralizing""",https://github.com/pytorch/pytorch/issues/186496,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: docs,"decide whether this is bug, use case, docs, or support",2026-06-07T15:05:24Z
issue,56340,log_softmax(x) != x - logsumexp(x),https://github.com/pytorch/pytorch/issues/56340,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: numerical-stability,"decide whether this is bug, use case, docs, or support",2026-06-07T03:04:19Z
issue,128396,torch.prod cannot be used with cudagraphs,https://github.com/pytorch/pytorch/issues/128396,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: cuda graphs,"decide whether this is bug, use case, docs, or support",2026-06-06T20:12:21Z
issue,95895,[PTD] dist.barrier() unreliable when using collectives from multiple threads.,https://github.com/pytorch/pytorch/issues/95895,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: c10d, oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-06T16:35:30Z
issue,91629,vmap + nn.SyncBatchNorm.convert_sync_batchnorm,https://github.com/pytorch/pytorch/issues/91629,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: data parallel, module: ddp, module: functorch, oncall: distributed","decide whether this is bug, use case, docs, or support",2026-06-06T16:35:22Z
issue,90848,[Distributed] Destruction order fiasco in ProcessGroupNCCL workCleanupLoop(),https://github.com/pytorch/pytorch/issues/90848,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: c10d, module: nccl, oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-06T16:35:21Z
issue,88838,[RFC] PyTorch DistributedTensor,https://github.com/pytorch/pytorch/issues/88838,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dtensor, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-06T16:34:10Z
issue,72025,[c10d] destruction of Store objects,https://github.com/pytorch/pytorch/issues/72025,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: c10d, oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-06T16:30:28Z
issue,72034,Bug about distributed launch,https://github.com/pytorch/pytorch/issues/72034,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-06T16:30:24Z
issue,72099,halt or exit function implementation,https://github.com/pytorch/pytorch/issues/72099,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: c10d, oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-06T16:30:20Z
issue,72107,DistributedDataParallel creates too many threads,https://github.com/pytorch/pytorch/issues/72107,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dataloader, module: ddp, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-06T16:30:16Z
issue,72346,torch.distributed.new_group() should consolidate the type of rank list,https://github.com/pytorch/pytorch/issues/72346,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: c10d, oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-06T16:30:12Z
issue,72366,Custom ProcessGroup Destructor Not Correctly Called in PT 1.10,https://github.com/pytorch/pytorch/issues/72366,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: c10d, module: ddp, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-06T16:30:07Z
issue,72435,ProcessGroupWrapper: Additional Improvements,https://github.com/pytorch/pytorch/issues/72435,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: c10d, module: ddp, oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-06T16:30:03Z
issue,72835,torch.distributed hangs at barrier(),https://github.com/pytorch/pytorch/issues/72835,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: c10d, oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-06T16:29:58Z
issue,73081,Additional out of the box DDP comm hooks,https://github.com/pytorch/pytorch/issues/73081,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: ddp, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-06T16:29:53Z
issue,73332,BatchNorm with LSTM in DistributedDataParallel throws one of the variables needed for gradient computation has been modified by an inplace operation,https://github.com/pytorch/pytorch/issues/73332,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: ddp, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-06T16:29:45Z
issue,73960,processes hang when executing cross-machine asynchronous P2P communication on NCCL backend,https://github.com/pytorch/pytorch/issues/73960,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: c10d, module: nccl, oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-06T16:29:36Z
issue,74162,consume_prefix_in_state_dict_if_present can not remove prefix of _metadata.,https://github.com/pytorch/pytorch/issues/74162,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: ddp, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-06T16:29:31Z
issue,74323,[c10d] `gather`/`scatter` inconsistency behavior on non-dst/src rank,https://github.com/pytorch/pytorch/issues/74323,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: c10d, oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-06T16:29:27Z
issue,74528,Supports for dist.send/dist.recv sending and recving torch.shorttensor,https://github.com/pytorch/pytorch/issues/74528,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: c10d, oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-06T16:29:23Z
issue,74602,My_rank in the implementation of torch.distributed.scatter_object_list should be global,https://github.com/pytorch/pytorch/issues/74602,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: c10d, oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-06T16:29:19Z
issue,74666,Re-initializing torch.distributed process_group hangs with destroy_process_group,https://github.com/pytorch/pytorch/issues/74666,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-06T16:29:15Z
issue,74734,RuntimeError: Unconvertible NCCL type Short when sending torch.cuda.ShortTensor.,https://github.com/pytorch/pytorch/issues/74734,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: c10d, oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-06T16:29:10Z
issue,74824,C10d Elastic Training master_addr ERROR,https://github.com/pytorch/pytorch/issues/74824,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra, oncall: r2p","decide whether this is bug, use case, docs, or support",2026-06-06T16:29:02Z
issue,75031,comm hook error in BWD pass,https://github.com/pytorch/pytorch/issues/75031,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: ddp, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-06T16:28:58Z
issue,75087,extend torch.distributed's `--tee` to log the nodename,https://github.com/pytorch/pytorch/issues/75087,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: elastic, oncall: distributed, oncall: distributed infra, oncall: r2p","decide whether this is bug, use case, docs, or support",2026-06-06T16:28:53Z
issue,75097,destroy_process_group() has a certain probability of hangs,https://github.com/pytorch/pytorch/issues/75097,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-06T16:28:49Z
issue,75334,ProcessGroupNCCL is relying on UB to support bool data type,https://github.com/pytorch/pytorch/issues/75334,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: nccl, oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-06T16:28:45Z
issue,75657,1.11.0 distribution train different with 1.8.1,https://github.com/pytorch/pytorch/issues/75657,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-06T16:28:40Z
issue,75936,AllGather with backward support async_op=True,https://github.com/pytorch/pytorch/issues/75936,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-06T16:28:32Z
issue,76103,Some test failed when running in parallel.,https://github.com/pytorch/pytorch/issues/76103,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-06T16:28:28Z
issue,76287,torch.elastic fails to shutdown despite crashed processes,https://github.com/pytorch/pytorch/issues/76287,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: elastic, oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-06T16:28:19Z
issue,76387,[checkpoint] Make prepare_sharded_tensor_read and prepare_sharded_tensor_write public,https://github.com/pytorch/pytorch/issues/76387,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed checkpointing","decide whether this is bug, use case, docs, or support",2026-06-06T16:28:15Z
issue,76389,Error in DistributedDataParallel with 'CPU' device,https://github.com/pytorch/pytorch/issues/76389,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: ddp, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-06T16:28:11Z
issue,76395,[checkpoint] SPMD distributed checkpoint coordination,https://github.com/pytorch/pytorch/issues/76395,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed checkpointing","decide whether this is bug, use case, docs, or support",2026-06-06T16:28:07Z
issue,76527,Non target rank receives result of 'reduce' op when backend is 'gloo',https://github.com/pytorch/pytorch/issues/76527,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-06T16:27:56Z
issue,148420,Float8_e4m3fn,https://github.com/pytorch/pytorch/issues/148420,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: floatx (formerly float8), module: mps","decide whether this is bug, use case, docs, or support",2026-06-06T07:18:25Z
issue,125557,result of 2 ** s differs between eager mode and inductor + triton + cuda when in float32 denormal range,https://github.com/pytorch/pytorch/issues/125557,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: oncall: pt2,"decide whether this is bug, use case, docs, or support",2026-06-06T06:32:16Z
issue,39242,Add numerically stable log1mexp = log(1 - exp(-|x|)) function,https://github.com/pytorch/pytorch/issues/39242,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: numerical-stability,"decide whether this is bug, use case, docs, or support",2026-06-06T04:26:47Z
issue,186405,arch_id.h: error: template constraint not satisfied,https://github.com/pytorch/pytorch/issues/186405,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: build, module: cuda, module: dependency bug","decide whether this is bug, use case, docs, or support",2026-06-06T03:21:32Z
issue,113708,`log_softmax` could be `2**124` to `2**1021` times more accurate on small outputs,https://github.com/pytorch/pytorch/issues/113708,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: nn,"decide whether this is bug, use case, docs, or support",2026-06-06T00:07:37Z
issue,186031,[torchcomms] FSDP+TP+PP+compile torchtitan flavor fails: functional collective cannot resolve torchcomms-managed TP process group under torch.compile,https://github.com/pytorch/pytorch/issues/186031,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: torchcomms, oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T22:37:10Z
issue,76760,hanging process with init_process_group(backend='mpi') cannot be killed,https://github.com/pytorch/pytorch/issues/76760,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T17:13:03Z
issue,76913,Pytorch return TCPStore( RuntimeError: Connection reset by peer),https://github.com/pytorch/pytorch/issues/76913,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: c10d, oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T17:12:54Z
issue,76927,Can't pickle model torch._C._distributed_c10d.ProcessGroupNCCL' object,https://github.com/pytorch/pytorch/issues/76927,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: c10d, module: ddp, oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T17:12:49Z
issue,77154,Distributed Weighted Sampler.,https://github.com/pytorch/pytorch/issues/77154,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: distributed_tool, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-05T17:12:44Z
issue,77523,DDP window TCP bug [socket.cpp:558] [c10d] The client socket has failed to,https://github.com/pytorch/pytorch/issues/77523,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: c10d, oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T17:12:39Z
issue,78638,"torch.distributed.init_process_group(backend=""nccl"") NCCL version error",https://github.com/pytorch/pytorch/issues/78638,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T17:12:34Z
issue,78842,Add TORCH_SHOW_CPP_STACKTRACES when TORCH_DISTRIBUTED_DEBUG = detail,https://github.com/pytorch/pytorch/issues/78842,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: bootcamp, module: c10d, module: distributed_tool, oncall: distributed","decide whether this is bug, use case, docs, or support",2026-06-05T17:12:29Z
issue,78885,[distributed_test.py] Improve `test_barrier`,https://github.com/pytorch/pytorch/issues/78885,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: tests, oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T17:12:24Z
issue,79388,Init connect timeout when use torch.distributed.run,https://github.com/pytorch/pytorch/issues/79388,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra, oncall: r2p","decide whether this is bug, use case, docs, or support",2026-06-05T17:12:19Z
issue,80496,DDP find_unused_parameters=True does not work for Sparse gradients,https://github.com/pytorch/pytorch/issues/80496,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: ddp, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-05T17:12:14Z
issue,80595,Overlapping Optimizer.step() with DDP backward,https://github.com/pytorch/pytorch/issues/80595,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: ddp, module: optimizer, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-05T17:12:09Z
issue,80774,RPC: Make RRefProxy callable,https://github.com/pytorch/pytorch/issues/80774,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: rpc, oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T17:12:04Z
issue,80824,"When running GPT trainning with megatron, the program quit due to torch.distributed.elastic.agent.server.api:Received 1 death signal, shutting down workers",https://github.com/pytorch/pytorch/issues/80824,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: elastic, oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T17:11:59Z
issue,80832,[DDP] doesn't support multiple backwards when static_graph=True,https://github.com/pytorch/pytorch/issues/80832,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: ddp, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-05T17:11:54Z
issue,81383,Investigate ncclRedOpCreatePreMulSum operator for gradient reduction,https://github.com/pytorch/pytorch/issues/81383,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: ddp, module: fsdp, module: performance, oncall: distributed","decide whether this is bug, use case, docs, or support",2026-06-05T17:11:49Z
issue,81622,[Distributed] test_dynamic_rpc_existing_rank_can_communicate_with_new_rank_cuda fails in caching allocator,https://github.com/pytorch/pytorch/issues/81622,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: rpc, oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T17:11:43Z
issue,82153,DistributedDataParallel hangs when not using GPU 0,https://github.com/pytorch/pytorch/issues/82153,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: ddp, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-05T17:11:32Z
issue,85949,"[Distributed] Loading distributed checkpoint with FSDP fails with varying key errors (pos.embedding, shared.weight)",https://github.com/pytorch/pytorch/issues/85949,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed checkpointing","decide whether this is bug, use case, docs, or support",2026-06-05T17:11:12Z
issue,86456,`SyncBatchNorm` doesn't work with subclass of `torch.Tensor`,https://github.com/pytorch/pytorch/issues/86456,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: ddp, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-05T17:11:07Z
issue,88006,C++ Extensions can't import c10d/reducer.hpp,https://github.com/pytorch/pytorch/issues/88006,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: ddp, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-05T17:10:57Z
issue,88626,Cloud-based rendezvous backend / distributed store?,https://github.com/pytorch/pytorch/issues/88626,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T17:10:46Z
issue,89197,Collective operations do not work with `torch.BoolTensor`s on `gloo` and raise `Invalid scalar type`,https://github.com/pytorch/pytorch/issues/89197,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T17:10:41Z
issue,89546,Unable to launch CUDA Graph with DDP model,https://github.com/pytorch/pytorch/issues/89546,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: ddp, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-05T17:10:37Z
issue,89716,DDP hangs on forward pass of transformer,https://github.com/pytorch/pytorch/issues/89716,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: ddp, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-05T17:10:32Z
issue,90171,Option to let DistributedDataParallel know in advance unused parameters at each forward pass,https://github.com/pytorch/pytorch/issues/90171,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: ddp, module: performance, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-05T17:10:27Z
issue,90245,[Distributed] `Invalid scalar type` when `dist.scatter()` boolean tensor,https://github.com/pytorch/pytorch/issues/90245,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T17:10:22Z
issue,90509,Can torchrun have a shell completion?,https://github.com/pytorch/pytorch/issues/90509,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T17:10:17Z
issue,90688,In distributed get SIGTERM and run crash,https://github.com/pytorch/pytorch/issues/90688,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: ddp, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-05T17:10:13Z
issue,91408,`NotImplementedError` when using `torch.distributed.launch`,https://github.com/pytorch/pytorch/issues/91408,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: ddp, module: functorch, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-05T17:10:08Z
issue,92694,API to check for errors in c10d.ProcessGroupNCCL,https://github.com/pytorch/pytorch/issues/92694,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: c10d, module: nccl, oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T17:09:57Z
issue,95724,"distributed training: lots of ""Exception ignored"" at the end of each epoch",https://github.com/pytorch/pytorch/issues/95724,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: ddp, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-05T17:09:52Z
issue,99614,Make broadcast_coalesced to a op for processgroup,https://github.com/pytorch/pytorch/issues/99614,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T17:09:42Z
issue,99652,DistributedDataParallel doesn't work with complex buffers,https://github.com/pytorch/pytorch/issues/99652,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: complex, module: ddp, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-05T17:09:37Z
issue,99908,Deprecate torch.distributed.algorithms._optimizer_overlap,https://github.com/pytorch/pytorch/issues/99908,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: ddp, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-05T17:09:32Z
issue,99978,torch.jit.trace can not trace buffer by Module.register_buffer() when use DDP Module.,https://github.com/pytorch/pytorch/issues/99978,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: ddp, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-05T17:09:27Z
issue,100582,"Synchronization issue when combining DPP and RPC - ""Parameter marked twice""",https://github.com/pytorch/pytorch/issues/100582,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: ddp, module: rpc, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-05T17:09:22Z
issue,101402,DataParallel for nested modules,https://github.com/pytorch/pytorch/issues/101402,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: data parallel, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-05T17:09:17Z
issue,102197,torch.func.jvp fails when acting on a DistributedDataParallel model,https://github.com/pytorch/pytorch/issues/102197,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: ddp, module: forward ad, module: functorch, oncall: distributed","decide whether this is bug, use case, docs, or support",2026-06-05T17:09:06Z
issue,102402,[Composable] Unified way to check if modules are managed by composable API,https://github.com/pytorch/pytorch/issues/102402,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: fsdp, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-05T17:09:01Z
issue,102438,DDP multi node multi gpu inconsistent params,https://github.com/pytorch/pytorch/issues/102438,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: ddp, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-05T17:08:56Z
issue,103622,ARM based GPU support for Distributed Data Parallelism Module,https://github.com/pytorch/pytorch/issues/103622,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: ddp, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-05T17:08:51Z
issue,103625,DISABLED test_backward_ddp_outside (__main__.TensorPipeDdpUnderDistAutogradTest),https://github.com/pytorch/pytorch/issues/103625,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T17:08:46Z
issue,103626,2D model checkpointing hangs on a ViT model,https://github.com/pytorch/pytorch/issues/103626,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed checkpointing","decide whether this is bug, use case, docs, or support",2026-06-05T17:08:41Z
issue,104011,DISABLED test_backward_ddp_outside_uneven_inputs (__main__.TensorPipeDdpUnderDistAutogradTest),https://github.com/pytorch/pytorch/issues/104011,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T17:08:36Z
issue,104164,DDP enhancement,https://github.com/pytorch/pytorch/issues/104164,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: ddp, module: performance, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-05T17:08:31Z
issue,105723,Parameter ... has been marked as ready twice,https://github.com/pytorch/pytorch/issues/105723,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: ddp, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-05T17:08:21Z
issue,107174,Surface NCCL and CUDA version incompatibility,https://github.com/pytorch/pytorch/issues/107174,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: nccl, oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T17:08:11Z
issue,107177,Timeout during NCCL initialization due to store,https://github.com/pytorch/pytorch/issues/107177,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T17:08:07Z
issue,108212,Is there a standard procedure to check the consistency of environment across all nodes in PyTorch DDP training?,https://github.com/pytorch/pytorch/issues/108212,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: distributed_tool, oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T17:08:02Z
issue,108381,"FSDP always puts parameters to fp32 when loading state_dict, even if state_dict has bf16 params",https://github.com/pytorch/pytorch/issues/108381,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: fsdp, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-05T17:07:57Z
issue,109103,"DDP - ""No backend type associated with device type cpu"" with new Model Phi 1.5 despite everything loaded on GPUs",https://github.com/pytorch/pytorch/issues/109103,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: ddp, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-05T17:07:52Z
issue,110281,feat(Pipeline Parallelism): use mincut optimization for local communication optimization,https://github.com/pytorch/pytorch/issues/110281,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: performance, module: pipelining, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-05T17:07:47Z
issue,110775,torch.distributed.pipeline skip module throws assert error that portal.grad is not None,https://github.com/pytorch/pytorch/issues/110775,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: pipelining, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-05T17:07:42Z
issue,111131,"""Invalid Scalar type"" when using bf16 allreduce with Gloo backend",https://github.com/pytorch/pytorch/issues/111131,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T17:07:37Z
issue,111646,torchrun: elastic training not restarted on missing keep-alive heartbeat/scale-down event,https://github.com/pytorch/pytorch/issues/111646,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T17:07:27Z
issue,111834,"Numerical inaccuracies in ""ddp_apply_optim_in_backward"" unit tests for gloo backend",https://github.com/pytorch/pytorch/issues/111834,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: ddp, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-05T17:07:22Z
issue,111943,Implement scatter and broadcast in functional collectives,https://github.com/pytorch/pytorch/issues/111943,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T17:07:17Z
issue,112029,DDPWrapper,https://github.com/pytorch/pytorch/issues/112029,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: ddp, module: fsdp, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-05T17:07:12Z
issue,112164,[Tracking] torch.compile + torch.distributed + set_grad_enabled/autocast,https://github.com/pytorch/pytorch/issues/112164,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: ddp, module: fsdp, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-05T17:07:02Z
issue,112277,test_DistributedDataParallel fails with static_graph=True,https://github.com/pytorch/pytorch/issues/112277,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: ddp, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-05T17:06:57Z
issue,113114,Allow to construct distributed Work object from Future,https://github.com/pytorch/pytorch/issues/113114,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T17:06:53Z
issue,113376,Add validation for send/recv sizes,https://github.com/pytorch/pytorch/issues/113376,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T17:06:48Z
issue,114765,DistributedDataParallel _verify_param_shape_across_processes causes reserved memory on GPU,https://github.com/pytorch/pytorch/issues/114765,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: ddp, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-05T17:06:43Z
issue,114857,recursive search for DDP ignored parameters,https://github.com/pytorch/pytorch/issues/114857,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: ddp, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-05T17:06:38Z
issue,116183,Pipeline parallel doesn't seem to work with mixed precision,https://github.com/pytorch/pytorch/issues/116183,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: pipelining, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-05T17:06:28Z
issue,117159,"If error is raised in distributed setting, top level exception formatting should include rank",https://github.com/pytorch/pytorch/issues/117159,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T17:06:23Z
issue,117755,[c10d] Enqueue work pointers instead of work copies,https://github.com/pytorch/pytorch/issues/117755,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T17:06:17Z
issue,118094,DTensorConverter::successful returns False on no Tensor args causing test failures,https://github.com/pytorch/pytorch/issues/118094,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dtensor, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-05T17:06:12Z
issue,119758,Flight Recorder dump_entries() segfaults when used with coalesced operations,https://github.com/pytorch/pytorch/issues/119758,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T17:06:06Z
issue,119856,Timeout for process termination upon receiving sigterm should be a user configurable parameter,https://github.com/pytorch/pytorch/issues/119856,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T17:05:56Z
issue,119890,dynamo remap collective improvement,https://github.com/pytorch/pytorch/issues/119890,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T17:05:51Z
issue,120516,[C10D] coalesced collectives are not well tested,https://github.com/pytorch/pytorch/issues/120516,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T17:05:41Z
issue,120787,Tensor Parallel Inference Performance Issue with Llama-7b-chat-hf Model caused by straggler problem,https://github.com/pytorch/pytorch/issues/120787,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dtensor, module: performance, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-05T17:05:36Z
issue,121181,Profiling tool for DDP and FSDP,https://github.com/pytorch/pytorch/issues/121181,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: ddp, module: distributed_tool, module: fsdp, oncall: distributed","decide whether this is bug, use case, docs, or support",2026-06-05T17:05:31Z
issue,121486,Kubernetes API server as Rendezvous backend,https://github.com/pytorch/pytorch/issues/121486,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: elastic, oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T17:05:26Z
issue,121543,[DCP] test_state_dict.test_fsdp is failing on unrelated tests in CI.,https://github.com/pytorch/pytorch/issues/121543,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed checkpointing","decide whether this is bug, use case, docs, or support",2026-06-05T17:05:20Z
issue,121587,Add clearer docs for `torch.distributed.<OP>` vs. `torch.distributed.nn.<OP>,https://github.com/pytorch/pytorch/issues/121587,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: docs, oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T17:05:16Z
issue,121605,Make the data type of local_used_map_ in the Reducer class configurable,https://github.com/pytorch/pytorch/issues/121605,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T17:05:11Z
issue,123815,my_role_info = role_infos group_rank index error,https://github.com/pytorch/pytorch/issues/123815,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T17:04:56Z
issue,124019,[FSDP+TP] RuntimeError: 'weight' must be 2-D,https://github.com/pytorch/pytorch/issues/124019,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dtensor, module: fsdp, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-05T17:04:40Z
issue,124101,[C10D] Standardize torch distributed launcher interface,https://github.com/pytorch/pytorch/issues/124101,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T17:04:35Z
issue,124204,RPC + DDP VRAM overflow to rank 0 GPU.,https://github.com/pytorch/pytorch/issues/124204,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T17:04:31Z
issue,124260,[FSDP] Unreleased reserved memory after FSDP wrap,https://github.com/pytorch/pytorch/issues/124260,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: fsdp, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-05T17:04:26Z
issue,124317,Many tests under test/distributed/elastic not running in OSS CI,https://github.com/pytorch/pytorch/issues/124317,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra, oncall: r2p","decide whether this is bug, use case, docs, or support",2026-06-05T17:04:20Z
issue,124372,Pipeline Parallelism support for ZeroBubble,https://github.com/pytorch/pytorch/issues/124372,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: pipelining, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-05T17:04:15Z
issue,124380,Error when running DDP examples to train between nodes,https://github.com/pytorch/pytorch/issues/124380,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: c10d, module: elastic, module: nccl, oncall: distributed","decide whether this is bug, use case, docs, or support",2026-06-05T17:04:11Z
issue,124468,Race in ProcessGroupNCCL shutdown,https://github.com/pytorch/pytorch/issues/124468,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: c10d, module: nccl, oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T17:04:05Z
issue,124680,Inconsistent results when training a model containing SyncBatchNorm with multiple GPUs,https://github.com/pytorch/pytorch/issues/124680,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: ddp, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-05T17:03:55Z
issue,126302,Make it easier to run distributed tests without run_test.py,https://github.com/pytorch/pytorch/issues/126302,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: distributed_tool, oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T17:03:50Z
issue,131737,Intra-Node one-shot/two-shot low latency kernels for fp32,https://github.com/pytorch/pytorch/issues/131737,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T17:03:30Z
issue,131739,One Shot Low-Latency kernel is 2x higher than expected,https://github.com/pytorch/pytorch/issues/131739,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: performance, oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T17:03:25Z
issue,131754,Coredump for pytorch ddp when job finished,https://github.com/pytorch/pytorch/issues/131754,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: cuda, oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T17:03:20Z
issue,135178,[DTensor] loss_parallel not worked with shifted logits and labels,https://github.com/pytorch/pytorch/issues/135178,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dtensor, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-05T17:03:14Z
issue,135511,RuntimeError: Modules with uninitialized parameters can't be used with `DistributedDataParallel` when using subclass of tensor.,https://github.com/pytorch/pytorch/issues/135511,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: ddp, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-05T17:03:10Z
issue,136122,[Distributed] Decouple ProcessGroupNCCL's regular exit from abort,https://github.com/pytorch/pytorch/issues/136122,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T17:03:05Z
issue,136187,Unable to Specify CUDA Stream for Collective Operations Using with torch.cuda.stream() context,https://github.com/pytorch/pytorch/issues/136187,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: cuda, oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T17:03:00Z
issue,136226,[pipelining] try not to dry run module when creating PipelineStage?,https://github.com/pytorch/pytorch/issues/136226,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: performance, module: pipelining, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-05T17:02:55Z
issue,136248,Async NCCL communication blocks CUDA kernel in the first run,https://github.com/pytorch/pytorch/issues/136248,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T17:02:50Z
issue,136363,[pipelining] output shape validation / lazy inference,https://github.com/pytorch/pytorch/issues/136363,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: pipelining, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-05T17:02:41Z
issue,136380,[pipelining] Improve recv buffer management,https://github.com/pytorch/pytorch/issues/136380,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: performance, module: pipelining, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-05T17:02:35Z
issue,136408,[RFC] Composable parallel documentation,https://github.com/pytorch/pytorch/issues/136408,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: fsdp, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-05T17:02:29Z
issue,136425,Any updates on AsyncCollectiveTensor support for all-gather along non-zero dims?,https://github.com/pytorch/pytorch/issues/136425,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T17:02:25Z
issue,136567,[RFC] Offload collectives to NVSwitch when possible,https://github.com/pytorch/pytorch/issues/136567,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: cuda, oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T17:02:20Z
issue,136866,fused_scaled_matmul_reduce_scatter report error with channel-wise scaling,https://github.com/pytorch/pytorch/issues/136866,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T17:02:15Z
issue,139280,Possible bug of tools::flight_recorder,https://github.com/pytorch/pytorch/issues/139280,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T17:02:09Z
issue,139581,Suggestion : enable one to adjust the interval between calls to the pytorch NCCL watchdog thread via env variable,https://github.com/pytorch/pytorch/issues/139581,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: nccl, module: performance, oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T17:01:59Z
issue,139681,Unexpected type cast from DTensor to Tensor when applying SequenceParallel on RMSNorm.,https://github.com/pytorch/pytorch/issues/139681,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dtensor, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-05T17:01:54Z
issue,139739,[Distributed][collectiveCoalesced] Why function(check_gpu_tensors_same_device) only works in allreduce_coalesced?,https://github.com/pytorch/pytorch/issues/139739,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T17:01:50Z
issue,140107,Support Joinable loss functions with DDP,https://github.com/pytorch/pytorch/issues/140107,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: ddp, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-05T17:01:45Z
issue,140227,"`Torch.distributed.tensor.parallel.Style.Rowwiseparallel` when send tensor, the use of the underlying operator is `aten.to.dtype_layout`, lack of distributed strategy. But after adding a policy, `AsyncCollectiveTensor` no placements error",https://github.com/pytorch/pytorch/issues/140227,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dtensor, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-05T17:01:40Z
issue,141237,test_fsdp_tp_integration fails for non-power-of 2 GPUs,https://github.com/pytorch/pytorch/issues/141237,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: fsdp, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-05T17:01:30Z
issue,144239,Integrate ULFM and FTI for OpenMPI (torch.distributed),https://github.com/pytorch/pytorch/issues/144239,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T17:01:20Z
issue,144779,Adding Infiniband to RDZV Backend for optimal torch run training,https://github.com/pytorch/pytorch/issues/144779,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: docs, oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T17:01:11Z
issue,144851,torch.distributed hangs between Linux (X86) and Mac (M2 Pro),https://github.com/pytorch/pytorch/issues/144851,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T17:01:07Z
issue,144856,Connection Limitation in PyTorch Distributed (Vanilla) with c10d Rendezvous Backend,https://github.com/pytorch/pytorch/issues/144856,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T17:01:02Z
issue,145170,CUDA initialization error with vLLM 0.5.4 and PyTorch 2.4.0+cu121,https://github.com/pytorch/pytorch/issues/145170,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T17:00:57Z
issue,145376,distributed.new_group with backend GLOO hangs when distributed.split_group was called before,https://github.com/pytorch/pytorch/issues/145376,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T17:00:53Z
issue,146094,[CI][Distributed] @skip_if_lt_x_gpu(2) seems to be broken,https://github.com/pytorch/pytorch/issues/146094,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: ci, module: tests, oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T17:00:48Z
issue,146326,Tracking issue for Improvements for DeepSeek,https://github.com/pytorch/pytorch/issues/146326,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: floatx (formerly float8), module: sdpa, oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T17:00:43Z
issue,146329,DeepSeek: a2a communication with metadata on the GPU,https://github.com/pytorch/pytorch/issues/146329,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T17:00:38Z
issue,146332,DeepSeek: fine-grained overlap,https://github.com/pytorch/pytorch/issues/146332,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: pipelining, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-05T17:00:28Z
issue,146565,use register_full_backward_hook with c10d in torch.compile and raise error,https://github.com/pytorch/pytorch/issues/146565,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T17:00:24Z
issue,146601,Enabling efficient model-level redistribution between FSPD-TP,https://github.com/pytorch/pytorch/issues/146601,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dtensor, module: fsdp, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-05T17:00:19Z
issue,146813,Oneshot AllReduce not being triggered when there's nested intra- and inter-node process groups,https://github.com/pytorch/pytorch/issues/146813,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T17:00:14Z
issue,152042,distrubuted: false positive Grad strides vs Bucket strides warning,https://github.com/pytorch/pytorch/issues/152042,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: ddp, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-05T17:00:03Z
issue,152427,Add switch to disable truncation to long list print,https://github.com/pytorch/pytorch/issues/152427,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T16:59:58Z
issue,152822,TorchRun: Option to specify which GPUs to run on,https://github.com/pytorch/pytorch/issues/152822,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T16:59:53Z
issue,153016,inconsistent grads between two types of `allgather`s,https://github.com/pytorch/pytorch/issues/153016,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: autograd, module: dtensor, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-05T16:59:49Z
issue,153122,[CUDA] test_c10d_nccl test_extra_cuda_context failure due to _helper_test_extra_cuda_context_by_memory,https://github.com/pytorch/pytorch/issues/153122,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: cuda, module: tests, oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T16:59:44Z
issue,153350,`distributed.checkpoint.state_dict.set_model_state_dict` returning incorrect `incompatible_keys` when `full_state_dict=True`,https://github.com/pytorch/pytorch/issues/153350,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed checkpointing","decide whether this is bug, use case, docs, or support",2026-06-05T16:59:40Z
issue,153369,Rank {rank} has different values for {name}: {scalar_tensor_value}.,https://github.com/pytorch/pytorch/issues/153369,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed checkpointing","decide whether this is bug, use case, docs, or support",2026-06-05T16:59:35Z
issue,153615,[DTensor] Backward Pass Failure with 2D Sharded DTensor After Mean Reduction,https://github.com/pytorch/pytorch/issues/153615,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dtensor, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-05T16:59:31Z
issue,153822,Passing string to devices in register_backend causes incorrect behavior,https://github.com/pytorch/pytorch/issues/153822,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T16:59:27Z
issue,154102,dist.init_process_group not work,https://github.com/pytorch/pytorch/issues/154102,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T16:59:22Z
issue,154116,Timeout in distributed training,https://github.com/pytorch/pytorch/issues/154116,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T16:59:18Z
issue,154297,Hangs and timeouts on dist.reduce_scatter on B200 GPU,https://github.com/pytorch/pytorch/issues/154297,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T16:59:14Z
issue,154327,`torch.distributed.checkpoint.state_dict.get_model_state_dict` does not update the state_dict._metadata key,https://github.com/pytorch/pytorch/issues/154327,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed checkpointing","decide whether this is bug, use case, docs, or support",2026-06-05T16:59:09Z
issue,154345,[RFC] : Remove Explicit Backend References from `torch.distributed` (`c10d`),https://github.com/pytorch/pytorch/issues/154345,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T16:59:04Z
issue,154370,[C10D] Blackwell B200: NotImplementedError: Could not run '_c10d_functional_autograd::all_to_all_single' with arguments from the 'CUDA' backend.,https://github.com/pytorch/pytorch/issues/154370,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T16:59:00Z
issue,154563,Distributed Breakpoint doesn't exit safely,https://github.com/pytorch/pytorch/issues/154563,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T16:58:55Z
issue,154621,functional all_gather does not work with scalar tensor and fail with torch.compile,https://github.com/pytorch/pytorch/issues/154621,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T16:58:50Z
issue,154787,Redistribute DTensor across different DeviceMeshes,https://github.com/pytorch/pytorch/issues/154787,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: devicemesh, module: dtensor, oncall: distributed, oncall: distributed checkpointing","decide whether this is bug, use case, docs, or support",2026-06-05T16:58:46Z
issue,155467,torch.distributed TCP init method binds socket to all interfaces in all cases,https://github.com/pytorch/pytorch/issues/155467,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T16:58:42Z
issue,155547,"[Misc] The keys of the sac_milp return dict were modified, but the test case was not updated",https://github.com/pytorch/pytorch/issues/155547,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: activation checkpointing, module: distributed_tool, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-05T16:58:37Z
issue,155664,[Misc] skip_if decorators aborts the test process in distributed/test_composability.py,https://github.com/pytorch/pytorch/issues/155664,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T16:58:33Z
issue,155892,[RFC] `DeviceGroup`: a mixin of `ProcessGroup` and `Backend`,https://github.com/pytorch/pytorch/issues/155892,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T16:58:29Z
issue,155916,[`torch.distributed`] Watchdog monitor error handling should match watchdog error handling,https://github.com/pytorch/pytorch/issues/155916,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T16:58:24Z
issue,155993,Keep gettting AssertionError: found no DeviceMesh from dtensor args for c10d.broadcast_.default!,https://github.com/pytorch/pytorch/issues/155993,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: ddp, module: dtensor, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-05T16:58:20Z
issue,156002,DCP save only saves one shard of tensor parallel model when using DP + TP,https://github.com/pytorch/pytorch/issues/156002,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed checkpointing","decide whether this is bug, use case, docs, or support",2026-06-05T16:58:16Z
issue,156130,[DDP][FSDP2] add unit test to showcase DDP mixed precision with FSDP2 mixed precision,https://github.com/pytorch/pytorch/issues/156130,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: ddp, module: fsdp, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-05T16:58:12Z
issue,156392,DataParallel gather NaN across multiple gpus,https://github.com/pytorch/pytorch/issues/156392,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: data parallel, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-05T16:58:08Z
issue,156627,`distributed.all_gather()` support for uneven tensors does not apply to all backends,https://github.com/pytorch/pytorch/issues/156627,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T16:58:03Z
issue,156842,FSDP2 + TP does not work,https://github.com/pytorch/pytorch/issues/156842,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dtensor, module: fsdp, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-05T16:57:59Z
issue,156942,Can't use pinned_memory to do async save ckpt when using torch.distributed.checkpoint.async_save,https://github.com/pytorch/pytorch/issues/156942,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed checkpointing","decide whether this is bug, use case, docs, or support",2026-06-05T16:57:55Z
issue,156966,Perf drop when running with FSDP and torch.compile,https://github.com/pytorch/pytorch/issues/156966,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: fsdp, module: performance, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-05T16:57:50Z
issue,157027,[RFC] Remove the FSDP data copy from compute stream critical path,https://github.com/pytorch/pytorch/issues/157027,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: fsdp, module: performance, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-05T16:57:46Z
issue,157318,Torch Elastic Wait timeout increase,https://github.com/pytorch/pytorch/issues/157318,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T16:57:41Z
issue,157395,[FSDP2] figure out the contract for mp_policy and tensor subclass extention,https://github.com/pytorch/pytorch/issues/157395,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: docs, module: fsdp, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-05T16:57:37Z
issue,157781,torch.distributed.checkpoint.state_dict.set_state_dict stucks with StateDictOptions(full_state_dict=True),https://github.com/pytorch/pytorch/issues/157781,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed checkpointing","decide whether this is bug, use case, docs, or support",2026-06-05T16:57:33Z
issue,157862,"Spurious ""Grad strides do not match bucket view strides"" warning for 1x1 convolution",https://github.com/pytorch/pytorch/issues/157862,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: ddp, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-05T16:57:28Z
issue,158016,"dist.all_to_all_single, when input tensors of different shapes result in undefined output behavior",https://github.com/pytorch/pytorch/issues/158016,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T16:57:24Z
issue,158443,DISABLED test_gradients_synchronizations (__main__.TensorPipeTensorPipeCudaDistAutogradTest),https://github.com/pytorch/pytorch/issues/158443,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T16:57:19Z
issue,158698,[fsdp2] Possible to fuse reduce-scatter (in fp32) + div + dtype conversion (to bf16) in post-backward?,https://github.com/pytorch/pytorch/issues/158698,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: fsdp, module: performance, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-05T16:57:15Z
issue,158775,Add docs for torchrun numa binding on website,https://github.com/pytorch/pytorch/issues/158775,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: docs, oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T16:57:06Z
issue,159563,About init_process_group with different backends,https://github.com/pytorch/pytorch/issues/159563,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T16:56:57Z
issue,159686,"c10d broadcast copies at::Tensor, creates footgun for downstream implementations",https://github.com/pytorch/pytorch/issues/159686,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T16:56:52Z
issue,159711,Support dict as input/output for pipeline parallelism,https://github.com/pytorch/pytorch/issues/159711,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: pipelining, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-05T16:56:48Z
issue,160006,Support NUMA Binding for Callable entrypoints to `elastic_launch`,https://github.com/pytorch/pytorch/issues/160006,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T16:56:44Z
issue,160130,Torchrun breaks with virtual environments,https://github.com/pytorch/pytorch/issues/160130,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: elastic, oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T16:56:39Z
issue,160731,Cannot initialise process group on CPU device with GLOO backend,https://github.com/pytorch/pytorch/issues/160731,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T16:56:35Z
issue,160732,GLOO backend should have MPS capability,https://github.com/pytorch/pytorch/issues/160732,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: mps, oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T16:56:32Z
issue,160950,fx.node._sort_key dones't work right with torch.distributed._functional_collectives,https://github.com/pytorch/pytorch/issues/160950,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T16:56:26Z
issue,161091,DTensor view op results in RuntimeError: shape '[38597376]' is invalid for input of size 38598144,https://github.com/pytorch/pytorch/issues/161091,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dtensor, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-05T16:56:22Z
issue,161153,torch.compile fails in FSDP due to .data assignment,https://github.com/pytorch/pytorch/issues/161153,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: fsdp, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-05T16:56:18Z
issue,161368,Incorrect offsets for 2D uneven sharding in DTensor in `compute_local_shape_and_global_offsetproduces`,https://github.com/pytorch/pytorch/issues/161368,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dtensor, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-05T16:56:14Z
issue,161776,FSDP&device_mesh Mesh_dim_names in init_device_mesh not work,https://github.com/pytorch/pytorch/issues/161776,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: devicemesh, module: fsdp, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-05T16:56:09Z
issue,162032,torch thread yield after launch nccl kernel,https://github.com/pytorch/pytorch/issues/162032,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T16:56:05Z
issue,162160,[BUG] Experts weight not release after backward when FSDP2 + moe expert parallel,https://github.com/pytorch/pytorch/issues/162160,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: fsdp, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-05T16:56:00Z
issue,162179,test_fully_shard_logging.py is not being run,https://github.com/pytorch/pytorch/issues/162179,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: fsdp, module: tests, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-05T16:55:56Z
issue,162237,Does it support new CPU group after passing `device_id` to `init_process_group`,https://github.com/pytorch/pytorch/issues/162237,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T16:55:51Z
issue,162290,local_output_for_attn: bool = False ( for Col/Row-wiseParallel ),https://github.com/pytorch/pytorch/issues/162290,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dtensor, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-05T16:55:47Z
issue,162638,Gradient Clipping in Pipeline Parallelism Schedules,https://github.com/pytorch/pytorch/issues/162638,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: autograd, module: pipelining, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-05T16:55:42Z
issue,162668,ReduceOp has incorrect py::class_ binding,https://github.com/pytorch/pytorch/issues/162668,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: c10d, oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T16:55:37Z
issue,162741,`torch.distributed.pipelining` Llama 2 and 3 Results Torch Dynamo Compiled Error,https://github.com/pytorch/pytorch/issues/162741,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: pipelining, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-05T16:55:32Z
issue,162745,[CI][CUDA][Distributed] python test/distributed/test_data_parallel.py TestDataParallel.test_data_parallel_rnn p1.allclose(p2) False,https://github.com/pytorch/pytorch/issues/162745,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: data parallel, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-05T16:55:28Z
issue,162748,[CI][CUDA][Distributed] test_3_level_hierarchical_model_averager Works on 4-GPU runner but requires explicit WORLD_SIZE=4 on 8-GPU runner,https://github.com/pytorch/pytorch/issues/162748,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T16:55:23Z
issue,162755,"Ideally, we should not use `require_exact_world_size(4)` if we only have 8-GPU runners",https://github.com/pytorch/pytorch/issues/162755,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: distributed_tool, module: tests, oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T16:55:19Z
issue,162871,[CI][CUDA][Distributed] TestDistBackendWithSpawn.test_gather_object Requires Exact World Size of 4,https://github.com/pytorch/pytorch/issues/162871,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: ci, oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T16:55:14Z
issue,162897,[CI][CUDA][Distributed][NVSHMEM][Triton] NVSHMEMTritonTest.test_triton_fence KeyError,https://github.com/pytorch/pytorch/issues/162897,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T16:55:10Z
issue,162917,[CI][CUDA][Distributed][Symmetric Memory] AsyncTPTest.test_fused_all_gather_matmul_native_symm_mem_input_False_is_b_row_major_False Failure,https://github.com/pytorch/pytorch/issues/162917,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: symm_mem, module: tests, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-05T16:55:05Z
issue,162940,[CI][CUDA][Distributed] test_fused_scaled_matmul_reduce_scatter_scatter_dim_0_rowwise_False fails due to accuracy mismatch,https://github.com/pytorch/pytorch/issues/162940,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T16:55:00Z
issue,163043,`parallelize_module` causing scatter,https://github.com/pytorch/pytorch/issues/163043,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dtensor, module: performance, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-05T16:54:55Z
issue,163546,Better nccl timeout handling,https://github.com/pytorch/pytorch/issues/163546,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T16:54:47Z
issue,163678,RuntimeError: NCCL error: unhandled cuda error,https://github.com/pytorch/pytorch/issues/163678,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T16:54:42Z
issue,163722,AdaptiveLogSoftmaxWithLoss with weight tying,https://github.com/pytorch/pytorch/issues/163722,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: ddp, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-05T16:54:38Z
issue,163913,test_triton_wait_until hangs,https://github.com/pytorch/pytorch/issues/163913,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T16:54:34Z
issue,164015,[CI][CUDA] test_multimem_one_shot_all_reduce_bfloat16_align_bytes_1 failures on B200 runner,https://github.com/pytorch/pytorch/issues/164015,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: cuda, module: symm_mem, oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T16:54:29Z
issue,165653,[distributed] test_symmetric_memory fails on odd-numbered GPU configurations (3 GPUs),https://github.com/pytorch/pytorch/issues/165653,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T16:54:25Z
issue,165933,[Distributed] fully_shard: support no_shard (ddp) strategy?,https://github.com/pytorch/pytorch/issues/165933,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: fsdp, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-05T16:54:16Z
issue,166523,DDP + MPI-based distributed backend: missing work/future completion assertion for Broadcast operations,https://github.com/pytorch/pytorch/issues/166523,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: ddp, module: mpi, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-05T16:54:02Z
issue,166930,[DTensor] DTensor randn inconsistent with single device behavior,https://github.com/pytorch/pytorch/issues/166930,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dtensor, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-05T16:53:58Z
issue,167036,FSDP2 moves my inputs to device when I don't want it to,https://github.com/pytorch/pytorch/issues/167036,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: fsdp, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-05T16:53:53Z
issue,167106,DTensor operations incur unnecessary gathering and memory consumption resulting therefrom,https://github.com/pytorch/pytorch/issues/167106,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dtensor, module: performance, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-05T16:53:49Z
issue,167288,"when use torchrun and set restarts for elastic run, cause NCCL INFO socketPollConnect: connect returned Connection refused",https://github.com/pytorch/pytorch/issues/167288,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T16:53:44Z
issue,167305,Pytorch distributed plugin backend _set_default_backend error when use multi-plugin backends,https://github.com/pytorch/pytorch/issues/167305,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T16:53:39Z
issue,167353,`use_mem_pool` is not active during backward,https://github.com/pytorch/pytorch/issues/167353,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: cuda, oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T16:53:35Z
issue,167425,confusion on pipeline parallelism implementation,https://github.com/pytorch/pytorch/issues/167425,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: pipelining, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-05T16:53:30Z
issue,167775,Graceful Ctrl+C handling from torchrun,https://github.com/pytorch/pytorch/issues/167775,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: devx, oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T16:53:26Z
issue,167832,PG splitting does not consider new groups with a different backend,https://github.com/pytorch/pytorch/issues/167832,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T16:53:21Z
issue,167994,CI Not Detecting Failing Tests in test/distributed/elastic/*,https://github.com/pytorch/pytorch/issues/167994,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: ci, oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T16:53:16Z
issue,168143,FSDP2 does not work with custom Tensor wrapper objects,https://github.com/pytorch/pytorch/issues/168143,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: fsdp, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-05T16:53:11Z
issue,169275,GLOO distribute train can not find each ranks,https://github.com/pytorch/pytorch/issues/169275,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T16:53:07Z
issue,169752,"if __name__ == ""__main__"": run_tests() cannot be found in the test files under distributed/elastic/rendezvous",https://github.com/pytorch/pytorch/issues/169752,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: tests, oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T16:53:02Z
issue,169841,ProcessGroup hash doesn't incorporate pg_options,https://github.com/pytorch/pytorch/issues/169841,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T16:52:58Z
issue,170290,"A lot of ""Failed to check the ""should dump"" flag on TCPStore"" output in ProcessGroupNCCL when rank 0 exit first",https://github.com/pytorch/pytorch/issues/170290,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T16:52:49Z
issue,171412,"[Feature Request][DCP] Add ""Broadcast Load"" arg to dcp.load for HSDP storage optimization",https://github.com/pytorch/pytorch/issues/171412,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed checkpointing","decide whether this is bug, use case, docs, or support",2026-06-05T16:52:44Z
issue,171495,[internode] Symmetric mem rendezvous erro,https://github.com/pytorch/pytorch/issues/171495,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: symm_mem, oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T16:52:39Z
issue,171604,NotImplementedError when using AsyncCollectiveTensor with torch.func.jvp,https://github.com/pytorch/pytorch/issues/171604,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: functorch, oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T16:52:34Z
issue,171656,torch.distributed.pipelining fails on models having DynamicCache (esp. Llama),https://github.com/pytorch/pytorch/issues/171656,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: pipelining, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-05T16:52:30Z
issue,173164,allgather throw exception with discontigous input tensor,https://github.com/pytorch/pytorch/issues/173164,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T16:52:25Z
issue,173362,"`dist.reduce_scatter` throws `ValueError: Tensors must be contiguous` when the output tensor is non-contiguous (e.g., from `permute()` or `transpose()`)",https://github.com/pytorch/pytorch/issues/173362,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T16:52:20Z
issue,173608,The program is stuck at function torch.distributed.new_group when parameter use_local_synchronization is True and len(_world.pg_names) is different between NPUs,https://github.com/pytorch/pytorch/issues/173608,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T16:52:16Z
issue,173789,[Devicemesh] Bug: `child_to_root_mapping` overwritten due to hash collision in DeviceMesh submeshes,https://github.com/pytorch/pytorch/issues/173789,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T16:52:11Z
issue,184394,[inductor] Merge SymPy printer path into typed ops/CSE path for Triton codegen,https://github.com/pytorch/pytorch/issues/184394,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-06-05T16:09:18Z
issue,24422,Label tracking meta-issue (edit me to get automatically CC'ed on issues! cc bot),https://github.com/pytorch/pytorch/issues/24422,needs maintainer decision,28,open issue has no stronger automated bucket,"decide whether this is bug, use case, docs, or support",2026-06-05T09:49:01Z
issue,170562,PadMMTest.test_original_aten_preserved_pad_mm fails,https://github.com/pytorch/pytorch/issues/170562,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-06-04T20:08:49Z
issue,186259,DISABLED test_redispatch_nn_functional_linear_cross_entropy_chunked_cuda_float16 (__main__.TestTorchFunctionRedispatchOpsCUDA),https://github.com/pytorch/pytorch/issues/186259,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: __torch_function__, module: flaky-tests","decide whether this is bug, use case, docs, or support",2026-06-04T19:28:03Z
issue,145189,DISABLED test_slice_scatter_reinplace_cuda (__main__.GPUTests),https://github.com/pytorch/pytorch/issues/145189,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-06-04T13:42:52Z
issue,137560,vmap specializing on dynamic shapes with index_put_,https://github.com/pytorch/pytorch/issues/137560,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dynamic shapes, module: functorch, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-06-04T09:45:26Z
issue,186213,CUDA expandable IPC receiver reserves ~9/8 of total GPU memory VA per imported handle and can exhaust device virtual address space,https://github.com/pytorch/pytorch/issues/186213,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: cuda, module: cudacachingallocator, module: multiprocessing","decide whether this is bug, use case, docs, or support",2026-06-04T09:19:34Z
issue,186203,DISABLED test_redispatch_nn_functional_linear_cross_entropy_cuda_float16 (__main__.TestTorchFunctionRedispatchOpsCUDA),https://github.com/pytorch/pytorch/issues/186203,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: __torch_function__, module: flaky-tests","decide whether this is bug, use case, docs, or support",2026-06-04T08:08:13Z
issue,166628,[WinError 1114] A dynamic link library (DLL) initialization routine failed,https://github.com/pytorch/pytorch/issues/166628,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: binaries, module: windows","decide whether this is bug, use case, docs, or support",2026-06-04T07:28:44Z
issue,181851,register_fake to throw error if signature doesn't match op's schema,https://github.com/pytorch/pytorch/issues/181851,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: compile ux, module: custom-operators, module: pt2-dispatcher, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-06-03T17:47:41Z
issue,186099,Triton linux nightly wheels fail to install on Python 3.15 (requires-python: <3.15),https://github.com/pytorch/pytorch/issues/186099,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: binaries, module: ci, module: python version","decide whether this is bug, use case, docs, or support",2026-06-03T16:56:20Z
issue,174373,Increase NUM_THREADS that embedded ARM OpenBLAS is compiled with,https://github.com/pytorch/pytorch/issues/174373,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: arm, module: binaries, module: build, module: openblas","decide whether this is bug, use case, docs, or support",2026-06-03T13:38:07Z
issue,181663,torch.softmax is not a true alias for torch.nn.functional.softmax,https://github.com/pytorch/pytorch/issues/181663,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: nn, module: ux","decide whether this is bug, use case, docs, or support",2026-06-03T09:00:52Z
issue,168652,Test: TestSelectAlgorithm.test_mm_dropout,https://github.com/pytorch/pytorch/issues/168652,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: inductor, module: rocm, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-06-02T23:20:08Z
issue,177259,Migrate device-agnostic skips from `op_db` to `@skipOps`,https://github.com/pytorch/pytorch/issues/177259,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: tests,"decide whether this is bug, use case, docs, or support",2026-06-02T16:51:31Z
issue,185746,CUDNN version issues in 2.13.0.dev20260530+cu132 nightly wheel,https://github.com/pytorch/pytorch/issues/185746,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: binaries, module: cudnn, module: windows","decide whether this is bug, use case, docs, or support",2026-06-02T15:29:29Z
issue,150121,torch.compile on MPS progress tracker,https://github.com/pytorch/pytorch/issues/150121,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: mps,"decide whether this is bug, use case, docs, or support",2026-06-02T11:05:43Z
issue,185198,[xpu] Intel GPU telemetry management APIs may not work on Arc A-Series and data center GPUs,https://github.com/pytorch/pytorch/issues/185198,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: xpu,"decide whether this is bug, use case, docs, or support",2026-06-02T10:13:37Z
issue,160472,PT2 Cache <> CustomOp issue,https://github.com/pytorch/pytorch/issues/160472,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: custom-operators, module: pt2-dispatcher, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-06-01T19:27:24Z
issue,185684,🐛 [Bug] Fake Tensor Updater Doesn't Exist on Windows Platform,https://github.com/pytorch/pytorch/issues/185684,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: faketensor, module: inductor, module: pt2-dispatcher, module: windows","decide whether this is bug, use case, docs, or support",2026-06-01T16:59:36Z
issue,175252,Slow reading tensor meta-data,https://github.com/pytorch/pytorch/issues/175252,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: performance, module: serialization","decide whether this is bug, use case, docs, or support",2026-06-01T14:54:10Z
issue,170487,Support `.requires_grad_(True)` in `torch.compile(fullgraph = True)`,https://github.com/pytorch/pytorch/issues/170487,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dynamo, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-06-01T04:01:19Z
issue,181006,[inductor] fails at caching stage with UnboundLocalError (bufXX referenced before assignment) when compiling torch.func.grad with MultiheadAttention + LayerNorm,https://github.com/pytorch/pytorch/issues/181006,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-06-01T03:59:09Z
issue,181653,torch.compile(torch.func.grad(fn)) hangs indefinitely on models with pooling ops — sympy.simplify exponential recursion in stride_at,https://github.com/pytorch/pytorch/issues/181653,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dynamic shapes, module: functorch, module: inductor, module: pooling","decide whether this is bug, use case, docs, or support",2026-06-01T03:58:28Z
issue,175636,`[feature request][discussion][linalg]` Polar decomposition,https://github.com/pytorch/pytorch/issues/175636,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: cuda, module: linear algebra","decide whether this is bug, use case, docs, or support",2026-06-01T01:46:50Z
issue,51002,Consider adding context manager support for torch.set_printoptions,https://github.com/pytorch/pytorch/issues/51002,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: printing,"decide whether this is bug, use case, docs, or support",2026-05-31T21:30:34Z
issue,185698,CUDA extension build failure on glibc 2.38+ and GCC 14+ due to math header noexcept mismatch - proposed clean fix,https://github.com/pytorch/pytorch/issues/185698,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: build, module: cpp-extensions, module: cuda, module: dependency bug","decide whether this is bug, use case, docs, or support",2026-05-30T04:58:00Z
issue,152028,[CUDA] MultiheadAttention with masks and dropout produces NaNs,https://github.com/pytorch/pytorch/issues/152028,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: correctness (silent), module: cuda, module: sdpa","decide whether this is bug, use case, docs, or support",2026-05-30T03:02:26Z
issue,159151,DISABLED test_fake_crossref_backward_no_amp_nn_functional_bilinear_cuda_float32 (__main__.TestFakeTensorCUDA),https://github.com/pytorch/pytorch/issues/159151,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, module: unknown, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-05-29T13:43:53Z
issue,159150,DISABLED test_fake_crossref_backward_amp_nn_functional_bilinear_cuda_float32 (__main__.TestFakeTensorCUDA),https://github.com/pytorch/pytorch/issues/159150,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: faketensor, module: flaky-tests, module: pt2-dispatcher, module: unknown","decide whether this is bug, use case, docs, or support",2026-05-29T07:59:21Z
issue,185590,[Test] PyTorch Test Case Refactoring and Track,https://github.com/pytorch/pytorch/issues/185590,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: tests,"decide whether this is bug, use case, docs, or support",2026-05-29T03:39:57Z
issue,74415,"Warning: ""Specified kernel cache directory could not be created""",https://github.com/pytorch/pytorch/issues/74415,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: jiterator,"decide whether this is bug, use case, docs, or support",2026-05-28T22:49:23Z
issue,184361,[pipelining] PP fails when trying to pipeline any HF transformers model that uses RoPE,https://github.com/pytorch/pytorch/issues/184361,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: pipelining, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-05-28T15:51:07Z
issue,155240,Dynamo Fails on torch_scatter.scatter_max with Fake Tensor Allocation Error During Graph Tracing,https://github.com/pytorch/pytorch/issues/155240,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: custom-operators, module: pt2-dispatcher, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-05-28T07:28:14Z
issue,179951,DISABLED test_scaled_mm (__main__.DistMatrixOpsTestWithLocalTensor),https://github.com/pytorch/pytorch/issues/179951,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: rocm, oncall: distributed","decide whether this is bug, use case, docs, or support",2026-05-28T06:30:50Z
issue,158186,Flash Attention 2 + Dynamo + FSDP accelerate plugin + torch.compile error,https://github.com/pytorch/pytorch/issues/158186,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: custom-operators, module: pt2-dispatcher, module: sdpa, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-05-27T19:29:07Z
issue,184903,[Inductor] Codegen Gluon,https://github.com/pytorch/pytorch/issues/184903,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-05-27T17:48:43Z
issue,185215,PyTorch with XPU support fails to build on Windows in Debug/RelWithDebInfo configurations,https://github.com/pytorch/pytorch/issues/185215,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: build, module: windows, module: xpu","decide whether this is bug, use case, docs, or support",2026-05-27T06:22:37Z
issue,131009,[cudagraph] an issue related to dynamic shapes for cudagraph backend with sd2.1,https://github.com/pytorch/pytorch/issues/131009,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: cuda graphs, module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-05-27T02:49:26Z
issue,185135,[inductor] Scheduler phases scale poorly on lowered aten.lstm/aten.gru graphs,https://github.com/pytorch/pytorch/issues/185135,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: compile-time, module: inductor, module: performance, module: rnn","decide whether this is bug, use case, docs, or support",2026-05-26T17:50:58Z
issue,181093,[RFC] Add Minimal torch.compile Backend to Validate PrivateUse1 Integration,https://github.com/pytorch/pytorch/issues/181093,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: backend, module: dynamo, module: openreg, module: privateuse1","decide whether this is bug, use case, docs, or support",2026-05-26T14:43:04Z
issue,173369,Exporting ONNX model with captum heatmaps generation,https://github.com/pytorch/pytorch/issues/173369,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: onnx, oncall: export, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-05-26T14:13:37Z
issue,183758,[Testing] Expand Testing for Non-Contiguous Tensor Ops Across All Platforms,https://github.com/pytorch/pytorch/issues/183758,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: memory format, module: testing","decide whether this is bug, use case, docs, or support",2026-05-26T13:25:38Z
issue,174469,[RFC] Overview of Test Case Refactoring in PyTorch,https://github.com/pytorch/pytorch/issues/174469,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: backend, module: privateuse1, module: tests","decide whether this is bug, use case, docs, or support",2026-05-26T07:54:29Z
issue,183060,PyTorch pytree equality comparison is too strict.,https://github.com/pytorch/pytorch/issues/183060,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: pytree,"decide whether this is bug, use case, docs, or support",2026-05-26T07:26:01Z
issue,162728,Significant Numerical Inconsistencies After torch.compile() for Local Attention Model,https://github.com/pytorch/pytorch/issues/162728,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: oncall: pt2,"decide whether this is bug, use case, docs, or support",2026-05-25T11:33:21Z
issue,180206,DISABLED test_opcheck_opinfo_NumpyMulScalarCustomOp_cpu_float32 (__main__.TestCustomOpTestingCPU),https://github.com/pytorch/pytorch/issues/180206,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: custom-operators, module: flaky-tests, module: pt2-dispatcher, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-05-25T08:07:11Z
issue,180437,DISABLED test_find_learning_rate_ensembling_FeatureAlphaDropout_mechanism_functional_call_cpu (__main__.TestExamplesCorrectnessCPU),https://github.com/pytorch/pytorch/issues/180437,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, module: functorch, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-05-25T08:07:02Z
issue,112827,Multi Scale Deformable Attention Support,https://github.com/pytorch/pytorch/issues/112827,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: nn, module: sdpa","decide whether this is bug, use case, docs, or support",2026-05-24T06:13:21Z
issue,14152,Failed to run 'bash ../tools/build_pytorch_libs.sh --use-cuda --use-nnpack --use-mkldnn --use-qnnpack caffe2',https://github.com/pytorch/pytorch/issues/14152,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: build,"decide whether this is bug, use case, docs, or support",2026-05-24T01:04:50Z
issue,80821,Add typing support to ModuleList and ModuleDict,https://github.com/pytorch/pytorch/issues/80821,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: typing,"decide whether this is bug, use case, docs, or support",2026-05-23T22:54:20Z
issue,150168,[ROCm] PyTorch slow on TTS,https://github.com/pytorch/pytorch/issues/150168,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: rocm,"decide whether this is bug, use case, docs, or support",2026-05-23T12:46:25Z
issue,88096,Add nondeterministic alert to `torch.Tensor.scatter()`,https://github.com/pytorch/pytorch/issues/88096,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: determinism,"decide whether this is bug, use case, docs, or support",2026-05-22T21:35:21Z
issue,184769,Optimize Metal Kernels with sincos,https://github.com/pytorch/pytorch/issues/184769,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: complex, module: mps, module: performance","decide whether this is bug, use case, docs, or support",2026-05-22T19:38:33Z
issue,178134,torch.compile (CUDA fp32) vs eager gives large max relative error for Conv2d,https://github.com/pytorch/pytorch/issues/178134,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: convolution, module: cuda, module: inductor, module: pt2 accuracy","decide whether this is bug, use case, docs, or support",2026-05-22T09:58:07Z
issue,55777,torch.load is very slow with gzip.open,https://github.com/pytorch/pytorch/issues/55777,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: serialization,"decide whether this is bug, use case, docs, or support",2026-05-22T05:22:09Z
issue,170636,[xpu] No XPU implementation of multiprocessing tensor reduction.,https://github.com/pytorch/pytorch/issues/170636,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: xpu,"decide whether this is bug, use case, docs, or support",2026-05-22T01:54:20Z
issue,177277,Conv3D does not preserve dim_order for some inputs,https://github.com/pytorch/pytorch/issues/177277,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: convolution, module: faketensor, module: memory format, module: pt2-dispatcher","decide whether this is bug, use case, docs, or support",2026-05-21T17:31:43Z
issue,103222,compilation fails `error: invalid argument '-std=c++17' not allowed with 'C'`,https://github.com/pytorch/pytorch/issues/103222,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: build,"decide whether this is bug, use case, docs, or support",2026-05-21T13:20:38Z
issue,160333,Tensor subclasses don't work with torch.library.triton_op,https://github.com/pytorch/pytorch/issues/160333,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: pt2-dispatcher, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-05-21T09:44:28Z
issue,164057,The device (Intel(R) Arc(TM) A380 Graphics) doesn't support querying the available free memory,https://github.com/pytorch/pytorch/issues/164057,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: xpu,"decide whether this is bug, use case, docs, or support",2026-05-21T08:49:54Z
issue,149953,`RuntimeError: UR error` with XPU,https://github.com/pytorch/pytorch/issues/149953,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: binaries, module: xpu","decide whether this is bug, use case, docs, or support",2026-05-21T08:43:11Z
issue,87961,crash in `torch.package.PackageExporter`,https://github.com/pytorch/pytorch/issues/87961,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: edge cases,"decide whether this is bug, use case, docs, or support",2026-05-20T18:32:06Z
issue,87556,The behavior of cast `NaN` is different on cpu and cuda,https://github.com/pytorch/pytorch/issues/87556,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: edge cases, module: nans and infs","decide whether this is bug, use case, docs, or support",2026-05-20T18:31:59Z
issue,87172,Unstable results in sin/arcsin/arccos calls,https://github.com/pytorch/pytorch/issues/87172,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: edge cases, module: numerical-stability","decide whether this is bug, use case, docs, or support",2026-05-20T18:31:57Z
issue,83494,`torch.pinverse` produces wrong output!,https://github.com/pytorch/pytorch/issues/83494,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: docs, module: linear algebra","decide whether this is bug, use case, docs, or support",2026-05-20T18:31:55Z
issue,165501,Inductor custom pass mechanism that maintains config constraints,https://github.com/pytorch/pytorch/issues/165501,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: inductor, module: vllm, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-05-20T01:21:01Z
issue,140970,`torch._inductor.cpu_vec_isa.pick_vec_isa` takes ~9 seconds to run,https://github.com/pytorch/pytorch/issues/140970,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: cpu, module: inductor, oncall: cpu inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-05-20T01:14:28Z
issue,182490,[Docathon] Convert `notes/hip.rst` from rST to MyST Markdown,https://github.com/pytorch/pytorch/issues/182490,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: docs,"decide whether this is bug, use case, docs, or support",2026-05-19T20:17:09Z
issue,161609,Bucketize lowering crash when using a sliced computed tensor as boundaries,https://github.com/pytorch/pytorch/issues/161609,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-05-19T19:21:49Z
issue,168042,TorchInductor compile breaks when called in a multithreaded context with GIL off,https://github.com/pytorch/pytorch/issues/168042,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-05-19T18:05:58Z
issue,138980,compile_fx inplace modifly the input graph,https://github.com/pytorch/pytorch/issues/138980,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: inductor, module: vllm, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-05-19T18:04:25Z
issue,106006,[torch.compile] assertion sometimes ignored with inductor backend,https://github.com/pytorch/pytorch/issues/106006,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-05-19T18:03:57Z
issue,184397,[ROCm][Windows] Track proper fix for HIP per-config flag injection bypass (follow-up to #183856),https://github.com/pytorch/pytorch/issues/184397,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: build, module: rocm, module: windows","decide whether this is bug, use case, docs, or support",2026-05-19T16:21:10Z
issue,179143,[RFC] AdamTR: Adam variant for deterministic Token-Routed architectures,https://github.com/pytorch/pytorch/issues/179143,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: optimizer,"decide whether this is bug, use case, docs, or support",2026-05-19T15:36:22Z
issue,130015,Improve the Inductor generated kernel for the pattern of `output1 = pointwise(intput); output2 = transpose(output1)`,https://github.com/pytorch/pytorch/issues/130015,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: floatx (formerly float8), module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-05-19T15:25:17Z
issue,143337,An idea to improve Welford implementation in Inductor,https://github.com/pytorch/pytorch/issues/143337,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: oncall: pt2,"decide whether this is bug, use case, docs, or support",2026-05-19T15:24:27Z
issue,165785,Harden CuTeDSL Inductor Path,https://github.com/pytorch/pytorch/issues/165785,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-05-19T15:21:31Z
issue,177578,[Inductor][NVGEMM] Extend nvMatmulHeuristics config key tuple for stages/split_k,https://github.com/pytorch/pytorch/issues/177578,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: cuda, module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-05-19T15:20:15Z
issue,179652,`torch.library.custom.register_autograd` incompatible with `jacrev`,https://github.com/pytorch/pytorch/issues/179652,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: autograd, module: functorch","decide whether this is bug, use case, docs, or support",2026-05-19T08:32:30Z
issue,130039,`min_cut_rematerialization_partition` is not optimal memory usage.,https://github.com/pytorch/pytorch/issues/130039,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-05-19T05:09:16Z
issue,145401,torch.compile has different numerics for var_mean,https://github.com/pytorch/pytorch/issues/145401,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-05-18T21:43:30Z
issue,144362,Some operators miss dtype check when using `torch.compile`,https://github.com/pytorch/pytorch/issues/144362,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: error checking, module: inductor, module: structured kernels, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-05-18T21:43:25Z
issue,120902,RuntimeError: cuFFT error: CUFFT_INTERNAL_ERROR,https://github.com/pytorch/pytorch/issues/120902,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: cuda, module: fft, module: sparse","decide whether this is bug, use case, docs, or support",2026-05-18T21:00:04Z
issue,183391,"torch.fmin with CUDA out=int16 casts inputs before computation, producing incorrect result",https://github.com/pytorch/pytorch/issues/183391,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: cuda, module: edge cases, module: type promotion","decide whether this is bug, use case, docs, or support",2026-05-18T17:33:01Z
issue,113287,Provide a way to AOT torch.compile and serialize a model,https://github.com/pytorch/pytorch/issues/113287,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: compile-time, module: dynamo, oncall: distributed, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-05-18T16:11:18Z
issue,130614,"Significant Memory Increase with Conv2d using padding_mode 'replicate', 'reflect', or 'circular'",https://github.com/pytorch/pytorch/issues/130614,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: convolution, module: memory usage, module: nn","decide whether this is bug, use case, docs, or support",2026-05-18T14:26:35Z
issue,176611,[cudagraph][torch.compile] `cudagraph_or_error` can be silently ignored even when the cudagraph is not captured.,https://github.com/pytorch/pytorch/issues/176611,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: cuda graphs, module: error checking, module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-05-18T06:40:31Z
issue,170762,[BUG][Inconsistency] aten.cudnn_batch_norm_backward produces numerical mismatch between eager and inductor,https://github.com/pytorch/pytorch/issues/170762,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: cudnn, module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-05-18T06:39:45Z
issue,165280,[inductor] avoid triton kernel when creating zero tensor,https://github.com/pytorch/pytorch/issues/165280,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-05-18T06:39:28Z
issue,163932,[inductor] a potential numerical instability issue inside a softmax model.,https://github.com/pytorch/pytorch/issues/163932,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: inductor, module: nans and infs, module: numerical-stability, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-05-18T06:39:23Z
issue,161842,Exhaustive Autotuning causes `cudaErrorIllegalAddress`,https://github.com/pytorch/pytorch/issues/161842,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-05-18T06:39:15Z
issue,159413,Inductor Assertion Error with extremely large input,https://github.com/pytorch/pytorch/issues/159413,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-05-18T06:39:00Z
issue,134606,VMAP over GRU: Batching rule not implemented for aten::gru.input,https://github.com/pytorch/pytorch/issues/134606,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: functorch, module: vmap","decide whether this is bug, use case, docs, or support",2026-05-17T15:15:34Z
issue,181094,Optimization: pin memory management optimization based on pre-allocation and buddy system,https://github.com/pytorch/pytorch/issues/181094,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: cuda, module: performance","decide whether this is bug, use case, docs, or support",2026-05-17T08:22:19Z
issue,87131,torch.clamp does not clamp out of -0 from 0 when ran on the CPU,https://github.com/pytorch/pytorch/issues/87131,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: edge cases, module: numpy","decide whether this is bug, use case, docs, or support",2026-05-17T03:28:49Z
issue,183768,torch.compile returns inf for torch.acosh on large finite float32 inputs while eager returns finite values,https://github.com/pytorch/pytorch/issues/183768,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: correctness (silent), module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-05-17T02:51:31Z
issue,183607,torch.compile changes fp16 overflow behavior for cast-to-float16 multiplication,https://github.com/pytorch/pytorch/issues/183607,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: correctness (silent), module: half, module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-05-17T02:51:09Z
issue,180811,"[AOTInductor] max_autotune non-deterministically selects cuBLAS over fused Triton kernels for small GEMM, causing significant performance regression",https://github.com/pytorch/pytorch/issues/180811,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: aotinductor, module: cuda, module: determinism, module: inductor","decide whether this is bug, use case, docs, or support",2026-05-17T02:48:31Z
issue,155679,Torch compile CUDA graphs leads to a large number of CUDA streams,https://github.com/pytorch/pytorch/issues/155679,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: cuda graphs, module: dependency bug, module: inductor, oncall: distributed","decide whether this is bug, use case, docs, or support",2026-05-16T01:05:57Z
issue,182272,[Windows] RTX 5070 Ti (Blackwell sm_120) - cuInit error 100 on hybrid graphics laptops,https://github.com/pytorch/pytorch/issues/182272,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: binaries, module: cuda, module: docs, module: windows","decide whether this is bug, use case, docs, or support",2026-05-15T22:40:31Z
issue,177719,[Discussion] Disable size_asserts by default in inductor for vllm x compile inference,https://github.com/pytorch/pytorch/issues/177719,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: cuda graphs, module: inductor, module: performance, module: vllm","decide whether this is bug, use case, docs, or support",2026-05-15T22:16:23Z
issue,37410,Reset a `torch.optim.Optimizer`,https://github.com/pytorch/pytorch/issues/37410,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: optimizer,"decide whether this is bug, use case, docs, or support",2026-05-15T07:08:39Z
issue,152270,Question about that support of torch.compile for a custom CUDA operator?,https://github.com/pytorch/pytorch/issues/152270,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: custom-operators, module: docs, module: pt2-dispatcher, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-05-15T03:38:02Z
issue,91469,torch.compile with aotautograd does not support double backwards,https://github.com/pytorch/pytorch/issues/91469,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: aotdispatch, module: autograd, module: pt2-dispatcher, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-05-15T03:37:00Z
issue,125977,[compile time] AOT Autograd is taking long time in tracing,https://github.com/pytorch/pytorch/issues/125977,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: aotdispatch, module: pt2-dispatcher, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-05-15T03:07:08Z
issue,150877,DISABLED test_multiple_module (__main__.TestInvokeSubgraphExportStrict),https://github.com/pytorch/pytorch/issues/150877,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, module: higher order operators, module: pt2-dispatcher, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-05-15T03:06:40Z
issue,127445,PyTorch version agnostic C++ extensions,https://github.com/pytorch/pytorch/issues/127445,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: cpp-extensions, module: custom-operators, module: pt2-dispatcher, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-05-15T03:06:32Z
issue,150062,Support HOPs in fx_graph_runnable,https://github.com/pytorch/pytorch/issues/150062,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: higher order operators, module: pt2-dispatcher, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-05-15T02:04:05Z
issue,157456,Deprecation of CUTLASS Python interface,https://github.com/pytorch/pytorch/issues/157456,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: cuda,"decide whether this is bug, use case, docs, or support",2026-05-14T23:37:56Z
issue,139257,Error compiling the torch.library.custom_op with input mutations with set_,https://github.com/pytorch/pytorch/issues/139257,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: custom-operators, module: pt2-dispatcher, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-05-14T22:34:17Z
issue,136446,Allow Inductor to Compose with FakeTensorMode to Estimate Memory Usage,https://github.com/pytorch/pytorch/issues/136446,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: faketensor, module: inductor, module: pt2-dispatcher, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-05-14T22:21:42Z
issue,133721,[HOP] Add BC tests for hop in export,https://github.com/pytorch/pytorch/issues/133721,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: higher order operators, module: pt2-dispatcher, oncall: export, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-05-14T22:21:42Z
issue,183795,Clean up XNNPACK_NO_CODE_CACHE,https://github.com/pytorch/pytorch/issues/183795,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: xnnpack,"decide whether this is bug, use case, docs, or support",2026-05-14T21:54:54Z
issue,131974,Missing grad_fn information while torch.compile with customized gradient function,https://github.com/pytorch/pytorch/issues/131974,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: aotdispatch, module: pt2-dispatcher, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-05-14T19:05:37Z
issue,129430,OperatorBase.py_impl should support with_keyset=True extra argument.,https://github.com/pytorch/pytorch/issues/129430,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: pt2-dispatcher, oncall: export, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-05-14T19:05:16Z
issue,113089,Add docs for __tensor_flatten__ / __tensor_unflatten__,https://github.com/pytorch/pytorch/issues/113089,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: __torch_dispatch__, module: docs, module: pt2-dispatcher, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-05-14T17:41:45Z
issue,142388,User-defined way to hook into Inductor autotuning (+ custom op),https://github.com/pytorch/pytorch/issues/142388,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: custom-operators, module: inductor, module: pt2-dispatcher, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-05-14T17:41:42Z
issue,145522,use_const_ref_for_mutable_tensors doesn't work with out= overloads,https://github.com/pytorch/pytorch/issues/145522,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: pt2-dispatcher,"decide whether this is bug, use case, docs, or support",2026-05-14T15:35:50Z
issue,121837,Revive libtorch-shared-without-deps flavor,https://github.com/pytorch/pytorch/issues/121837,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: abi, module: binaries, module: cpp","decide whether this is bug, use case, docs, or support",2026-05-14T10:01:18Z
issue,183686,[triton-cpu] All HuggingFace models of the dynamo benchmark suites failed with the latest triton-cpu backend,https://github.com/pytorch/pytorch/issues/183686,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: cpu, module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-05-14T08:54:26Z
issue,127621,custom_op API: better type anntation for Tuple,https://github.com/pytorch/pytorch/issues/127621,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: custom-operators, module: pt2-dispatcher, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-05-14T04:03:04Z
issue,129395,[custom_op] Figure out what to do with random custom ops,https://github.com/pytorch/pytorch/issues/129395,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: custom-operators, module: pt2-dispatcher, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-05-14T04:02:55Z
issue,147210,Consider relaxing the in-place mutation restriction for torch.cond and torch.while_loop,https://github.com/pytorch/pytorch/issues/147210,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: higher order operators, module: pt2-dispatcher, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-05-14T03:44:31Z
issue,146941,Make HOPs more debuggable,https://github.com/pytorch/pytorch/issues/146941,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: higher order operators, module: pt2-dispatcher, module: tests, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-05-14T03:44:31Z
issue,146860,[BE] Put all the HOP tests in one location,https://github.com/pytorch/pytorch/issues/146860,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: higher order operators, module: pt2-dispatcher, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-05-14T02:27:56Z
issue,146936,flexible custom operators: custom operators that accept arbitrary input/output type,https://github.com/pytorch/pytorch/issues/146936,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: custom-operators, module: pt2-dispatcher, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-05-14T02:27:55Z
issue,146861,[BE] the majority of HOP OpInfo tests should be in one place,https://github.com/pytorch/pytorch/issues/146861,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: higher order operators, module: pt2-dispatcher, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-05-14T02:27:54Z
issue,146111,[scan] Support lowering and lifted arguments for inductor,https://github.com/pytorch/pytorch/issues/146111,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: higher order operators, module: pt2-dispatcher, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-05-13T23:37:38Z
issue,182817,[MPS] Switch test_output_match / test_output_grad_match to magnitude-floored relative tolerance,https://github.com/pytorch/pytorch/issues/182817,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: mps, module: numerical-stability, module: tests","decide whether this is bug, use case, docs, or support",2026-05-13T20:54:44Z
issue,111824,GroupNorm & InstanceNorm does not handle channels_last correctly,https://github.com/pytorch/pytorch/issues/111824,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: cuda, module: memory format, module: nn","decide whether this is bug, use case, docs, or support",2026-05-13T18:29:11Z
issue,144805,"""GenericHOPVariable"" / abstract out Dynamo support for HOPs",https://github.com/pytorch/pytorch/issues/144805,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dynamo, module: higher order operators, module: pt2-dispatcher, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-05-13T04:03:13Z
issue,183036,Adadelta uses SGD in its examples,https://github.com/pytorch/pytorch/issues/183036,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: docs, module: optimizer","decide whether this is bug, use case, docs, or support",2026-05-11T18:45:21Z
issue,182560,import torch fails with concurrent nested imports despite GIL ('torch._C' is not a package),https://github.com/pytorch/pytorch/issues/182560,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: edge cases, module: python frontend","decide whether this is bug, use case, docs, or support",2026-05-11T17:15:02Z
issue,182960,"`torch.cuda.ExternalStream(0, device=...)` silently returns a fresh pooled stream instead of wrapping handle `0x0`",https://github.com/pytorch/pytorch/issues/182960,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: correctness (silent), module: cuda, module: edge cases, module: python frontend","decide whether this is bug, use case, docs, or support",2026-05-11T17:07:32Z
issue,142135,Device assert throws a runtime error in cuda backend and results in a crash in xpu backend,https://github.com/pytorch/pytorch/issues/142135,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: xpu,"decide whether this is bug, use case, docs, or support",2026-05-11T07:56:53Z
issue,143239,"xpu: torch.nn.DataParallel fails on multi-XPU environment with ""module 'torch._C' has no attribute '_scatter'""",https://github.com/pytorch/pytorch/issues/143239,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: xpu, oncall: distributed","decide whether this is bug, use case, docs, or support",2026-05-10T08:19:10Z
issue,179707,Improve static type hints for dataloader with known `collate_fn`,https://github.com/pytorch/pytorch/issues/179707,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dataloader, module: typing","decide whether this is bug, use case, docs, or support",2026-05-09T17:48:15Z
issue,7887,[Caffe2] convert ONNX to caffe2,https://github.com/pytorch/pytorch/issues/7887,needs maintainer decision,28,open issue has no stronger automated bucket,"decide whether this is bug, use case, docs, or support",2026-05-09T14:32:45Z
issue,9867,How can I sure that I run caffe2 with GPU,https://github.com/pytorch/pytorch/issues/9867,needs maintainer decision,28,open issue has no stronger automated bucket,"decide whether this is bug, use case, docs, or support",2026-05-09T14:32:41Z
issue,11869,[caffe2] Unaligned AVX instruction operands in LayerNorm implementation in OSS build,https://github.com/pytorch/pytorch/issues/11869,needs maintainer decision,28,open issue has no stronger automated bucket,"decide whether this is bug, use case, docs, or support",2026-05-09T14:32:25Z
issue,10818,[Caffe2] Can't load pretrained model,https://github.com/pytorch/pytorch/issues/10818,needs maintainer decision,28,open issue has no stronger automated bucket,"decide whether this is bug, use case, docs, or support",2026-05-09T14:32:21Z
issue,10464,caffe2 CI does not test header install,https://github.com/pytorch/pytorch/issues/10464,needs maintainer decision,28,open issue has no stronger automated bucket,"decide whether this is bug, use case, docs, or support",2026-05-09T14:32:09Z
issue,10440,"[caffe2] enforce fail at context_gpu.cu:285, cannot get GPU memory usage statistics",https://github.com/pytorch/pytorch/issues/10440,needs maintainer decision,28,open issue has no stronger automated bucket,"decide whether this is bug, use case, docs, or support",2026-05-09T14:32:05Z
issue,9707,"[caffe2]mpirun multi-node multi GPU in Distributed mode ,run resnet50_trainer.py get RuntimeError",https://github.com/pytorch/pytorch/issues/9707,needs maintainer decision,28,open issue has no stronger automated bucket,"decide whether this is bug, use case, docs, or support",2026-05-09T14:31:57Z
issue,9349,[Caffe2] Can't find Caffe2<->ONNX conversion tools,https://github.com/pytorch/pytorch/issues/9349,needs maintainer decision,28,open issue has no stronger automated bucket,"decide whether this is bug, use case, docs, or support",2026-05-09T14:31:52Z
issue,8595,[Caffe2] How to build caffe2/mobile/ulp2/ulp_test?,https://github.com/pytorch/pytorch/issues/8595,needs maintainer decision,28,open issue has no stronger automated bucket,"decide whether this is bug, use case, docs, or support",2026-05-09T14:31:48Z
issue,8533,[Caffe2] Runtime error while using a pre-trained style_transfer model,https://github.com/pytorch/pytorch/issues/8533,needs maintainer decision,28,open issue has no stronger automated bucket,"decide whether this is bug, use case, docs, or support",2026-05-09T14:31:44Z
issue,8442,[caffe2] Why the GPU memory consumption has no change after enable optimize_gradient_memory?,https://github.com/pytorch/pytorch/issues/8442,needs maintainer decision,28,open issue has no stronger automated bucket,"decide whether this is bug, use case, docs, or support",2026-05-09T14:31:40Z
issue,7835,[caffe2] how to using the Mul operator to mul the input vetor,https://github.com/pytorch/pytorch/issues/7835,needs maintainer decision,28,open issue has no stronger automated bucket,"decide whether this is bug, use case, docs, or support",2026-05-09T14:31:36Z
issue,7614,[Caffe2] modify op in the net from init and predict files,https://github.com/pytorch/pytorch/issues/7614,needs maintainer decision,28,open issue has no stronger automated bucket,"decide whether this is bug, use case, docs, or support",2026-05-09T14:31:32Z
issue,7569,[Caffe2]How to set lr_mult and decay_mult in Conv layer?,https://github.com/pytorch/pytorch/issues/7569,needs maintainer decision,28,open issue has no stronger automated bucket,"decide whether this is bug, use case, docs, or support",2026-05-09T14:31:28Z
issue,7491,[caffe2] How to know the input shape for certain pre-trained network,https://github.com/pytorch/pytorch/issues/7491,needs maintainer decision,28,open issue has no stronger automated bucket,"decide whether this is bug, use case, docs, or support",2026-05-09T14:31:24Z
issue,7490,"[Caffe2]: caffe2.python.caffe_translator.py script doesn't convert ""Split"" and ""Slice"" Layer",https://github.com/pytorch/pytorch/issues/7490,needs maintainer decision,28,open issue has no stronger automated bucket,"decide whether this is bug, use case, docs, or support",2026-05-09T14:31:20Z
issue,7374,[Caffe2] Can't use resnet50_trainer.py through redis.,https://github.com/pytorch/pytorch/issues/7374,needs maintainer decision,28,open issue has no stronger automated bucket,"decide whether this is bug, use case, docs, or support",2026-05-09T14:31:16Z
issue,7362,[caffe2] Can I load the gpu data to cpu in .cu file?,https://github.com/pytorch/pytorch/issues/7362,needs maintainer decision,28,open issue has no stronger automated bucket,"decide whether this is bug, use case, docs, or support",2026-05-09T14:31:12Z
issue,7060,[feature request] Provide Caffe2 CUDA dockerfile with USE_REDIS=ON,https://github.com/pytorch/pytorch/issues/7060,needs maintainer decision,28,open issue has no stronger automated bucket,"decide whether this is bug, use case, docs, or support",2026-05-09T14:31:07Z
issue,6979,[Caffe2] Is there any way to load custom operators in c++?,https://github.com/pytorch/pytorch/issues/6979,needs maintainer decision,28,open issue has no stronger automated bucket,"decide whether this is bug, use case, docs, or support",2026-05-09T14:31:03Z
issue,6785,[caffe2] Training and inference,https://github.com/pytorch/pytorch/issues/6785,needs maintainer decision,28,open issue has no stronger automated bucket,"decide whether this is bug, use case, docs, or support",2026-05-09T14:30:59Z
issue,6549,caffe2/cuda_rtc can throw during the destructor,https://github.com/pytorch/pytorch/issues/6549,needs maintainer decision,28,open issue has no stronger automated bucket,"decide whether this is bug, use case, docs, or support",2026-05-09T14:30:55Z
issue,179784,"torch.xlogy(0, 0) returns NaN, while in many numerical computing libraries (e.g., SciPy) and common machine learning practice, it is defined as 0.",https://github.com/pytorch/pytorch/issues/179784,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: scipy compatibility, module: special","decide whether this is bug, use case, docs, or support",2026-05-08T19:49:25Z
issue,181946,[MPS] Remove Unnecessary Tensor Memory Gathers and Contiguous Calls for Stride-Aware MPS Backend Inputs,https://github.com/pytorch/pytorch/issues/181946,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: mps, module: performance","decide whether this is bug, use case, docs, or support",2026-05-08T16:26:55Z
issue,178492,native_batch_norm_backward extremely slow on MPS for 3D tensors (slower than CPU),https://github.com/pytorch/pytorch/issues/178492,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: mps, module: norms and normalization, module: performance","decide whether this is bug, use case, docs, or support",2026-05-08T14:26:39Z
issue,180343,Accelerator-independent API to determine if a specific backend is available?,https://github.com/pytorch/pytorch/issues/180343,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: accelerator, module: backend, module: privateuse1","decide whether this is bug, use case, docs, or support",2026-05-08T06:40:05Z
issue,146045,Redundant collectives are deduplicated but stick around as dead code,https://github.com/pytorch/pytorch/issues/146045,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dtensor, oncall: distributed, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-05-07T18:37:49Z
issue,182842,[v2.11.0] Cannot jvp through vmap,https://github.com/pytorch/pytorch/issues/182842,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: forward ad, module: functorch, module: vmap","decide whether this is bug, use case, docs, or support",2026-05-07T17:05:37Z
issue,180824,"`torch.compile fails with RecursionError on nn.Module with custom __getattr__, while eager execution succeeds`",https://github.com/pytorch/pytorch/issues/180824,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: compile ux, module: dynamo, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-05-07T11:08:38Z
issue,139313,BFloat16 support for `torch.fft.rfft`,https://github.com/pytorch/pytorch/issues/139313,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: complex,"decide whether this is bug, use case, docs, or support",2026-05-07T07:22:10Z
issue,167616,UNSTABLE trunk-rocm-sandbox / linux-jammy-rocm-py3.10 / test,https://github.com/pytorch/pytorch/issues/167616,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: ci, module: rocm","decide whether this is bug, use case, docs, or support",2026-05-07T06:48:26Z
issue,169857,"[ROCm] MIOpen 3D convolution kernel fails with ""invalid configuration argument"" for Qwen2.5-VL vision encoder - silent fallback causes severe performance regression",https://github.com/pytorch/pytorch/issues/169857,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: rocm,"decide whether this is bug, use case, docs, or support",2026-05-06T23:02:58Z
issue,167411,Torch nightly rocm6.4 pytorch-triton-rocm requirement mismatch,https://github.com/pytorch/pytorch/issues/167411,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: binaries, module: rocm","decide whether this is bug, use case, docs, or support",2026-05-06T21:47:28Z
issue,182684,[`SDPA`] Checking for FA backend eligibility via SDPAParams fails under fullgraph compile,https://github.com/pytorch/pytorch/issues/182684,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dynamo, module: sdpa, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-05-06T18:57:14Z
issue,182450,Spin codegen run doesn't use self-contained environement,https://github.com/pytorch/pytorch/issues/182450,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: codegen, module: lint, module: spin","decide whether this is bug, use case, docs, or support",2026-05-06T15:42:06Z
issue,71389,Create a nested directory while saving objects using `torch.save`,https://github.com/pytorch/pytorch/issues/71389,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: serialization,"decide whether this is bug, use case, docs, or support",2026-05-06T15:39:21Z
issue,160514,BenchmarkingTest.test_benchmark_on_non_zero_device mock hit failed,https://github.com/pytorch/pytorch/issues/160514,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: tests,"decide whether this is bug, use case, docs, or support",2026-05-06T15:33:27Z
issue,180284,Unused jvp function on autograd function causes inscrutable shape error,https://github.com/pytorch/pytorch/issues/180284,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: compile ux, module: compiled autograd, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-05-06T15:21:40Z
issue,181718,bf16 matmul throughput drops on RTX 4090 for shapes where N%16==8,https://github.com/pytorch/pytorch/issues/181718,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: bfloat16, module: cublas, module: cuda, module: linear algebra","decide whether this is bug, use case, docs, or support",2026-05-06T12:43:17Z
issue,149639,Pylint error: ` torch.linalg.vector_norm is not callable`,https://github.com/pytorch/pytorch/issues/149639,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: lint, module: typing","decide whether this is bug, use case, docs, or support",2026-05-06T10:53:55Z
issue,168871,TestModule: test/test_transformers.py::TestSDPACudaOnly,https://github.com/pytorch/pytorch/issues/168871,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: rocm, module: sdpa","decide whether this is bug, use case, docs, or support",2026-05-04T22:37:10Z
issue,182381,[HOP] Backward pass for scan HOP can result in invalid tensor metadata,https://github.com/pytorch/pytorch/issues/182381,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: autograd, module: faketensor, module: higher order operators, module: pt2-dispatcher","decide whether this is bug, use case, docs, or support",2026-05-04T22:25:38Z
issue,144459,"In AOTI Runtime, acquiring a write lock on model_exec_mutex_ may cause starvation.",https://github.com/pytorch/pytorch/issues/144459,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: aotinductor, oncall: export","decide whether this is bug, use case, docs, or support",2026-05-04T21:42:34Z
issue,182336,Adding new lint passes PR CI but not trunk CI,https://github.com/pytorch/pytorch/issues/182336,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: ci, module: devx, module: lint, module: spin","decide whether this is bug, use case, docs, or support",2026-05-04T20:03:28Z
issue,181723,torch.cuda.make_graphed_callables can silently corrupt parameter gradient accumulation by returning static grad buffers,https://github.com/pytorch/pytorch/issues/181723,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: autograd, module: correctness (silent), module: cuda, module: cuda graphs","decide whether this is bug, use case, docs, or support",2026-05-04T17:17:56Z
issue,181801,Inconsistent Signed-Zero Handling Between CPU and CUDA Backends,https://github.com/pytorch/pytorch/issues/181801,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: correctness (silent), module: cuda, module: edge cases, module: nn","decide whether this is bug, use case, docs, or support",2026-05-04T17:15:15Z
issue,182286,[Bug] CUDA caching allocator unable to satisfy 3-18 GiB contiguous allocation despite 50+ GiB free on RTX PRO 6000 Blackwell + WSL2 (16 GiB hidden CUDA overhead),https://github.com/pytorch/pytorch/issues/182286,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: cuda, module: cudacachingallocator, module: memory usage, module: wsl","decide whether this is bug, use case, docs, or support",2026-05-04T13:50:31Z
issue,113228,dataloader use lmdb stuck,https://github.com/pytorch/pytorch/issues/113228,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: deadlock,"decide whether this is bug, use case, docs, or support",2026-05-04T08:16:53Z
issue,180968,Initializing weights after parametrization silently fails,https://github.com/pytorch/pytorch/issues/180968,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: initialization, module: nn, module: nn.utils.parametrize","decide whether this is bug, use case, docs, or support",2026-05-03T20:40:30Z
issue,172807,"[CUDA] Auto-detected arch flags omit accelerated suffix (sm_120a → sm_120), breaking Blackwell NVFP4/block-scaled MMA",https://github.com/pytorch/pytorch/issues/172807,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: build, module: cuda","decide whether this is bug, use case, docs, or support",2026-05-03T13:13:57Z
issue,74839,Feature request: Root-finding functionality (like scipy.optimize.root_scalar),https://github.com/pytorch/pytorch/issues/74839,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: scientific computing,"decide whether this is bug, use case, docs, or support",2026-05-03T12:15:55Z
issue,182156,Add torch.utils.nan_guard for forward-pass NaN/Inf debugging,https://github.com/pytorch/pytorch/issues/182156,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: nans and infs, module: nn","decide whether this is bug, use case, docs, or support",2026-05-01T18:23:22Z
issue,112903,Support for Bazel workspace function or Bazel module,https://github.com/pytorch/pytorch/issues/112903,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: bazel, module: build","decide whether this is bug, use case, docs, or support",2026-05-01T16:13:37Z
issue,181737,image diffusion model output changes for the worse after 530ff92655,https://github.com/pytorch/pytorch/issues/181737,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: cublas, module: cuda, module: numerical-stability","decide whether this is bug, use case, docs, or support",2026-05-01T14:49:33Z
issue,182105,Adding a UserWarning when pickling / deepcopying tensor views,https://github.com/pytorch/pytorch/issues/182105,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: molly-guard, module: pickle, module: serialization, module: ux","decide whether this is bug, use case, docs, or support",2026-05-01T10:35:50Z
issue,182008,Windows Application Control (WDAC) blocks loading unsigned torch DLL,https://github.com/pytorch/pytorch/issues/182008,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: binaries, module: windows","decide whether this is bug, use case, docs, or support",2026-04-30T17:48:24Z
issue,145555,need to document `FlopCounterMode`,https://github.com/pytorch/pytorch/issues/145555,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: docs,"decide whether this is bug, use case, docs, or support",2026-04-30T17:28:06Z
issue,153546,Make implicit packages (PEP420) explicit PyTorch,https://github.com/pytorch/pytorch/issues/153546,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: build, module: lint","decide whether this is bug, use case, docs, or support",2026-04-30T17:28:05Z
issue,75225,One of the backends for `lu_factor` fails on windows.,https://github.com/pytorch/pytorch/issues/75225,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: linear algebra, module: windows","decide whether this is bug, use case, docs, or support",2026-04-30T07:30:24Z
issue,180839,[Dynamo] Attempted to call function marked as skipped: torch.sparse_csr_tensor cannot be traced,https://github.com/pytorch/pytorch/issues/180839,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dynamo, module: sparse, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-04-30T00:48:31Z
issue,141265,rope perf improvement,https://github.com/pytorch/pytorch/issues/141265,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-04-29T23:22:13Z
issue,181494,Add support for sparse block-diagonal matrices,https://github.com/pytorch/pytorch/issues/181494,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: sparse,"decide whether this is bug, use case, docs, or support",2026-04-29T18:54:21Z
issue,103841,[question] [docs] Short/mid/long-term status of TorchScript / JIT / torch.jit.trace / FX / symbolic tracing and its replacement by Dynamo,https://github.com/pytorch/pytorch/issues/103841,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: docs, oncall: jit","decide whether this is bug, use case, docs, or support",2026-04-28T14:37:09Z
issue,152355,[source.wheel] Pin setuptools runtime dependency,https://github.com/pytorch/pytorch/issues/152355,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: binaries, module: build, module: cpp-extensions","decide whether this is bug, use case, docs, or support",2026-04-28T06:58:10Z
issue,181649,UNSTABLE rocm-nightly / linux-noble-rocm-nightly-py3.12-gfx942 / test (distributed),https://github.com/pytorch/pytorch/issues/181649,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: ci, module: rocm","decide whether this is bug, use case, docs, or support",2026-04-27T20:45:25Z
issue,181648,UNSTABLE rocm-nightly / linux-noble-rocm-nightly-py3.12-gfx942 / test (inductor),https://github.com/pytorch/pytorch/issues/181648,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: ci, module: inductor, module: rocm, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-04-27T20:45:13Z
issue,181647,UNSTABLE rocm-nightly / linux-noble-rocm-nightly-py3.12-gfx942 / test (default),https://github.com/pytorch/pytorch/issues/181647,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: ci, module: rocm","decide whether this is bug, use case, docs, or support",2026-04-27T20:45:10Z
issue,152710,"[FSDP2] NO_SHARD as fully_shard(mesh=(Replicate, Shard)) with shard of world size 1",https://github.com/pytorch/pytorch/issues/152710,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: oncall: distributed,"decide whether this is bug, use case, docs, or support",2026-04-27T13:40:13Z
issue,179008,[privateuse1] Update the privateuse1 tutorial to mention registering HooksInterface,https://github.com/pytorch/pytorch/issues/179008,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: privateuse1,"decide whether this is bug, use case, docs, or support",2026-04-27T10:54:18Z
issue,179010,[privateuse1] Update privateuse1 tutorial to show the Python backend approach,https://github.com/pytorch/pytorch/issues/179010,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: privateuse1,"decide whether this is bug, use case, docs, or support",2026-04-27T10:52:58Z
issue,154318,[BUG] DataLoader low GPU utilization and extremely slow compared to manual batching,https://github.com/pytorch/pytorch/issues/154318,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: data, module: dataloader, module: performance","decide whether this is bug, use case, docs, or support",2026-04-27T09:46:06Z
issue,158714,Feature request: `DataLoader` using multithreading instead of multiprocessing,https://github.com/pytorch/pytorch/issues/158714,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: data, module: dataloader, module: multithreading","decide whether this is bug, use case, docs, or support",2026-04-27T06:18:53Z
issue,154363,`scaled_dot_product_attention` broadcasting (GQA) is a memory footgun,https://github.com/pytorch/pytorch/issues/154363,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: memory usage, module: sdpa","decide whether this is bug, use case, docs, or support",2026-04-26T22:24:54Z
issue,181525,explore using gmpy2 to accelerate expression eval in sympy,https://github.com/pytorch/pytorch/issues/181525,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dynamic shapes, module: performance, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-04-26T20:20:56Z
issue,179294,[MPS] `scaled_dot_product_attention` (SDPA) improvements,https://github.com/pytorch/pytorch/issues/179294,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: mps, module: performance, module: sdpa","decide whether this is bug, use case, docs, or support",2026-04-26T01:54:18Z
issue,179430,Add Frechet/Inverse Weibull distribution,https://github.com/pytorch/pytorch/issues/179430,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: distributions,"decide whether this is bug, use case, docs, or support",2026-04-25T18:55:22Z
issue,96602,"[MPS] softmax returns NaN attention probabilities for large tensors, in float16 and float32.",https://github.com/pytorch/pytorch/issues/96602,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: mps,"decide whether this is bug, use case, docs, or support",2026-04-25T15:36:08Z
issue,181488,[TensorIterator] reorder_dimensions is sensitive to operand order for commutative ops,https://github.com/pytorch/pytorch/issues/181488,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: performance, module: tensoriterator","decide whether this is bug, use case, docs, or support",2026-04-25T11:51:14Z
issue,181379,"[cuDNN][SDPA] head_dim limit stuck at 128 on sm_120 (RTX 5090), blocking head_dim=256 models",https://github.com/pytorch/pytorch/issues/181379,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: cuda, module: cudnn, module: sdpa","decide whether this is bug, use case, docs, or support",2026-04-24T17:33:44Z
issue,28648,Can't successfully install pytorch with python3.6 on my pi 4,https://github.com/pytorch/pytorch/issues/28648,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: build,"decide whether this is bug, use case, docs, or support",2026-04-23T17:41:42Z
issue,180620,"@custom_op doesn't support ""maybe_out semantics""",https://github.com/pytorch/pytorch/issues/180620,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: custom-operators, module: pt2-dispatcher, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-04-23T14:47:32Z
issue,180960,[cudagraphs] torch.cond does not work with cudagraphs when nccl collectives are used,https://github.com/pytorch/pytorch/issues/180960,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: cuda graphs, module: higher order operators, module: nccl, module: pt2-dispatcher","decide whether this is bug, use case, docs, or support",2026-04-23T13:46:02Z
issue,178600,RuntimeError of use_libuv and ReduceOp.AVG with DDP + Gloo on Windows multi-gpu train,https://github.com/pytorch/pytorch/issues/178600,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: c10d, module: ddp, oncall: distributed","decide whether this is bug, use case, docs, or support",2026-04-23T05:33:41Z
issue,171022,In-place real-domain FFT operators for memory-efficient training (bf16 support),https://github.com/pytorch/pytorch/issues/171022,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: fft,"decide whether this is bug, use case, docs, or support",2026-04-23T02:45:24Z
issue,181108,Use of dynamic shapes impedes Inductor fusion for XSA projection,https://github.com/pytorch/pytorch/issues/181108,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dynamic shapes, module: inductor, module: performance, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-04-22T17:24:08Z
issue,113149,Compilation error on loongarch64,https://github.com/pytorch/pytorch/issues/113149,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: build, module: third_party","decide whether this is bug, use case, docs, or support",2026-04-22T15:07:22Z
issue,181076,[Inductor] Improve compile time of `build_fusion_regions`,https://github.com/pytorch/pytorch/issues/181076,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: inductor, module: performance, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-04-22T07:49:54Z
issue,180975,[Tracking] RISC-V PyTorch enablement,https://github.com/pytorch/pytorch/issues/180975,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: build, module: ci, module: cpu, module: risc-v","decide whether this is bug, use case, docs, or support",2026-04-21T13:12:07Z
issue,177654,`import functorch.dim` silently breaks `torch.Tensor.expand(size=...)`,https://github.com/pytorch/pytorch/issues/177654,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: correctness (silent), module: functorch","decide whether this is bug, use case, docs, or support",2026-04-21T12:34:52Z
issue,52241,CTCLoss gradient is incorrect,https://github.com/pytorch/pytorch/issues/52241,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: autograd, module: docs, module: loss","decide whether this is bug, use case, docs, or support",2026-04-20T22:17:26Z
issue,180915,accelerate sympy computations using gmpy2,https://github.com/pytorch/pytorch/issues/180915,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: performance, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-04-20T21:11:06Z
issue,179710,AArch64 unit test failure TestFX.test_shape_prop_layout_3d,https://github.com/pytorch/pytorch/issues/179710,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: arm, module: tests","decide whether this is bug, use case, docs, or support",2026-04-20T10:09:37Z
issue,174970,[RFC] Hooks to trigger recompilation,https://github.com/pytorch/pytorch/issues/174970,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: aotdispatch, module: dynamo, module: fx, module: inductor","decide whether this is bug, use case, docs, or support",2026-04-20T08:23:22Z
issue,158917,[Feat] Tracking for OpenReg Improvements,https://github.com/pytorch/pytorch/issues/158917,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: openreg, module: privateuse1","decide whether this is bug, use case, docs, or support",2026-04-20T06:41:25Z
issue,180650,`torch.linalg._powsum` is slow on CPU for ord != 2,https://github.com/pytorch/pytorch/issues/180650,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: cpu, module: linear algebra, module: performance","decide whether this is bug, use case, docs, or support",2026-04-20T03:00:29Z
issue,119613,init_device_mesh doesn't have a timeout kwarg,https://github.com/pytorch/pytorch/issues/119613,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: devicemesh, oncall: distributed","decide whether this is bug, use case, docs, or support",2026-04-19T07:56:23Z
issue,116697,Tracing GPU usage of `torch.vmap` vectorized operation using `torch.profiler` cause CUDA OOM,https://github.com/pytorch/pytorch/issues/116697,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: vmap, oncall: profiler","decide whether this is bug, use case, docs, or support",2026-04-19T01:34:35Z
issue,131972,backward of adaptive max pool (adaptive_max_pool2d_backward_cuda) doesn't have a deterministic implementation,https://github.com/pytorch/pytorch/issues/131972,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: autograd, module: determinism","decide whether this is bug, use case, docs, or support",2026-04-18T22:09:51Z
issue,158551,CUDAGraphs RuntimeError: Accessing Overwritten Tensor Output Despite Clone and cudagraph_mark_step_begin in PyTorch 2.3.0,https://github.com/pytorch/pytorch/issues/158551,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: cuda graphs, module: ddp, oncall: distributed, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-04-18T15:22:15Z
issue,151204,[Dynamo] Dynamo fails to trace reduce_scatter_v,https://github.com/pytorch/pytorch/issues/151204,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: c10d, module: dynamo, oncall: distributed, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-04-18T15:09:31Z
issue,172231,[cuBLAS] setting `cublas/cublaslt` backend seems to be a no-op for `torch.addmm` when dispatched to cuBLASLt,https://github.com/pytorch/pytorch/issues/172231,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: cublas,"decide whether this is bug, use case, docs, or support",2026-04-18T00:05:37Z
issue,158343,Upgrade to latest XNNPACK,https://github.com/pytorch/pytorch/issues/158343,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: xnnpack,"decide whether this is bug, use case, docs, or support",2026-04-17T21:17:16Z
issue,180712,Add a lazy import hook for autoloading?,https://github.com/pytorch/pytorch/issues/180712,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: backend, module: ux","decide whether this is bug, use case, docs, or support",2026-04-17T20:10:18Z
issue,157807,[RFC] Replace setuptools build backend with scikit-build-core,https://github.com/pytorch/pytorch/issues/157807,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: build,"decide whether this is bug, use case, docs, or support",2026-04-16T23:16:52Z
issue,180596,capture_dynamic_output_shape_ops=True crashes 8/50 HuggingFace models with GuardOnDataDependentSymNode,https://github.com/pytorch/pytorch/issues/180596,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dynamic shapes, module: guards, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-04-16T19:42:40Z
issue,180595,capture_scalar_outputs=True crashes 7/50 HuggingFace models with PendingUnbackedSymbolNotFound,https://github.com/pytorch/pytorch/issues/180595,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dynamic shapes, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-04-16T19:42:31Z
issue,179520,Different results for with and without torch.compile on flux.1-dev model,https://github.com/pytorch/pytorch/issues/179520,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-04-16T17:50:16Z
issue,173897,XPU dist.all_reduce is async: torch.xpu.synchronize does not wait for collective completion,https://github.com/pytorch/pytorch/issues/173897,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: xpu, oncall: distributed","decide whether this is bug, use case, docs, or support",2026-04-16T14:52:44Z
issue,134956,Pytorch XPU Windows build failed in cmake rerun loop due to the source code deep path,https://github.com/pytorch/pytorch/issues/134956,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: xpu,"decide whether this is bug, use case, docs, or support",2026-04-16T08:38:48Z
issue,179858,torch.compile doesn't use async all_to_all,https://github.com/pytorch/pytorch/issues/179858,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dtensor, module: dynamo, oncall: distributed, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-04-15T19:10:38Z
issue,178883,`torch.compile` produces different output for model with `amax`/`amin` + partial reuse reduction pattern,https://github.com/pytorch/pytorch/issues/178883,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: aotdispatch, module: correctness (silent), module: cuda, module: decompositions","decide whether this is bug, use case, docs, or support",2026-04-15T14:43:28Z
issue,177225,mps test got Exception: mean(): could not infer output dtype.,https://github.com/pytorch/pytorch/issues/177225,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: mps, module: primtorch, module: tests","decide whether this is bug, use case, docs, or support",2026-04-15T11:00:01Z
issue,148402,Generate two reduction loops for vectorization,https://github.com/pytorch/pytorch/issues/148402,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-04-15T00:56:02Z
issue,154356,"Q.size(-1) == m INTERNAL ASSERT FAILED at ""/pytorch/aten/src/ATen/native/BatchLinearAlgebra.cpp",https://github.com/pytorch/pytorch/issues/154356,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: error checking, module: linear algebra","decide whether this is bug, use case, docs, or support",2026-04-14T19:23:25Z
issue,180362,RFC: Power (ppc64le) CI and Build Infrastructure Plan,https://github.com/pytorch/pytorch/issues/180362,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: build, module: ci, module: power","decide whether this is bug, use case, docs, or support",2026-04-14T18:22:31Z
issue,170753,[CI][Windows][Infrastructure] Support for RTX GPU workflows on Windows,https://github.com/pytorch/pytorch/issues/170753,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: ci, module: windows","decide whether this is bug, use case, docs, or support",2026-04-14T16:12:33Z
issue,169277,[Accelerator] Hardware native support versus Software simulated support for Device Capabilties,https://github.com/pytorch/pytorch/issues/169277,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: accelerator,"decide whether this is bug, use case, docs, or support",2026-04-14T11:10:33Z
issue,179709,Cannot export model with CenterCrop to ONNX,https://github.com/pytorch/pytorch/issues/179709,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: onnx,"decide whether this is bug, use case, docs, or support",2026-04-13T19:12:22Z
issue,147478,Blackwell Only Bugs,https://github.com/pytorch/pytorch/issues/147478,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: cuda, module: inductor, module: tests, oncall: export","decide whether this is bug, use case, docs, or support",2026-04-13T17:23:04Z
issue,150725,Continuous calls to nn.Linear in fp32 on the 5090D cause severe performance degradation,https://github.com/pytorch/pytorch/issues/150725,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: cuda, module: nn, module: performance","decide whether this is bug, use case, docs, or support",2026-04-13T17:23:03Z
issue,153555,Updated Scaled_mm to support more scaling formats via CuBlas,https://github.com/pytorch/pytorch/issues/153555,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: cuda, module: floatx (formerly float8)","decide whether this is bug, use case, docs, or support",2026-04-13T17:23:02Z
issue,169015,[CI][CUDA][B200] Test_weight_norm_conv2d_cuda Failure,https://github.com/pytorch/pytorch/issues/169015,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: cuda,"decide whether this is bug, use case, docs, or support",2026-04-13T17:22:55Z
issue,178839,[ROCm] MIOpen Gemm solver receives workspace_size=0 via legacy Find1 API,https://github.com/pytorch/pytorch/issues/178839,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: convolution, module: rocm","decide whether this is bug, use case, docs, or support",2026-04-13T17:12:02Z
issue,146542,DeepSeek: mixed precision optimizers (BF16AdamW),https://github.com/pytorch/pytorch/issues/146542,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: optimizer,"decide whether this is bug, use case, docs, or support",2026-04-13T15:43:07Z
issue,159770,extract statistics from attention weights in FlexAttention,https://github.com/pytorch/pytorch/issues/159770,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flex attention, module: higher order operators, module: pt2-dispatcher, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-04-13T14:50:21Z
issue,48972,non-negative least squares solver feature request,https://github.com/pytorch/pytorch/issues/48972,needs maintainer decision,28,open issue has no stronger automated bucket,"decide whether this is bug, use case, docs, or support",2026-04-13T08:50:09Z
issue,127176,"RuntimeError: ""_amp_foreach_non_finite_check_and_unscale_cuda"" not implemented for 'BFloat16'",https://github.com/pytorch/pytorch/issues/127176,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: amp (automated mixed precision), module: bfloat16, module: mta","decide whether this is bug, use case, docs, or support",2026-04-13T05:08:08Z
issue,121122,Evaluate defaulting `TORCHINDUCTOR_CACHE_DIR` to `~/.cache` on Linux systems,https://github.com/pytorch/pytorch/issues/121122,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-04-11T23:49:42Z
issue,25174,BCEWithLogitsLoss expects wrong shape of weight (#classes instead of batch size),https://github.com/pytorch/pytorch/issues/25174,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: loss, module: nn","decide whether this is bug, use case, docs, or support",2026-04-11T17:08:12Z
issue,157345,nn.rmsnorm is super slower than nn.layernorm,https://github.com/pytorch/pytorch/issues/157345,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: nn, module: norms and normalization, module: performance","decide whether this is bug, use case, docs, or support",2026-04-11T07:46:04Z
issue,179510,RuntimeError: Argument 'primals_out' of Node 'results' was used before it has been defined! Please check that Nodes in the graph are topologically ordered,https://github.com/pytorch/pytorch/issues/179510,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: functorch, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-04-10T21:10:22Z
issue,59194,Add numerically-stable function for angle between vectors,https://github.com/pytorch/pytorch/issues/59194,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: distance functions, module: numerical-stability","decide whether this is bug, use case, docs, or support",2026-04-09T18:22:45Z
issue,128121,2nd compile of deepcopy(model) fails on multiple ubuntu-pc (fatal error: Python.h: file not found),https://github.com/pytorch/pytorch/issues/128121,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dynamo, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-04-09T18:09:18Z
issue,145374,Memory Leak in MPS Backend During LSTM Iterations (Out of Memory Error),https://github.com/pytorch/pytorch/issues/145374,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: memory usage, module: mps, module: rnn","decide whether this is bug, use case, docs, or support",2026-04-09T02:11:29Z
issue,146691,MPS support `torch.linalg.norm` on complex numbers,https://github.com/pytorch/pytorch/issues/146691,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: complex, module: linear algebra","decide whether this is bug, use case, docs, or support",2026-04-09T00:13:29Z
issue,179643,Call Hierarchy Metadata,https://github.com/pytorch/pytorch/issues/179643,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dynamo, module: fx, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-04-08T18:00:48Z
issue,127605,RuntimeError: CUDA error: an illegal memory access was encountered with specific input shape and Conv1d,https://github.com/pytorch/pytorch/issues/127605,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: cuda, module: cudnn, module: nn","decide whether this is bug, use case, docs, or support",2026-04-08T14:51:38Z
issue,127078,Wrong Flag defined for ppc64le arch,https://github.com/pytorch/pytorch/issues/127078,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: cpu,"decide whether this is bug, use case, docs, or support",2026-04-07T19:15:04Z
issue,150655,Create better alerting for binary size validations and time it takes to build the binary,https://github.com/pytorch/pytorch/issues/150655,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: binaries,"decide whether this is bug, use case, docs, or support",2026-04-07T18:58:17Z
issue,169670,"Energy normalization error on torch.fft.fft2 and torch.fft.ifft2 when the matrix size = [2048, 2048]",https://github.com/pytorch/pytorch/issues/169670,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: intel, module: linear algebra, module: mkl","decide whether this is bug, use case, docs, or support",2026-04-07T13:05:23Z
issue,55937,test_stream_event_nogil: Is the test making a wrong assumption?,https://github.com/pytorch/pytorch/issues/55937,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: rocm, module: tests","decide whether this is bug, use case, docs, or support",2026-04-07T11:42:56Z
issue,174913,Test: TestLinalg.test_tensorinv,https://github.com/pytorch/pytorch/issues/174913,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: linear algebra, module: rocm","decide whether this is bug, use case, docs, or support",2026-04-06T21:56:35Z
issue,113045,[Feature][DTensor] Manage additional `_padded_local_tensor` attribute,https://github.com/pytorch/pytorch/issues/113045,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dtensor, oncall: distributed","decide whether this is bug, use case, docs, or support",2026-04-06T17:57:20Z
issue,179448,[DTensor] Inconsistent output placements of sharded DTensor indexing,https://github.com/pytorch/pytorch/issues/179448,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dtensor, oncall: distributed","decide whether this is bug, use case, docs, or support",2026-04-06T17:46:20Z
issue,179382,Support dropout in varlen_attn,https://github.com/pytorch/pytorch/issues/179382,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: sdpa,"decide whether this is bug, use case, docs, or support",2026-04-05T14:37:43Z
issue,178749,[FSDP] Multi-GPU distributed tests crash with NCCL 2.29.7 NVLS multicast slot exhaustion on H200,https://github.com/pytorch/pytorch/issues/178749,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: fsdp, oncall: distributed","decide whether this is bug, use case, docs, or support",2026-04-05T09:12:37Z
issue,133168,The more detailed GELU formula should be added to the doc of `GELU()`:,https://github.com/pytorch/pytorch/issues/133168,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: nn,"decide whether this is bug, use case, docs, or support",2026-04-05T06:23:39Z
issue,160875,Add `uv` installation commands to pytorch.org,https://github.com/pytorch/pytorch/issues/160875,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: docs, oncall: releng","decide whether this is bug, use case, docs, or support",2026-04-03T15:05:30Z
issue,171158,`torch.func.grad` to allow some inplace ops,https://github.com/pytorch/pytorch/issues/171158,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: autograd, module: functorch","decide whether this is bug, use case, docs, or support",2026-04-03T12:16:41Z
issue,167540,"[Dtensor]:change the test_mm shape from (12,8) * (8,16) to (512, 512) * (512, 512), throw assert error",https://github.com/pytorch/pytorch/issues/167540,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dtensor, oncall: distributed","decide whether this is bug, use case, docs, or support",2026-04-02T23:19:19Z
issue,178482,torch.compile(dynamic=True) ignores incompatible `out` tensor for `torch.empty` and returns wrong shape,https://github.com/pytorch/pytorch/issues/178482,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: correctness (silent), module: tensor creation","decide whether this is bug, use case, docs, or support",2026-04-02T22:40:43Z
issue,178930,[torch.compile] Better way to override the default backend,https://github.com/pytorch/pytorch/issues/178930,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: backend, module: compile ux, module: dynamo, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-04-02T21:23:58Z
issue,158710,[feature request] Inplace downcast dtype conversion,https://github.com/pytorch/pytorch/issues/158710,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: memory usage,"decide whether this is bug, use case, docs, or support",2026-04-02T18:34:56Z
issue,175695,Multi Tensor Apply ops have low occupancy due to a maximum of 320 thread blocks per kernel launch,https://github.com/pytorch/pytorch/issues/175695,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: cuda, module: mta, module: performance","decide whether this is bug, use case, docs, or support",2026-04-02T14:35:49Z
issue,179124,grid_sampler_2d_backward_cuda is not deterministic when torch.use_deterministic_algorithms(True),https://github.com/pytorch/pytorch/issues/179124,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: cuda, module: determinism, module: interpolation","decide whether this is bug, use case, docs, or support",2026-04-02T06:44:57Z
issue,178864,torch.multinomial on CUDA with zero-sum probabilities triggers device-side assert instead of invalid multinomial distribution error,https://github.com/pytorch/pytorch/issues/178864,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: cuda, module: edge cases, module: error checking","decide whether this is bug, use case, docs, or support",2026-04-01T22:23:35Z
issue,178988,Cannot build documentation (`RecursionError`),https://github.com/pytorch/pytorch/issues/178988,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: build, module: docs","decide whether this is bug, use case, docs, or support",2026-04-01T20:24:49Z
issue,113063,Rework Dynamic Benchmarks To Actually Vary Shapes,https://github.com/pytorch/pytorch/issues/113063,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: benchmark, module: dynamic shapes, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-04-01T18:58:08Z
issue,151997,torch.compile doesnot support index with tensor,https://github.com/pytorch/pytorch/issues/151997,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dynamic shapes, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-04-01T15:40:20Z
issue,140463,Ban all operations on sympy expressions in Inductor,https://github.com/pytorch/pytorch/issues/140463,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dynamic shapes, module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-04-01T15:37:42Z
issue,90900,Check that SymPy semantics match Python semantics,https://github.com/pytorch/pytorch/issues/90900,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: dynamic shapes,"decide whether this is bug, use case, docs, or support",2026-03-31T23:20:19Z
issue,44282,PyTorch wheel's own OpenMP library clashing with system-wide OpenMP library at runtime,https://github.com/pytorch/pytorch/issues/44282,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: openmp,"decide whether this is bug, use case, docs, or support",2026-03-31T14:48:09Z
issue,178820,Add Dynamo support for dist.record_comm (capture_profiler_record_comm),https://github.com/pytorch/pytorch/issues/178820,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: c10d, module: dynamo, oncall: distributed, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-03-31T02:38:06Z
issue,178616,[DTensor] reshape is dispatched to aten.view instead of aten.reshape,https://github.com/pytorch/pytorch/issues/178616,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dtensor, oncall: distributed","decide whether this is bug, use case, docs, or support",2026-03-30T20:21:12Z
issue,165070,Dynamically load CUDSS library using pypi wheel to save wheel space,https://github.com/pytorch/pytorch/issues/165070,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: build, module: cuda","decide whether this is bug, use case, docs, or support",2026-03-30T18:39:43Z
issue,141827,Compiled and masked MHA running with DDP and `no_grad()` leads to Graph break due to unsupported builtin torch._C._distributed_c10d.PyCapsule._broadcast_coalesced,https://github.com/pytorch/pytorch/issues/141827,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: nn, oncall: distributed","decide whether this is bug, use case, docs, or support",2026-03-30T18:21:46Z
issue,166563,[RFC] Modifying Getting started page for Experimental Wheel Variant Support,https://github.com/pytorch/pytorch/issues/166563,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: docs,"decide whether this is bug, use case, docs, or support",2026-03-30T17:49:33Z
issue,178593,torch.gcd returns incorrect results due to automatic dtype casting between signed and unsigned integers,https://github.com/pytorch/pytorch/issues/178593,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: correctness (silent), module: type promotion, module: unsigned int","decide whether this is bug, use case, docs, or support",2026-03-30T17:30:40Z
issue,178614,bag boundry is not calculated correctly in EmbeddingBag when setting include_last_offset = True,https://github.com/pytorch/pytorch/issues/178614,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: correctness (silent), module: cuda, module: embedding, module: nn","decide whether this is bug, use case, docs, or support",2026-03-30T17:28:57Z
issue,177599,arange CPU kernel has different precisions for scalar and vectorized executions,https://github.com/pytorch/pytorch/issues/177599,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: cpu, module: half, module: numerical-stability","decide whether this is bug, use case, docs, or support",2026-03-30T17:09:11Z
issue,178345,[build env combo] elainewy,https://github.com/pytorch/pytorch/issues/178345,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: ci,"decide whether this is bug, use case, docs, or support",2026-03-30T17:07:13Z
issue,177548,Update Internal fbsource CK,https://github.com/pytorch/pytorch/issues/177548,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: rocm, module: third_party","decide whether this is bug, use case, docs, or support",2026-03-30T15:53:36Z
issue,162872,[MPS] dead lock when calling `Event.synchronize()` before `Event.elapsed_time()`,https://github.com/pytorch/pytorch/issues/162872,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: deadlock, module: mps","decide whether this is bug, use case, docs, or support",2026-03-30T07:16:53Z
issue,178472,conv1d on meta device path skips groups validation and returns an invalid output shape,https://github.com/pytorch/pytorch/issues/178472,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: convolution, module: error checking, module: meta tensors","decide whether this is bug, use case, docs, or support",2026-03-27T02:39:11Z
issue,175022,[RFC] Cross-Repository CI Relay for PyTorch Out-of-Tree Backends,https://github.com/pytorch/pytorch/issues/175022,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: ci,"decide whether this is bug, use case, docs, or support",2026-03-26T20:33:42Z
issue,175409,Choose rounding mode when casting to dtype,https://github.com/pytorch/pytorch/issues/175409,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: cuda, module: floatx (formerly float8), module: inductor, module: type promotion","decide whether this is bug, use case, docs, or support",2026-03-26T18:01:13Z
issue,178511,[torch.compile] allow_in_graph called inside compiled function causes Unsupported error,https://github.com/pytorch/pytorch/issues/178511,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dynamo, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-03-26T15:11:33Z
issue,174045,Suboptimal `affine_grid` implementation,https://github.com/pytorch/pytorch/issues/174045,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: nn, module: performance","decide whether this is bug, use case, docs, or support",2026-03-26T14:50:48Z
issue,154299,ImportError: libcudnn.so.9: cannot open shared object file: No such file or directory,https://github.com/pytorch/pytorch/issues/154299,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: binaries, module: cuda","decide whether this is bug, use case, docs, or support",2026-03-26T07:46:36Z
issue,154052,Most requested ops for the MPS backend,https://github.com/pytorch/pytorch/issues/154052,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: mps,"decide whether this is bug, use case, docs, or support",2026-03-26T04:58:41Z
issue,178091,"propagate_single_input_strategy copies input tensor_meta to output, producing wrong dtype for ops like convert_element_type",https://github.com/pytorch/pytorch/issues/178091,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dtensor, oncall: distributed","decide whether this is bug, use case, docs, or support",2026-03-25T21:10:29Z
issue,176065,Errors in Activation Checkpointing tutorial,https://github.com/pytorch/pytorch/issues/176065,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: docs,"decide whether this is bug, use case, docs, or support",2026-03-25T20:48:47Z
issue,170837,MPS backend - inconsistent results for batched inference on BERT/RoBERTa,https://github.com/pytorch/pytorch/issues/170837,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: correctness (silent), module: mps","decide whether this is bug, use case, docs, or support",2026-03-25T20:48:04Z
issue,71381,Cannot import PyTorch in Alpine Docker Container,https://github.com/pytorch/pytorch/issues/71381,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: binaries, module: build","decide whether this is bug, use case, docs, or support",2026-03-25T20:18:38Z
issue,44027,Immutable (read-only) tensors,https://github.com/pytorch/pytorch/issues/44027,needs maintainer decision,28,open issue has no stronger automated bucket,"decide whether this is bug, use case, docs, or support",2026-03-25T13:02:47Z
issue,166291,"""FSDP does not support non-contiguous parameters yet"" prevents using channels-last convolutions",https://github.com/pytorch/pytorch/issues/166291,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: oncall: distributed,"decide whether this is bug, use case, docs, or support",2026-03-25T04:57:00Z
issue,177935,[RFC] Add A Cross-Backend Device Name API: `torch.accelerator.current_device_name()`,https://github.com/pytorch/pytorch/issues/177935,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: accelerator,"decide whether this is bug, use case, docs, or support",2026-03-25T03:25:24Z
issue,177021,linear attention / mamba2 OP,https://github.com/pytorch/pytorch/issues/177021,needs maintainer decision,28,open issue has no stronger automated bucket,"decide whether this is bug, use case, docs, or support",2026-03-24T20:20:38Z
issue,177014,`cudaErrorInvalidDeviceFunction` error due to use of `__CUDA_ARCH__` in host code,https://github.com/pytorch/pytorch/issues/177014,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: build, module: cuda","decide whether this is bug, use case, docs, or support",2026-03-24T19:08:09Z
issue,178052,Deterministic implementation of CTC Loss,https://github.com/pytorch/pytorch/issues/178052,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: cuda, module: determinism, module: loss","decide whether this is bug, use case, docs, or support",2026-03-24T17:50:12Z
issue,177910,nvshmem symm memory backend causes stuck sometimes for helion distributed kernel autotuning,https://github.com/pytorch/pytorch/issues/177910,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: cuda, module: deadlock, module: helion, module: symm_mem","decide whether this is bug, use case, docs, or support",2026-03-24T15:11:32Z
issue,99037,AttributeError: type object 'torch._C._profiler.ProfilerActivity' has no attribute 'MPS',https://github.com/pytorch/pytorch/issues/99037,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: mps, oncall: profiler","decide whether this is bug, use case, docs, or support",2026-03-24T08:22:46Z
issue,177771,Accelerators: torch.Stream.record_event doesn't handle torch.cuda.Events even though torch.cuda.Event is a torch.Event subclass,https://github.com/pytorch/pytorch/issues/177771,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: accelerator, module: cuda","decide whether this is bug, use case, docs, or support",2026-03-23T22:17:19Z
issue,177477,Misleading error message (INTERNAL ASSERT FAILED) in `jiterator._create_multi_output_jit_fn` when passing CPU tensors,https://github.com/pytorch/pytorch/issues/177477,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: cuda, module: error checking, module: jiterator","decide whether this is bug, use case, docs, or support",2026-03-23T22:17:19Z
issue,177322,Support TORCH_SHOW_CPP_STACKTRACES=1 on Windows,https://github.com/pytorch/pytorch/issues/177322,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: devx, module: windows","decide whether this is bug, use case, docs, or support",2026-03-23T22:17:15Z
issue,165506,[torch.compile] dynamic shape kernel mismatch,https://github.com/pytorch/pytorch/issues/165506,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dynamic shapes, module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-03-23T22:17:12Z
issue,170361,[Dynamo] Use VariableBuilder/SourcelessBuilder consistently,https://github.com/pytorch/pytorch/issues/170361,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dynamo, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-03-23T19:34:33Z
issue,177663,[DTensor] greedy redistribution cost favors all_gather over all_to_all,https://github.com/pytorch/pytorch/issues/177663,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dtensor, oncall: distributed","decide whether this is bug, use case, docs, or support",2026-03-23T18:11:23Z
issue,177969,test utilities should intersect `torch.accelerator.get_device_capability` when resolving dtypes used for a test,https://github.com/pytorch/pytorch/issues/177969,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: accelerator, module: tests","decide whether this is bug, use case, docs, or support",2026-03-23T17:17:44Z
issue,74746,Change the type hint for nn.Module.__call__ to be friendly to overrides.,https://github.com/pytorch/pytorch/issues/74746,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: nn, module: typing","decide whether this is bug, use case, docs, or support",2026-03-23T16:22:00Z
issue,178132,torch.compile streams support in AOTI,https://github.com/pytorch/pytorch/issues/178132,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: compiled streams, module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-03-23T08:29:51Z
issue,177957,api shows unexpected difference for flatten(): numpy vs. pytorch,https://github.com/pytorch/pytorch/issues/177957,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: docs,"decide whether this is bug, use case, docs, or support",2026-03-22T14:45:46Z
issue,178072,[PERFORMANCE] [NN] Change `torch._C._get_tracing_state()` to `torch._C._is_tracing()` for around 15% improved performance of `module.py` `_call_impl()`,https://github.com/pytorch/pytorch/issues/178072,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: nn, module: performance","decide whether this is bug, use case, docs, or support",2026-03-21T20:07:52Z
issue,53721,PyTorch on Embedded Hardware,https://github.com/pytorch/pytorch/issues/53721,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: cuda,"decide whether this is bug, use case, docs, or support",2026-03-21T12:46:32Z
issue,177260,Standardize device-agnostic test cases via instantiate_device_type_tests,https://github.com/pytorch/pytorch/issues/177260,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: tests,"decide whether this is bug, use case, docs, or support",2026-03-20T22:16:51Z
issue,177830,Why register_replacement fails to support fuzzy matching for optional operator parameters in search_fn/search_fn_pattern,https://github.com/pytorch/pytorch/issues/177830,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: compile ux, module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-03-20T21:32:20Z
issue,177970,`TEST_DEVICE_BIAS` linter scope should be expanded,https://github.com/pytorch/pytorch/issues/177970,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: ci, module: tests","decide whether this is bug, use case, docs, or support",2026-03-20T16:02:33Z
issue,159635,[BUG] DTensor (PyTorch native TP) + dynamic shapes + torch.compile failed!,https://github.com/pytorch/pytorch/issues/159635,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-03-20T06:58:25Z
issue,177896,Bot to notify PoC if ci pin updater fails,https://github.com/pytorch/pytorch/issues/177896,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: ci, module: devx","decide whether this is bug, use case, docs, or support",2026-03-19T20:30:03Z
issue,155720,[rocm] HIP Graph capture raises segmentation fault on AMD GPU but CUDA Graph capture succeeds on Nvidia GPU,https://github.com/pytorch/pytorch/issues/155720,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: rocm,"decide whether this is bug, use case, docs, or support",2026-03-19T16:12:54Z
issue,177859,Combining `bfloat16` with complex scalars raises `NotImplementedError`,https://github.com/pytorch/pytorch/issues/177859,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: bfloat16, module: complex","decide whether this is bug, use case, docs, or support",2026-03-19T15:39:14Z
issue,176274,PyTorch + Math libraries: redundant `copy_` calls when working with buffer views,https://github.com/pytorch/pytorch/issues/176274,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: cpu, module: cuda, module: linear algebra, module: memory usage","decide whether this is bug, use case, docs, or support",2026-03-19T13:11:42Z
issue,177833,StepLR accepts step_size=0 and crashes with ZeroDivisionError on first step(),https://github.com/pytorch/pytorch/issues/177833,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: error checking, module: lrscheduler, module: optimizer","decide whether this is bug, use case, docs, or support",2026-03-19T10:38:49Z
issue,134767,"Setting a negative, `int` and `bool` value to `label_smoothing` argument of `nn.CrossEntropyLoss()` works against the error messages",https://github.com/pytorch/pytorch/issues/134767,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: loss,"decide whether this is bug, use case, docs, or support",2026-03-19T10:24:29Z
issue,164429,Add support for Intel GPU Memory Information,https://github.com/pytorch/pytorch/issues/164429,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: xpu,"decide whether this is bug, use case, docs, or support",2026-03-19T02:07:19Z
issue,136495,[dynamo] enable TorchDispatchMode for eager part when graph breaks,https://github.com/pytorch/pytorch/issues/136495,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dynamo, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-03-18T23:59:16Z
issue,137749,Graph break on wait() of Torchrec's Awaitable instance,https://github.com/pytorch/pytorch/issues/137749,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dynamo, oncall: distributed, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-03-18T23:58:16Z
issue,94792,Compiling libtorch from Source on Mac Beyond v1.11.0,https://github.com/pytorch/pytorch/issues/94792,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: build, module: cpp, module: macos","decide whether this is bug, use case, docs, or support",2026-03-18T19:56:02Z
issue,123148,GRU is super slow on MPS backend,https://github.com/pytorch/pytorch/issues/123148,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: mps, module: performance, module: rnn","decide whether this is bug, use case, docs, or support",2026-03-18T18:34:28Z
issue,167297,"DISABLED test_max_autotune_addmm_max_autotune_gemm_backends_ATen,Triton,CK_x_shape1 (__main__.TestCKBackend)",https://github.com/pytorch/pytorch/issues/167297,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, module: inductor, module: rocm, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-03-18T13:37:08Z
issue,167272,"DISABLED test_max_autotune_addmm_max_autotune_gemm_backends_ATen,Triton,CK_x_shape2 (__main__.TestCKBackend)",https://github.com/pytorch/pytorch/issues/167272,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, module: inductor, module: rocm, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-03-18T13:36:22Z
issue,97668,sparse_csr_tensor matmul wrong output in bfloat16,https://github.com/pytorch/pytorch/issues/97668,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: sparse,"decide whether this is bug, use case, docs, or support",2026-03-18T11:49:23Z
issue,157989,[XPU] torch.xpu.mem_get_info() query failed or reports Out-Of-Memory even when there are enough free memory,https://github.com/pytorch/pytorch/issues/157989,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: windows, module: xpu","decide whether this is bug, use case, docs, or support",2026-03-18T06:14:16Z
issue,164966,XPU OOM when allocate tensor according to its reported available memory,https://github.com/pytorch/pytorch/issues/164966,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: memory usage, module: xpu","decide whether this is bug, use case, docs, or support",2026-03-18T05:51:30Z
issue,76324,Bessel and Related Functions,https://github.com/pytorch/pytorch/issues/76324,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: special,"decide whether this is bug, use case, docs, or support",2026-03-18T00:42:34Z
issue,28224,Allow to disable polling for CUDA synchronization,https://github.com/pytorch/pytorch/issues/28224,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: cuda, module: performance","decide whether this is bug, use case, docs, or support",2026-03-17T20:00:22Z
issue,164593,[CI] Make sure B200 benchmark testing covers regression such as https://github.com/triton-lang/triton/issues/8328,https://github.com/pytorch/pytorch/issues/164593,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: benchmark, module: ci","decide whether this is bug, use case, docs, or support",2026-03-17T19:22:55Z
issue,112344,[test/dynamo] BE: cleanup `test_misc.py`,https://github.com/pytorch/pytorch/issues/112344,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dynamo, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-03-16T19:50:06Z
issue,177393,[release 2.11][vllm] Language Models Test (Extended Generation) test_models[True-False-5-32-bigcode/starcoder2-3b],https://github.com/pytorch/pytorch/issues/177393,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: tests, module: vllm","decide whether this is bug, use case, docs, or support",2026-03-16T17:08:52Z
issue,177053,[inductor] realize_reads_threshold could be tuned differently with auto-chunking,https://github.com/pytorch/pytorch/issues/177053,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: oncall: pt2,"decide whether this is bug, use case, docs, or support",2026-03-16T15:34:24Z
issue,177520,"Memory management, memory not returned after CPU -> GPU",https://github.com/pytorch/pytorch/issues/177520,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: cpp, module: cuda, module: memory usage, module: windows","decide whether this is bug, use case, docs, or support",2026-03-16T14:22:57Z
issue,174481,Modular Dynamo FX Graph in torch 2.10,https://github.com/pytorch/pytorch/issues/174481,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dynamo, module: fx, oncall: export, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-03-16T10:47:57Z
issue,177425,binary_populate_env.sh: PYTORCH_ROOT optional check is inconsistent with later required use,https://github.com/pytorch/pytorch/issues/177425,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: ci,"decide whether this is bug, use case, docs, or support",2026-03-13T21:59:10Z
issue,163974,C++ build dependencies,https://github.com/pytorch/pytorch/issues/163974,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: cpp, module: doc infra, module: docs","decide whether this is bug, use case, docs, or support",2026-03-13T21:35:30Z
issue,119455,Users confusing torch.distributed.nn.functional with functional collectives,https://github.com/pytorch/pytorch/issues/119455,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: c10d, oncall: distributed, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-03-13T19:23:05Z
issue,177386,`CatBackward0` rejects valid gradient after padded round-trip on jagged nested tensors,https://github.com/pytorch/pytorch/issues/177386,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: autograd, module: nestedtensor","decide whether this is bug, use case, docs, or support",2026-03-13T13:23:52Z
issue,125678,Unnecessary warning when numpy not installed,https://github.com/pytorch/pytorch/issues/125678,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: build, module: numpy","decide whether this is bug, use case, docs, or support",2026-03-13T13:11:05Z
issue,177372,Add torch.narrow_scatter operator to complete the *_copy/*_scatter operator family.,https://github.com/pytorch/pytorch/issues/177372,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: functionalization, module: functorch, module: pt2-dispatcher, module: scatter & gather ops","decide whether this is bug, use case, docs, or support",2026-03-13T08:20:09Z
issue,177050,PyTorch 2.10-2.11 Linux x86 Binary size increase investigation,https://github.com/pytorch/pytorch/issues/177050,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: binaries,"decide whether this is bug, use case, docs, or support",2026-03-12T20:05:37Z
issue,177268,[OpenReg] Add testing operator framework usage examples,https://github.com/pytorch/pytorch/issues/177268,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: openreg, module: tests","decide whether this is bug, use case, docs, or support",2026-03-12T15:07:49Z
issue,136748,Inconsistent behavior of cdist with half-precision inputs,https://github.com/pytorch/pytorch/issues/136748,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: cpu, module: norms and normalization","decide whether this is bug, use case, docs, or support",2026-03-12T14:08:58Z
issue,122806,Deprecate accessing FakeTensor.data_ptr outside of torch.compile,https://github.com/pytorch/pytorch/issues/122806,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: faketensor,"decide whether this is bug, use case, docs, or support",2026-03-11T21:19:09Z
issue,118492,`torch.compile(dynamic=True)` compiles forever,https://github.com/pytorch/pytorch/issues/118492,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: compile-time, module: dynamic shapes, module: performance, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-03-11T21:18:22Z
issue,118521,Improve constant propagation in the dynamic shape case,https://github.com/pytorch/pytorch/issues/118521,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dynamic shapes, module: fx.passes","decide whether this is bug, use case, docs, or support",2026-03-11T20:56:45Z
issue,120105,Weird Artifact name: 'not_implemented' not registered being raised,https://github.com/pytorch/pytorch/issues/120105,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: faketensor, module: logging","decide whether this is bug, use case, docs, or support",2026-03-11T20:49:01Z
issue,118211,"Dynamo codegen for mapping outputs back to user variables should contain safety checks (no fake tensors, no variable trackers)",https://github.com/pytorch/pytorch/issues/118211,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: dynamo,"decide whether this is bug, use case, docs, or support",2026-03-11T20:40:11Z
issue,121818,torch._check but trust me,https://github.com/pytorch/pytorch/issues/121818,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: dynamic shapes,"decide whether this is bug, use case, docs, or support",2026-03-11T20:39:54Z
issue,107631,torch.fx.Interpreter modules don't get compiled,https://github.com/pytorch/pytorch/issues/107631,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dynamo, module: graph breaks","decide whether this is bug, use case, docs, or support",2026-03-11T20:34:53Z
issue,119407,"convolution_backward has different size results on ""CPU"", FakeTensor",https://github.com/pytorch/pytorch/issues/119407,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: faketensor,"decide whether this is bug, use case, docs, or support",2026-03-11T20:33:53Z
issue,113589,Fake tensor produces incorrect values w/ is_coalesced and sparse_coo,https://github.com/pytorch/pytorch/issues/113589,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: faketensor, module: sparse","decide whether this is bug, use case, docs, or support",2026-03-11T20:23:41Z
issue,88109,Python Dispatcher registrations beyond BackendSelect do nothing,https://github.com/pytorch/pytorch/issues/88109,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dispatch, module: python dispatcher","decide whether this is bug, use case, docs, or support",2026-03-11T20:17:22Z
issue,113643,Distinguish immutable/mutable fake tensor mode,https://github.com/pytorch/pytorch/issues/113643,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dynamo, module: faketensor","decide whether this is bug, use case, docs, or support",2026-03-11T20:15:04Z
issue,127772,Run existing eager DTensor tests under torch.compile,https://github.com/pytorch/pytorch/issues/127772,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dtensor, oncall: distributed, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-03-11T18:25:28Z
issue,170241,Allow for multiple stream captures to a cuda graph cudagraph at once.,https://github.com/pytorch/pytorch/issues/170241,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: cuda graphs,"decide whether this is bug, use case, docs, or support",2026-03-11T17:44:10Z
issue,104194,View ops on fake tensors can dispatch `detach`es to backend kernels,https://github.com/pytorch/pytorch/issues/104194,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: faketensor,"decide whether this is bug, use case, docs, or support",2026-03-11T16:13:41Z
issue,107582,[FakeTensor] `to` doesn't error with `allow_non_fake_inputs=False`,https://github.com/pytorch/pytorch/issues/107582,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: faketensor,"decide whether this is bug, use case, docs, or support",2026-03-11T15:59:16Z
issue,102187,Extend fake fast path to more situations,https://github.com/pytorch/pytorch/issues/102187,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: faketensor,"decide whether this is bug, use case, docs, or support",2026-03-11T15:47:36Z
issue,100985,"native_batch_norm has different size results on ""CPU"" vs ""META"" device",https://github.com/pytorch/pytorch/issues/100985,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: faketensor, module: meta tensors","decide whether this is bug, use case, docs, or support",2026-03-11T15:33:58Z
issue,91655,FakeTensors not moving between device properly on Module.cuda(),https://github.com/pytorch/pytorch/issues/91655,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: faketensor,"decide whether this is bug, use case, docs, or support",2026-03-11T15:27:43Z
issue,82616,FakeTensor Support For Pickling,https://github.com/pytorch/pytorch/issues/82616,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: faketensor,"decide whether this is bug, use case, docs, or support",2026-03-11T15:17:25Z
issue,84925,`is_pinned()` support in PrimTorch and FakeTensor.,https://github.com/pytorch/pytorch/issues/84925,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: faketensor, module: primtorch","decide whether this is bug, use case, docs, or support",2026-03-11T15:12:31Z
issue,177135,release notes commit script should respect user-specified labels,https://github.com/pytorch/pytorch/issues/177135,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: ci,"decide whether this is bug, use case, docs, or support",2026-03-11T14:01:42Z
issue,176625,The `padding` type should be set as Scaler for torch.nested.to_padded_tensor (currently set to double),https://github.com/pytorch/pytorch/issues/176625,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: correctness (silent), module: edge cases, module: nestedtensor","decide whether this is bug, use case, docs, or support",2026-03-11T00:05:22Z
issue,167766,Fix cudagraph flaky tests,https://github.com/pytorch/pytorch/issues/167766,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: cuda graphs, module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-03-10T22:12:36Z
issue,170535,Graphsafe RNG clone_state,https://github.com/pytorch/pytorch/issues/170535,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: cuda, module: cuda graphs","decide whether this is bug, use case, docs, or support",2026-03-10T22:10:38Z
issue,175557,torch.compile crashes during autoregressive decode with growing KV cache in cudagraph_trees,https://github.com/pytorch/pytorch/issues/175557,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: cuda graphs, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-03-10T22:06:16Z
issue,168820,Test: TestFP8Matmul.test_scaled_mm_vs_emulated_row_wise,https://github.com/pytorch/pytorch/issues/168820,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: floatx (formerly float8), module: linear algebra, module: rocm","decide whether this is bug, use case, docs, or support",2026-03-10T17:48:40Z
issue,176592,custom_op : custom_op overhead is much higher than autograd.Function (in eager),https://github.com/pytorch/pytorch/issues/176592,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: custom-operators, module: performance, module: pt2-dispatcher, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-03-10T17:18:50Z
issue,168279,Add flag to error on any Inductor specialization,https://github.com/pytorch/pytorch/issues/168279,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dynamic shapes, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-03-10T16:07:13Z
issue,156805,[RFC] Support the unified memory API GetAllocator and getDeviceAllocator for in-tree and out-of-tree backend,https://github.com/pytorch/pytorch/issues/156805,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: accelerator,"decide whether this is bug, use case, docs, or support",2026-03-10T07:48:48Z
issue,113669,document functional_collectives,https://github.com/pytorch/pytorch/issues/113669,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: docs, oncall: distributed, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-03-10T06:02:22Z
issue,155562,[Misc] test_foreach_add_different_mesh cannot work on machines with less than 4 GPUs,https://github.com/pytorch/pytorch/issues/155562,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dtensor, oncall: distributed","decide whether this is bug, use case, docs, or support",2026-03-10T05:57:51Z
issue,108971,Really slow compilation times for torch.compile causing distributed training errors,https://github.com/pytorch/pytorch/issues/108971,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: compile-time, module: ddp, oncall: distributed, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-03-10T05:56:23Z
issue,145726,Add correct `__repr__` for parallel distributed modules,https://github.com/pytorch/pytorch/issues/145726,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: oncall: distributed,"decide whether this is bug, use case, docs, or support",2026-03-10T05:45:45Z
issue,152442,`torch.compile` causes assertion error in distributed checkpoint wrapper test,https://github.com/pytorch/pytorch/issues/152442,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-03-10T04:58:34Z
issue,158519,`get_model_state_dict` output inconsistent before and after `set_model_state_dict` when using `torch.compile`,https://github.com/pytorch/pytorch/issues/158519,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dynamo, oncall: distributed, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-03-10T04:48:32Z
issue,155632,"Functional all_gather_into_tensor does not support stacking, fails when compiled",https://github.com/pytorch/pytorch/issues/155632,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dynamo, oncall: distributed, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-03-10T04:48:30Z
issue,168823,Test: TestFP8Matmul.test_blockwise_mxfp8_nvfp4_mxfp4_numerics,https://github.com/pytorch/pytorch/issues/168823,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: floatx (formerly float8), module: linear algebra, module: rocm","decide whether this is bug, use case, docs, or support",2026-03-10T01:27:44Z
issue,169182,Kineto library not installed,https://github.com/pytorch/pytorch/issues/169182,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: aotinductor, module: build, module: cpp-extensions","decide whether this is bug, use case, docs, or support",2026-03-09T23:25:35Z
issue,163716,torch.compile bug because aot autograd trace ignores requires_grad,https://github.com/pytorch/pytorch/issues/163716,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: aotdispatch, module: pt2-dispatcher, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-03-09T23:23:20Z
issue,54881,Often the documentation does not match the actual signature of the function,https://github.com/pytorch/pytorch/issues/54881,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: __torch_function__, module: docs","decide whether this is bug, use case, docs, or support",2026-03-09T23:23:12Z
issue,62982,Don't use try-catch to handle overload resolution handling in torch.ops,https://github.com/pytorch/pytorch/issues/62982,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: __torch_dispatch__, module: bootcamp","decide whether this is bug, use case, docs, or support",2026-03-09T23:23:04Z
issue,174711,add an alternative to compile time benchmarking in inductor for determinism that is reasonably performant,https://github.com/pytorch/pytorch/issues/174711,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: determinism, module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-03-09T23:09:21Z
issue,168099,Unify pointwise DTensor and NestedTensor OP Coverage. Adds over 100 op overloads to DTensor and about to 10 to NestedTensor,https://github.com/pytorch/pytorch/issues/168099,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dtensor, oncall: distributed","decide whether this is bug, use case, docs, or support",2026-03-09T20:16:34Z
issue,137505,RuntimeError: Backend nccl does not support allgather_into_tensor_coalesced,https://github.com/pytorch/pytorch/issues/137505,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: oncall: distributed,"decide whether this is bug, use case, docs, or support",2026-03-09T18:13:31Z
issue,176692,Dynamo does not know how to trace the builtin `<unknown module>.TensorBase._make_wrapper_subclass.`,https://github.com/pytorch/pytorch/issues/176692,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dynamo, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-03-09T16:19:17Z
issue,26207,Dispatch Tracing/Debugging,https://github.com/pytorch/pytorch/issues/26207,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: dispatch,"decide whether this is bug, use case, docs, or support",2026-03-08T04:04:10Z
issue,152213,Outdated install commands,https://github.com/pytorch/pytorch/issues/152213,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: docs,"decide whether this is bug, use case, docs, or support",2026-03-07T19:28:18Z
issue,165987,`triu_/tril_` ignores memory overlaps which can lead to silent correctness issues,https://github.com/pytorch/pytorch/issues/165987,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: correctness (silent), module: error checking, module: memory format","decide whether this is bug, use case, docs, or support",2026-03-07T07:07:44Z
issue,171620,Feedback about torch.autograd.graph.Node.next_functions,https://github.com/pytorch/pytorch/issues/171620,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: autograd, module: docs","decide whether this is bug, use case, docs, or support",2026-03-07T04:21:25Z
issue,163997,`F.scaled_dot_product_attention` returns all-`NaN` rows during evaluation but not training,https://github.com/pytorch/pytorch/issues/163997,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: nans and infs, module: sdpa","decide whether this is bug, use case, docs, or support",2026-03-07T00:37:30Z
issue,133511,Do dynamo/inductor support custom classes like TorchScript.,https://github.com/pytorch/pytorch/issues/133511,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: docs, module: dynamo, module: vllm, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-03-06T23:40:51Z
issue,174949,[vllm] CUBLAS_STATUS_INVALID_VALUE in cublasGemmEx after upgrading to PyTorch 2.10,https://github.com/pytorch/pytorch/issues/174949,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: cublas, module: cuda, module: vllm","decide whether this is bug, use case, docs, or support",2026-03-06T22:11:49Z
issue,176730,scaled_dot_product_attention_math usage and perf implications of returning the second value,https://github.com/pytorch/pytorch/issues/176730,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: mps, module: nn, module: performance, module: sdpa","decide whether this is bug, use case, docs, or support",2026-03-06T19:03:47Z
issue,176188,test_structured_trace.py's test_ddp_graphs and test_graph_breaks flaky on all platforms,https://github.com/pytorch/pytorch/issues/176188,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: rocm,"decide whether this is bug, use case, docs, or support",2026-03-06T14:56:17Z
issue,176687,Proposal: Pytorch agent harness for external contributors,https://github.com/pytorch/pytorch/issues/176687,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: ci,"decide whether this is bug, use case, docs, or support",2026-03-06T07:22:50Z
issue,176628,"[NN] Performance improvement in `nn.Module._call_impl()` by around 9% by optimizing fast-path for `forward_call()`, hook iterations and checks plus some minor changes",https://github.com/pytorch/pytorch/issues/176628,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: nn, module: performance","decide whether this is bug, use case, docs, or support",2026-03-06T02:01:35Z
issue,175538,[Inductor] Need a CustomIRPass Abstraction for Loop‑Level IR Transformations,https://github.com/pytorch/pytorch/issues/175538,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-03-05T17:02:56Z
issue,176166,DataLoader: collate_fn leads to high RAM usage,https://github.com/pytorch/pytorch/issues/176166,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dataloader, module: performance, module: windows","decide whether this is bug, use case, docs, or support",2026-03-05T14:50:31Z
issue,136003,[MPS] Inconsistent performance issues,https://github.com/pytorch/pytorch/issues/136003,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: mps, module: performance","decide whether this is bug, use case, docs, or support",2026-03-05T05:53:00Z
issue,101850,Unexpected modification to CPU affinity of Dataloader workers,https://github.com/pytorch/pytorch/issues/101850,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: multiprocessing, module: openmp","decide whether this is bug, use case, docs, or support",2026-03-05T04:01:15Z
issue,176507,Security hardening: Add HMAC integrity verification and authentication support to torch.compile Redis remote cache,https://github.com/pytorch/pytorch/issues/176507,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-03-05T00:22:39Z
issue,149809,LoadHIP.cmake should find_package(composable_kernel),https://github.com/pytorch/pytorch/issues/149809,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: build, module: rocm","decide whether this is bug, use case, docs, or support",2026-03-04T19:16:07Z
issue,176415,Output difference anomaly under equivalent migration in logdet operator,https://github.com/pytorch/pytorch/issues/176415,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: edge cases, module: linear algebra, module: nans and infs, module: windows","decide whether this is bug, use case, docs, or support",2026-03-04T09:54:18Z
issue,176413,Output difference anomaly under equivalent migration in logdet operator,https://github.com/pytorch/pytorch/issues/176413,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: linear algebra, module: windows","decide whether this is bug, use case, docs, or support",2026-03-04T09:52:26Z
issue,176412,Output difference on 2D float32 tensors with negative out-of-domain values,https://github.com/pytorch/pytorch/issues/176412,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: edge cases, module: nans and infs, module: windows","decide whether this is bug, use case, docs, or support",2026-03-04T09:51:34Z
issue,173885,[Inductor][CPU][float16] LayerNorm outputs NaN when input contains Inf,https://github.com/pytorch/pytorch/issues/173885,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: edge cases, module: inductor, module: nans and infs, module: nn","decide whether this is bug, use case, docs, or support",2026-03-04T06:38:36Z
issue,169926,Failed to build PyTorch with CUDA by using clang with undefined reference of const_data_ptr in TensorBase,https://github.com/pytorch/pytorch/issues/169926,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: arm, module: build, module: cuda","decide whether this is bug, use case, docs, or support",2026-03-04T06:34:33Z
issue,176385,Inconsistent cross-framework behavior (nan vs +/-inf) for float64 out-of-domain inputs,https://github.com/pytorch/pytorch/issues/176385,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: edge cases, module: nans and infs, module: special","decide whether this is bug, use case, docs, or support",2026-03-04T05:07:51Z
issue,176384,Output difference anomaly under equivalent migration in erfinv operator,https://github.com/pytorch/pytorch/issues/176384,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: edge cases, module: python frontend","decide whether this is bug, use case, docs, or support",2026-03-04T05:06:26Z
issue,175311,[RFC][Inductor] PyTorch Inductor cuTile Backend,https://github.com/pytorch/pytorch/issues/175311,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: cuda, module: helion, module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-03-04T03:24:08Z
issue,159155,Audit all usages of constrain_to_fx_strides. Many of them might want constrain_to_fake_tensor,https://github.com/pytorch/pytorch/issues/159155,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: correctness (silent), module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-03-03T19:57:48Z
issue,172987,Leveraging Unified Memory for MPS Tensors on Apple Silicone,https://github.com/pytorch/pytorch/issues/172987,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: memory usage, module: mps","decide whether this is bug, use case, docs, or support",2026-03-03T18:08:14Z
issue,175479,Feedback about LSTM,https://github.com/pytorch/pytorch/issues/175479,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: docs, module: rnn","decide whether this is bug, use case, docs, or support",2026-03-03T17:49:59Z
issue,173793,Numerical stability discrepancy in torch.compile: LayerNorm produces NaN for large inputs (~1e37) while Eager mode is stable,https://github.com/pytorch/pytorch/issues/173793,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: inductor, module: nans and infs, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-03-03T14:36:33Z
issue,154077,"accept 0-d tensor for `fill_value` parameter of `full`, `full_like`",https://github.com/pytorch/pytorch/issues/154077,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: cuda, module: python frontend","decide whether this is bug, use case, docs, or support",2026-03-03T14:02:36Z
issue,158371,Transition Adam and maybe other optimizers to fused code path by default to avoid `foreach=True`-specific VRAM peak due to temp TensorList for bias-corrected moments,https://github.com/pytorch/pytorch/issues/158371,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: optimizer, module: performance","decide whether this is bug, use case, docs, or support",2026-03-03T09:34:33Z
issue,175269,[dynamo] Refactor variables/builtin.py,https://github.com/pytorch/pytorch/issues/175269,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dynamo, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-03-02T21:48:35Z
issue,175738,"torch.export + torch.onnx.export(dynamo=True) gives incorrect results for scatter_reduce_(reduce=""mean"")",https://github.com/pytorch/pytorch/issues/175738,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: onnx,"decide whether this is bug, use case, docs, or support",2026-03-02T18:17:51Z
issue,176091,Windows CPU: SDPA FlashAttention gives wrong results with multiple intra-op threads (mixed OpenMP runtimes),https://github.com/pytorch/pytorch/issues/176091,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: correctness (silent), module: cpu, module: openmp, module: sdpa","decide whether this is bug, use case, docs, or support",2026-03-02T18:12:16Z
issue,176153,Libtorch - make call reset_parameters during init optional,https://github.com/pytorch/pytorch/issues/176153,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: cpp, module: nn","decide whether this is bug, use case, docs, or support",2026-03-02T14:10:51Z
issue,166123,"Can't compile torch-cublas-hgemm project in Windows on a torch newer than 2.8.0, I get complains involving compiled_autograd.h",https://github.com/pytorch/pytorch/issues/166123,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: cpp-extensions, module: windows","decide whether this is bug, use case, docs, or support",2026-03-02T10:18:48Z
issue,109592,test_memory_timeline fails on PPC due to extra temopraries,https://github.com/pytorch/pytorch/issues/109592,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: power, oncall: profiler","decide whether this is bug, use case, docs, or support",2026-03-01T20:03:11Z
issue,176078,custom_op with mutable optional arguments,https://github.com/pytorch/pytorch/issues/176078,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: python frontend,"decide whether this is bug, use case, docs, or support",2026-03-01T00:06:22Z
issue,72388,ArgMax for Multi Dimensional Tensor,https://github.com/pytorch/pytorch/issues/72388,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: reductions, module: ux","decide whether this is bug, use case, docs, or support",2026-02-28T06:21:42Z
issue,126804,"Torch fails to build with cub from CCCL v2.3.2 or newer, due to -D__CUDA_NO_HALF_OPERATORS__",https://github.com/pytorch/pytorch/issues/126804,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: build, module: cuda","decide whether this is bug, use case, docs, or support",2026-02-27T06:47:53Z
issue,175899,[CI][VLLM] Track disabled vLLM tests in PyTorch CI,https://github.com/pytorch/pytorch/issues/175899,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: ci,"decide whether this is bug, use case, docs, or support",2026-02-26T22:09:32Z
issue,175873,[MPS][Performance] SDPA op implementation is wasting resources by returning attention weights,https://github.com/pytorch/pytorch/issues/175873,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: mps, module: performance","decide whether this is bug, use case, docs, or support",2026-02-26T18:24:58Z
issue,50335,Function Request: scipy.interpolate.interp1d,https://github.com/pytorch/pytorch/issues/50335,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: interpolation, module: numpy","decide whether this is bug, use case, docs, or support",2026-02-26T16:30:18Z
issue,175850,RFC: Vmap Extension Points for Custom Container Types,https://github.com/pytorch/pytorch/issues/175850,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: functorch, module: vmap","decide whether this is bug, use case, docs, or support",2026-02-26T11:19:32Z
issue,170117,PyTorch Nightly's Torch Compile Feature Does Not Work on THOR due to Triton 3.6.x regression.,https://github.com/pytorch/pytorch/issues/170117,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: arm, module: binaries, module: cuda","decide whether this is bug, use case, docs, or support",2026-02-26T08:45:49Z
issue,175829,Why the same test is running twice in regular and slow shard,https://github.com/pytorch/pytorch/issues/175829,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: ci, module: tests","decide whether this is bug, use case, docs, or support",2026-02-26T04:44:08Z
issue,175737,GPUDirectStorage Status?,https://github.com/pytorch/pytorch/issues/175737,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: cuda, module: docs, module: serialization","decide whether this is bug, use case, docs, or support",2026-02-25T18:23:44Z
issue,174417,[release 2.11] [triton] test/inductor/test_loop_ordering.py::LoopOrderingTest::test_fp8_cast_and_t,https://github.com/pytorch/pytorch/issues/174417,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: floatx (formerly float8), module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-02-25T16:22:31Z
issue,174165,Fallback to Composite Implicit Autograd for Purely Functional Composite Custom Ops,https://github.com/pytorch/pytorch/issues/174165,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: autograd, module: custom-operators, module: dispatch, module: pt2-dispatcher","decide whether this is bug, use case, docs, or support",2026-02-25T05:10:54Z
issue,164696,Support torch._inductor.config.inplace_buffers for custom_op whenever possible,https://github.com/pytorch/pytorch/issues/164696,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: custom-operators, module: inductor, module: pt2-dispatcher, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-02-25T02:43:09Z
issue,173525,[MPS] layer_norm backward numerical issues,https://github.com/pytorch/pytorch/issues/173525,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: edge cases, module: mps","decide whether this is bug, use case, docs, or support",2026-02-25T00:33:10Z
issue,146129,torch.compile on Mamba2 model produces NaNs,https://github.com/pytorch/pytorch/issues/146129,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dynamo, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-02-24T18:38:46Z
issue,174066,How to prevent operator fusion at specific points in TorchInductor?,https://github.com/pytorch/pytorch/issues/174066,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-02-24T16:09:26Z
issue,170558,[dynamo] Remove all Unsupported subclasses,https://github.com/pytorch/pytorch/issues/170558,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dynamo, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-02-24T04:13:42Z
issue,127882,tensor.size does not accept `dim=None` argument despite documentation because of some confusion with named dimensions,https://github.com/pytorch/pytorch/issues/127882,needs maintainer decision,28,open issue has no stronger automated bucket,"decide whether this is bug, use case, docs, or support",2026-02-24T00:10:37Z
issue,174602,[nondeterminism][inconsistency] ctc_loss sometimes return incorrect results,https://github.com/pytorch/pytorch/issues/174602,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: error checking,"decide whether this is bug, use case, docs, or support",2026-02-23T18:38:08Z
issue,175193,ZLib Reference outdated in riscv ci dockerfile,https://github.com/pytorch/pytorch/issues/175193,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: build, module: infra, module: risc-v","decide whether this is bug, use case, docs, or support",2026-02-23T18:25:01Z
issue,141259,Consider relaxing torch.cond's requirement that the outputs of each branch have matching requires_grad metadata,https://github.com/pytorch/pytorch/issues/141259,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: higher order operators, module: pt2-dispatcher, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-02-23T05:10:33Z
issue,141784,[`FlexAttention`] Request for the Support of Dropout,https://github.com/pytorch/pytorch/issues/141784,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flex attention, module: pt2-dispatcher, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-02-23T04:22:40Z
issue,139518,OpOverloads are slow?,https://github.com/pytorch/pytorch/issues/139518,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: custom-operators, module: pt2-dispatcher, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-02-23T03:35:01Z
issue,139152,Dynamo capture of tensor.data assignment doesn't identical to eager call of tensor.data assignment,https://github.com/pytorch/pytorch/issues/139152,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: aotdispatch, module: pt2-dispatcher, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-02-22T20:51:09Z
issue,138355,tlparse's fx_graph_runnable should include custom op info,https://github.com/pytorch/pytorch/issues/138355,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: custom-operators, module: pt2-dispatcher, oncall: fx, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-02-22T20:31:48Z
issue,138220,Update auto_functionalized_v2 to pass view subgraphs,https://github.com/pytorch/pytorch/issues/138220,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: functionalization, module: pt2-dispatcher, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-02-22T20:25:59Z
issue,153221,FlexAttention: Support for post_mod,https://github.com/pytorch/pytorch/issues/153221,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flex attention, module: higher order operators, module: pt2-dispatcher, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-02-22T20:09:13Z
issue,136272,AOTDispatcher debug mode,https://github.com/pytorch/pytorch/issues/136272,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: aotdispatch, module: pt2-dispatcher, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-02-22T18:33:11Z
issue,136149,Dynamo: warning when we incorrectly inline autograd.Function.backward due to subclass gradients,https://github.com/pytorch/pytorch/issues/136149,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dynamo, module: pt2-dispatcher, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-02-22T18:21:44Z
issue,119589,Free-floating fake tensor without the mode (FakeTensorMode),https://github.com/pytorch/pytorch/issues/119589,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: faketensor, module: pt2-dispatcher, module: serialization, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-02-22T18:17:33Z
issue,115626,Alternatives to compiling a for loop,https://github.com/pytorch/pytorch/issues/115626,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dynamo, module: higher order operators, module: pt2-dispatcher, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-02-22T05:42:08Z
issue,175484,Unable to disable warning about tf32 precision,https://github.com/pytorch/pytorch/issues/175484,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: compile ux, module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-02-22T05:17:11Z
issue,122540,[RFC] Solving Warm Start in torch.compile,https://github.com/pytorch/pytorch/issues/122540,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: aotdispatch, module: dynamo, module: faketensor, module: inductor","decide whether this is bug, use case, docs, or support",2026-02-22T04:56:00Z
issue,132865,aotautograd CSE interferes with Inductor reinplacing,https://github.com/pytorch/pytorch/issues/132865,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: inductor, module: pt2-dispatcher, module: reinplacing, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-02-22T04:53:01Z
issue,174496,Improve performance of torch._foreach_copy_ on CUDA tensors,https://github.com/pytorch/pytorch/issues/174496,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: cuda, module: mta, module: performance, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-02-22T04:26:56Z
issue,164299,MPS Memory Leaks across core MPS files,https://github.com/pytorch/pytorch/issues/164299,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: mps,"decide whether this is bug, use case, docs, or support",2026-02-21T21:13:42Z
issue,175478,torch.distributions.Gamma().sample() should take a torch.Generator,https://github.com/pytorch/pytorch/issues/175478,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: determinism, module: distributions, module: random","decide whether this is bug, use case, docs, or support",2026-02-21T19:40:55Z
issue,89757,third-order gradient of torch.pow with tensor args and certain input returns NaN,https://github.com/pytorch/pytorch/issues/89757,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: autograd, module: nans and infs","decide whether this is bug, use case, docs, or support",2026-02-21T18:00:29Z
issue,129964,run dynamo CI shard with aot_eager_decomp_partition backend to see fallout,https://github.com/pytorch/pytorch/issues/129964,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: aotdispatch, module: pt2-dispatcher, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-02-21T05:24:38Z
issue,117115,request: torch.compile support for tensor subclass subgraph boundaries,https://github.com/pytorch/pytorch/issues/117115,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: __torch_dispatch__, module: pt2-dispatcher, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-02-21T05:07:00Z
issue,123252,FakeTensor support of pin_memory,https://github.com/pytorch/pytorch/issues/123252,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dynamo, module: faketensor, module: pt2-dispatcher, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-02-21T05:02:37Z
issue,125030,"Allow operators with no returns to be ""out="" operations",https://github.com/pytorch/pytorch/issues/125030,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: custom-operators, module: pt2-dispatcher, module: reinplacing, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-02-21T04:53:19Z
issue,151761,Inconsistent `sum`/`dot`/`norm` behavior,https://github.com/pytorch/pytorch/issues/151761,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: linear algebra,"decide whether this is bug, use case, docs, or support",2026-02-21T04:08:27Z
issue,173452,Feature request: Support forward hooks with torch.compile fullgraph=True for capturing intermediate activations,https://github.com/pytorch/pytorch/issues/173452,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dynamo, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-02-21T00:13:40Z
issue,173255,Torch.compile is decreasing training speed instead increasing it,https://github.com/pytorch/pytorch/issues/173255,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dynamo, module: performance, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-02-20T23:43:37Z
issue,173621,Feasibility of Persistent Caching for Dynamo Frontend,https://github.com/pytorch/pytorch/issues/173621,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: compile ux, module: dynamo, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-02-20T23:34:58Z
issue,40770,Incremental version of pca_lowrank,https://github.com/pytorch/pytorch/issues/40770,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: cuda,"decide whether this is bug, use case, docs, or support",2026-02-20T22:13:52Z
issue,173160,Feature Request: load_state_dict should optionally support partial tensor filling in non-strict mode,https://github.com/pytorch/pytorch/issues/173160,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: nn,"decide whether this is bug, use case, docs, or support",2026-02-20T21:03:20Z
issue,103359,Libtorch compile error when defining D_GLIBCXX_DEBUG,https://github.com/pytorch/pytorch/issues/103359,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: abi, module: build","decide whether this is bug, use case, docs, or support",2026-02-20T20:01:18Z
issue,97894,Dropout traces poorly with AotAutograd/make_fx,https://github.com/pytorch/pytorch/issues/97894,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: aotdispatch, module: pt2-dispatcher, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-02-20T20:01:02Z
issue,113914,Register a torch op for `functorch.experimental.control_flow.map` to lower,https://github.com/pytorch/pytorch/issues/113914,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: functorch, module: pt2-dispatcher, module: xla, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-02-20T19:56:55Z
issue,114411,higher order ops + torch.compile + __torch_dispatch__ subclasses,https://github.com/pytorch/pytorch/issues/114411,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: __torch_dispatch__, module: higher order operators, module: pt2-dispatcher, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-02-20T19:53:44Z
issue,114413,Support sending __torch_dispatch__ subclasses in torch.compile that do not desugar,https://github.com/pytorch/pytorch/issues/114413,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: __torch_dispatch__, module: pt2-dispatcher, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-02-20T19:50:04Z
issue,114397,__torch_dispatch__ + compile: run functionalization before the subclass,https://github.com/pytorch/pytorch/issues/114397,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: __torch_dispatch__, module: functionalization, module: pt2-dispatcher, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-02-20T19:48:22Z
issue,115741,torch_function mode explodes with torch.compile,https://github.com/pytorch/pytorch/issues/115741,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: __torch_function__, module: dynamo, module: pt2-dispatcher, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-02-20T19:47:14Z
issue,106732,Hugging Face safetensor does not work with FakeTensorMode,https://github.com/pytorch/pytorch/issues/106732,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: faketensor, module: pt2-dispatcher, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-02-20T19:40:20Z
issue,104146,make_fx: torch.where scalar promotion burns in device,https://github.com/pytorch/pytorch/issues/104146,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: aotdispatch, module: pt2-dispatcher, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-02-20T18:54:38Z
issue,103650,AOT autograd: Avoid dependency on strides for manual regeneration of outputs that are aliased to inputs,https://github.com/pytorch/pytorch/issues/103650,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: aotdispatch, module: pt2-dispatcher, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-02-20T18:52:46Z
issue,114412,torch.compile <> __torch_dispatch__ support tracker issue,https://github.com/pytorch/pytorch/issues/114412,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: __torch_dispatch__, module: aotdispatch, module: pt2-dispatcher, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-02-20T18:39:00Z
issue,109237,"AOTAutograd should track view chains so it can replay them, instead of using as_strided.",https://github.com/pytorch/pytorch/issues/109237,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: aotdispatch, module: pt2-dispatcher, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-02-20T16:34:17Z
issue,130028,Support python attribute mutations of tensor subclass graph inputs from within __torch_dispatch__,https://github.com/pytorch/pytorch/issues/130028,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: aotdispatch, module: pt2-dispatcher, oncall: distributed, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-02-20T16:16:49Z
issue,141349,Burn down test/functorch/test_aotdispatch.py OpInfo test failures,https://github.com/pytorch/pytorch/issues/141349,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: aotdispatch, module: pt2-dispatcher, module: tests, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-02-20T16:01:59Z
issue,175372,python property setter ignored when assigning nn.Parameter() to nn.Module,https://github.com/pytorch/pytorch/issues/175372,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: nn,"decide whether this is bug, use case, docs, or support",2026-02-20T12:49:24Z
issue,23430,[Feature request] Let DistributedSampler take a Sampler as input,https://github.com/pytorch/pytorch/issues/23430,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dataloader, oncall: distributed","decide whether this is bug, use case, docs, or support",2026-02-20T12:18:58Z
issue,170556,[dynamo] Add linter rule to disallow `raise Unsupported` in torch/_dynamo,https://github.com/pytorch/pytorch/issues/170556,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dynamo, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-02-19T10:09:18Z
issue,161757,[RFC] Flash decoding optimization based on FlexAttention on X86 CPU,https://github.com/pytorch/pytorch/issues/161757,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: sdpa,"decide whether this is bug, use case, docs, or support",2026-02-19T08:07:21Z
issue,175306,SAC should treat inductor_compiled_code and other HOPs specially (pull out a more unique name) to prevent mixups,https://github.com/pytorch/pytorch/issues/175306,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: activation checkpointing, module: compile ux, module: higher order operators, module: pt2-dispatcher","decide whether this is bug, use case, docs, or support",2026-02-19T02:22:28Z
issue,129279,[ONNX] Create unit tests for the new export path by adapting all existing tests,https://github.com/pytorch/pytorch/issues/129279,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: onnx,"decide whether this is bug, use case, docs, or support",2026-02-18T20:01:25Z
issue,150296,[RFC] zentorch Integration,https://github.com/pytorch/pytorch/issues/150296,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-02-18T18:45:35Z
issue,174227,[RFC] Support physical memory accounting for VMM-based pluggable allocators,https://github.com/pytorch/pytorch/issues/174227,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: cuda, module: cudacachingallocator, module: memory usage, module: vllm","decide whether this is bug, use case, docs, or support",2026-02-18T16:59:51Z
issue,119930,[dtensor] ops coverage tracker,https://github.com/pytorch/pytorch/issues/119930,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dtensor, oncall: distributed","decide whether this is bug, use case, docs, or support",2026-02-18T00:31:09Z
issue,151782,JVP: Option to Disable Gradient Caching for Tangents,https://github.com/pytorch/pytorch/issues/151782,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: functorch,"decide whether this is bug, use case, docs, or support",2026-02-17T20:49:45Z
issue,136092,User API For Donated Buffers,https://github.com/pytorch/pytorch/issues/136092,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: aotdispatch, module: inductor, module: pt2-dispatcher, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-02-17T19:27:12Z
issue,173802,Batching rule of aten::_weight_norm,https://github.com/pytorch/pytorch/issues/173802,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: batching,"decide whether this is bug, use case, docs, or support",2026-02-16T19:21:44Z
issue,174763,"torch._grouped_mm is not autocast-compatible, causes dtype mismatch",https://github.com/pytorch/pytorch/issues/174763,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: amp (automated mixed precision), module: docs, module: linear algebra","decide whether this is bug, use case, docs, or support",2026-02-16T10:04:43Z
issue,175032,Dead code: reduction of post AI tech debt,https://github.com/pytorch/pytorch/issues/175032,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: nn,"decide whether this is bug, use case, docs, or support",2026-02-14T21:25:51Z
issue,174983,Make `export CUDA_LOG_FILE=stdout` less noisy with spurious errors.,https://github.com/pytorch/pytorch/issues/174983,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: cuda, module: ux","decide whether this is bug, use case, docs, or support",2026-02-13T22:36:47Z
issue,166856,CPU `normal_()` Non-Deterministic Behavior with Non-Contiguous Tensors,https://github.com/pytorch/pytorch/issues/166856,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: determinism, module: random","decide whether this is bug, use case, docs, or support",2026-02-13T22:13:21Z
issue,45728,Low shared memory,https://github.com/pytorch/pytorch/issues/45728,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: dataloader,"decide whether this is bug, use case, docs, or support",2026-02-13T15:55:48Z
issue,174930,UNSTABLE inductor / inductor-test-cuda13 / test (inductor_torchbench),https://github.com/pytorch/pytorch/issues/174930,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: ci,"decide whether this is bug, use case, docs, or support",2026-02-13T00:50:19Z
issue,174929,UNSTABLE inductor / inductor-cpu-test / test (cpu_inductor_torchbench),https://github.com/pytorch/pytorch/issues/174929,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: ci,"decide whether this is bug, use case, docs, or support",2026-02-13T00:49:37Z
issue,171568,[ROCm][Inductor] Flash/efficient attention meta kernel shape mismatch on CDNA (MI200/MI300),https://github.com/pytorch/pytorch/issues/171568,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: rocm,"decide whether this is bug, use case, docs, or support",2026-02-12T23:36:37Z
issue,173904,Unexpected cuda memory usage when doing 4d+ matmul,https://github.com/pytorch/pytorch/issues/173904,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: cuda, module: edge cases, module: linear algebra, module: memory usage","decide whether this is bug, use case, docs, or support",2026-02-12T21:00:20Z
issue,141958,Support SDPA flash attention/ memory efficient attn on ROCm gfx908,https://github.com/pytorch/pytorch/issues/141958,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: rocm, module: sdpa","decide whether this is bug, use case, docs, or support",2026-02-12T00:45:34Z
issue,170758,bf16 layernorm has wrong answer,https://github.com/pytorch/pytorch/issues/170758,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: bfloat16, module: cpu, module: edge cases","decide whether this is bug, use case, docs, or support",2026-02-11T18:43:21Z
issue,174792,Provide a way to systematically run all the autotune configs provided by torch.compile,https://github.com/pytorch/pytorch/issues/174792,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: inductor, module: tests, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-02-11T18:36:26Z
issue,141550,RISC-V CI support,https://github.com/pytorch/pytorch/issues/141550,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: ci, module: risc-v","decide whether this is bug, use case, docs, or support",2026-02-11T10:28:48Z
issue,90574,Bfloat16 tensor .numpy() support,https://github.com/pytorch/pytorch/issues/90574,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: bfloat16, module: numpy","decide whether this is bug, use case, docs, or support",2026-02-11T06:58:23Z
issue,132989,Different random seeds can cause differences in arithmetic precision,https://github.com/pytorch/pytorch/issues/132989,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: numerical-stability,"decide whether this is bug, use case, docs, or support",2026-02-11T01:52:09Z
issue,165628,[Meta] Type Hints in Pytorch,https://github.com/pytorch/pytorch/issues/165628,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: typing,"decide whether this is bug, use case, docs, or support",2026-02-10T19:16:22Z
issue,174313,[release 2.11][triton] trunk / linux-jammy-rocm-py3.10 / test failures,https://github.com/pytorch/pytorch/issues/174313,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: inductor, module: rocm, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-02-10T18:39:17Z
issue,122588,[ONNX] Applying consecutive conv2d layers in a for loop,https://github.com/pytorch/pytorch/issues/122588,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: higher order operators, module: onnx, module: pt2-dispatcher, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-02-10T13:32:02Z
issue,82098,[feature request] Discover actually loaded shared libraries at runtime,https://github.com/pytorch/pytorch/issues/82098,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: build,"decide whether this is bug, use case, docs, or support",2026-02-10T12:03:55Z
issue,135508,"BUG: `torch.cuda.is_available()` returns `False` in certain torch, CUDA and driver version",https://github.com/pytorch/pytorch/issues/135508,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: binaries, module: cuda, module: docker","decide whether this is bug, use case, docs, or support",2026-02-09T19:23:10Z
issue,174359,The OpenReg runtime is stuck,https://github.com/pytorch/pytorch/issues/174359,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: deadlock, module: multithreading, module: openreg","decide whether this is bug, use case, docs, or support",2026-02-09T18:18:09Z
issue,61686,[discussion] Remove the need of mandatory super() nn.Module's call,https://github.com/pytorch/pytorch/issues/61686,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: nn,"decide whether this is bug, use case, docs, or support",2026-02-09T13:58:11Z
issue,174355,Forward AD not implemented for GRU cell when using CUDA,https://github.com/pytorch/pytorch/issues/174355,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: autograd, module: cuda, module: nn","decide whether this is bug, use case, docs, or support",2026-02-09T01:31:34Z
issue,126158,[torch.arange] Output mismatch with Numpy / TensorFlow,https://github.com/pytorch/pytorch/issues/126158,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: numerical-stability,"decide whether this is bug, use case, docs, or support",2026-02-07T15:34:57Z
issue,108521,"resutl of (torch.mm(a,b) does not match result of (a[:part,:], b)",https://github.com/pytorch/pytorch/issues/108521,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: numerical-stability,"decide whether this is bug, use case, docs, or support",2026-02-07T15:13:45Z
issue,139964,Report issue for torch.nn.Linear when forwarding a 3-dim tensor.,https://github.com/pytorch/pytorch/issues/139964,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: numerical-stability,"decide whether this is bug, use case, docs, or support",2026-02-07T15:13:04Z
issue,76528,Discrepancy in einsum when done in batch vs non-batch,https://github.com/pytorch/pytorch/issues/76528,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: linear algebra, module: tf32","decide whether this is bug, use case, docs, or support",2026-02-07T07:59:57Z
issue,76632,RestrictPtrTraits in CUDA potentially has no effect.,https://github.com/pytorch/pytorch/issues/76632,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: cuda,"decide whether this is bug, use case, docs, or support",2026-02-07T07:57:09Z
issue,171398,ISAAC inference,https://github.com/pytorch/pytorch/issues/171398,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: cuda, module: rocm","decide whether this is bug, use case, docs, or support",2026-02-05T23:27:40Z
issue,124275,Performance Impact of `torch.autograd.profiler.record_function` in DataLoader's `_BaseDataLoaderIter` Iteration,https://github.com/pytorch/pytorch/issues/124275,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dataloader, module: performance","decide whether this is bug, use case, docs, or support",2026-02-05T20:55:13Z
issue,143887,`torch.accelerator` cross-device utilities and properties,https://github.com/pytorch/pytorch/issues/143887,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: accelerator,"decide whether this is bug, use case, docs, or support",2026-02-05T20:11:09Z
issue,159520,[RFC] Enable native Windows CI/CD on ROCm,https://github.com/pytorch/pytorch/issues/159520,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: ci, module: rocm","decide whether this is bug, use case, docs, or support",2026-02-05T17:44:34Z
issue,174266,BatchNorm1d CPU outputs differ across devices/threads with identical inputs,https://github.com/pytorch/pytorch/issues/174266,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: cpu, module: edge cases, module: nn","decide whether this is bug, use case, docs, or support",2026-02-05T00:26:46Z
issue,172320,"[OpenReg] Extend device infrastructure tests (device, event, streams)",https://github.com/pytorch/pytorch/issues/172320,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: openreg,"decide whether this is bug, use case, docs, or support",2026-02-04T15:14:00Z
issue,150974,feature: tlparse summary,https://github.com/pytorch/pytorch/issues/150974,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: logging, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-02-04T12:56:50Z
issue,159516,Validate PyPI Wheel Binary Size failing for >3months,https://github.com/pytorch/pytorch/issues/159516,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: oncall: releng,"decide whether this is bug, use case, docs, or support",2026-02-04T04:31:01Z
issue,174220,Dr CI produces confusing issue attribution report,https://github.com/pytorch/pytorch/issues/174220,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: ci, module: devx","decide whether this is bug, use case, docs, or support",2026-02-03T18:36:52Z
issue,119461,Promote pytorch nightly docker image,https://github.com/pytorch/pytorch/issues/119461,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: binaries, module: docker","decide whether this is bug, use case, docs, or support",2026-02-03T16:12:29Z
issue,79197,[feature request] Support dataclass derivations of nn.Module,https://github.com/pytorch/pytorch/issues/79197,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: nn,"decide whether this is bug, use case, docs, or support",2026-02-03T15:26:07Z
issue,109204,Pytorch ROCM windows builds,https://github.com/pytorch/pytorch/issues/109204,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: rocm, module: windows","decide whether this is bug, use case, docs, or support",2026-02-03T14:49:49Z
issue,137574,[feature request] Provide FlexAttention as a new available/selectable backend for SDPA,https://github.com/pytorch/pytorch/issues/137574,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: sdpa,"decide whether this is bug, use case, docs, or support",2026-02-03T13:35:39Z
issue,119196,Possible NCCL race condition `destroy_process_group` followed by `init_process_group`.,https://github.com/pytorch/pytorch/issues/119196,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: nccl, oncall: distributed","decide whether this is bug, use case, docs, or support",2026-02-03T01:04:49Z
issue,154444,[RFC] Removing ideep git submodule dependency from PyTorch & add oneDNN git submodule,https://github.com/pytorch/pytorch/issues/154444,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: arm, module: intel, module: third_party","decide whether this is bug, use case, docs, or support",2026-02-03T00:11:59Z
issue,174081,"[vllm] [release 2.10] V1 e2e + engine test_ngram_and_suffix_correctness[speculative_config1], test_draft_model_realistic_example",https://github.com/pytorch/pytorch/issues/174081,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: inductor, module: vllm, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-02-02T19:21:45Z
issue,174088,Gold doesn't work with GB200,https://github.com/pytorch/pytorch/issues/174088,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: arm, module: build, module: cuda","decide whether this is bug, use case, docs, or support",2026-02-02T17:38:51Z
issue,163408,RFC: Inherit some modules from torchtune.,https://github.com/pytorch/pytorch/issues/163408,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: nn,"decide whether this is bug, use case, docs, or support",2026-02-02T09:28:12Z
issue,133559,Support for variable number of dimensions in functorch.dim dims() and autonaming induced inconsistencies,https://github.com/pytorch/pytorch/issues/133559,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: first class dims, module: functorch","decide whether this is bug, use case, docs, or support",2026-02-01T16:20:34Z
issue,137093,Use SymInt instead of sympy.Expr uniformly in Inductor,https://github.com/pytorch/pytorch/issues/137093,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dynamic shapes, module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-01-31T18:23:40Z
issue,125254,AssertionError: Device mps not supported,https://github.com/pytorch/pytorch/issues/125254,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: inductor, module: mps, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-01-30T23:32:56Z
issue,171558,`torch.sparse.spdiags` produces incorrect results,https://github.com/pytorch/pytorch/issues/171558,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: sparse,"decide whether this is bug, use case, docs, or support",2026-01-30T21:54:46Z
issue,173950,[CI] enabling CUDA 13.0 periodic test,https://github.com/pytorch/pytorch/issues/173950,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: ci,"decide whether this is bug, use case, docs, or support",2026-01-30T20:30:22Z
issue,163327,[MPS] scatter_add_ is incorrect on MPS with non-zero offset slices,https://github.com/pytorch/pytorch/issues/163327,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: correctness (silent), module: mps","decide whether this is bug, use case, docs, or support",2026-01-30T15:50:21Z
issue,173638,`torch.linalg.slogdet` does not propagate `NaN` in CUDA,https://github.com/pytorch/pytorch/issues/173638,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: correctness (silent), module: cuda, module: linear algebra, module: nans and infs","decide whether this is bug, use case, docs, or support",2026-01-29T22:32:11Z
issue,168393,"`register_autograd` broken for `torch.func`, fails to register `setup_context`",https://github.com/pytorch/pytorch/issues/168393,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: functorch, module: vmap","decide whether this is bug, use case, docs, or support",2026-01-29T21:24:42Z
issue,173796,`torch.nn.functional.PoissonNLLLoss` returns `NaN` imaginary part on CPU (vs 0 on GPU) for overflowing complex inputs,https://github.com/pytorch/pytorch/issues/173796,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: nans and infs, module: nn","decide whether this is bug, use case, docs, or support",2026-01-29T20:53:16Z
issue,173661,[Perfomance issue] Dataloader has unexpected overhead,https://github.com/pytorch/pytorch/issues/173661,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dataloader, module: performance","decide whether this is bug, use case, docs, or support",2026-01-29T17:41:00Z
issue,34242,RuntimeError with torch.unique: radix_sort: failed on 2nd step: invalid argument,https://github.com/pytorch/pytorch/issues/34242,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: cuda, module: sorting and selection","decide whether this is bug, use case, docs, or support",2026-01-29T09:17:46Z
issue,133702,"Add `beta`, `betainc`, etc. to `torch.special`",https://github.com/pytorch/pytorch/issues/133702,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: distributions,"decide whether this is bug, use case, docs, or support",2026-01-29T05:41:00Z
issue,173642,[Feature Request][torch.compile] Improve UX when autotuning repeated layer,https://github.com/pytorch/pytorch/issues/173642,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: compile ux, module: dynamo, module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-01-29T00:51:49Z
issue,173729,Way to request a non-aliased torch.cuda.Stream,https://github.com/pytorch/pytorch/issues/173729,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: cuda,"decide whether this is bug, use case, docs, or support",2026-01-28T23:01:06Z
issue,173514,Improve documentation of assumptions for symmetric memory,https://github.com/pytorch/pytorch/issues/173514,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: docs, module: symm_mem","decide whether this is bug, use case, docs, or support",2026-01-28T22:37:30Z
issue,112398,[Tracker] Move nested tensors to beta,https://github.com/pytorch/pytorch/issues/112398,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: nestedtensor,"decide whether this is bug, use case, docs, or support",2026-01-28T22:28:59Z
issue,108794,[C++ Frontend] Simple Changes for Cleaner Options,https://github.com/pytorch/pytorch/issues/108794,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: cpp,"decide whether this is bug, use case, docs, or support",2026-01-28T14:54:34Z
issue,173516,[CUDA] MaxPool2d CUDA kernel lacks int64 support → one-side error,https://github.com/pytorch/pytorch/issues/173516,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: cuda,"decide whether this is bug, use case, docs, or support",2026-01-27T23:37:05Z
issue,76344,"[ONNX] Use topk to export max(dim,keepdim) to onnx",https://github.com/pytorch/pytorch/issues/76344,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: onnx,"decide whether this is bug, use case, docs, or support",2026-01-27T16:21:25Z
issue,170419,Cannot set version_counter for inference tensor,https://github.com/pytorch/pytorch/issues/170419,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: python frontend,"decide whether this is bug, use case, docs, or support",2026-01-27T13:35:05Z
issue,95487,Torch RPC on multiple nodes with GPU returns a EOF error,https://github.com/pytorch/pytorch/issues/95487,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: rpc, module: tensorpipe, oncall: distributed","decide whether this is bug, use case, docs, or support",2026-01-27T13:06:21Z
issue,171375,aten.lift triggers internal assert failure in Inductor functionalization pass,https://github.com/pytorch/pytorch/issues/171375,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: functionalization, module: pt2-dispatcher, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-01-27T09:23:46Z
issue,173104,Crash with `CUDACachingAllocator.cpp` when running Nvidia `nemotron-parse` at load,https://github.com/pytorch/pytorch/issues/173104,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: cudacachingallocator,"decide whether this is bug, use case, docs, or support",2026-01-26T16:33:25Z
issue,173290,Potential issue with lt_two in DistanceKernel.cu,https://github.com/pytorch/pytorch/issues/173290,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: cuda,"decide whether this is bug, use case, docs, or support",2026-01-26T16:30:30Z
issue,172886,libtorch_cpu.so have unresolved symbols,https://github.com/pytorch/pytorch/issues/172886,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: build, module: mpi","decide whether this is bug, use case, docs, or support",2026-01-25T19:02:42Z
issue,141288,Enhancement Request: improve this error message,https://github.com/pytorch/pytorch/issues/141288,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: nn,"decide whether this is bug, use case, docs, or support",2026-01-24T20:42:54Z
issue,133367,Adafactor foreach impl performance tracker,https://github.com/pytorch/pytorch/issues/133367,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: mta, module: optimizer, module: performance","decide whether this is bug, use case, docs, or support",2026-01-23T17:00:57Z
issue,173059,pandas/numpy binary incompatibility breaks test_dataloader and test_datapipe,https://github.com/pytorch/pytorch/issues/173059,needs maintainer decision,28,open issue has no stronger automated bucket,"decide whether this is bug, use case, docs, or support",2026-01-22T23:25:56Z
issue,119328,[pytree] Handling of `None` in torch.utils._pytree is inconsistent with JAX.,https://github.com/pytorch/pytorch/issues/119328,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: pytree,"decide whether this is bug, use case, docs, or support",2026-01-22T21:28:30Z
issue,171283,aten.uniform cannot be called with keyword argument from,https://github.com/pytorch/pytorch/issues/171283,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: distributions, module: docs, module: python frontend","decide whether this is bug, use case, docs, or support",2026-01-22T04:09:22Z
issue,172854,[test] Add unittest for nanmean output dtype validation (follow-up to #131043),https://github.com/pytorch/pytorch/issues/172854,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: reductions, module: tests","decide whether this is bug, use case, docs, or support",2026-01-21T23:01:54Z
issue,151705,Enable TorchInductor to Generate Matmuls Natively via `tl.dot`,https://github.com/pytorch/pytorch/issues/151705,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-01-21T19:59:07Z
issue,172797,autograd.Function.apply yields surprising strides,https://github.com/pytorch/pytorch/issues/172797,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: autograd,"decide whether this is bug, use case, docs, or support",2026-01-21T18:51:06Z
issue,85366,More windows for filtering and spectral analysis,https://github.com/pytorch/pytorch/issues/85366,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: scipy compatibility,"decide whether this is bug, use case, docs, or support",2026-01-21T17:33:29Z
issue,159444,#define CHECK causes issues with a bunch of other libraries,https://github.com/pytorch/pytorch/issues/159444,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: build, module: cpp","decide whether this is bug, use case, docs, or support",2026-01-21T16:18:46Z
issue,92250,torch.cuda.is_available() returns True even if the CUDA hardware can't run pytorch,https://github.com/pytorch/pytorch/issues/92250,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: cuda,"decide whether this is bug, use case, docs, or support",2026-01-21T08:50:37Z
issue,171804,[On Demand Profiling] Early CUDA initialization caused by `import torch` when `KINETO_USE_DAEMON=1`,https://github.com/pytorch/pytorch/issues/171804,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: profiler","decide whether this is bug, use case, docs, or support",2026-01-21T03:42:29Z
issue,162780,Prebuilt Wheels supporting T4G,https://github.com/pytorch/pytorch/issues/162780,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: arm, module: binaries, module: cuda","decide whether this is bug, use case, docs, or support",2026-01-20T23:12:34Z
issue,147662,GPUDirect Storage dcp.save/dcp.load torch.save/torch.load,https://github.com/pytorch/pytorch/issues/147662,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: serialization, oncall: distributed checkpointing","decide whether this is bug, use case, docs, or support",2026-01-20T18:28:08Z
issue,74282,RuntimeError: cuDNN error: CUDNN_STATUS_VERSION_MISMATCH for torchvision models,https://github.com/pytorch/pytorch/issues/74282,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: binaries, module: cudnn, module: vision","decide whether this is bug, use case, docs, or support",2026-01-20T18:27:16Z
issue,168573,TestModule: test/inductor/test_codecache.py::TestFxGraphCache,https://github.com/pytorch/pytorch/issues/168573,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: inductor, module: rocm, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-01-20T15:22:03Z
issue,144634,[MPS] `torch.mps.synchronize` hangs on error,https://github.com/pytorch/pytorch/issues/144634,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: deadlock, module: mps, module: third_party","decide whether this is bug, use case, docs, or support",2026-01-20T15:19:44Z
issue,68110,zeta(1/2) returns nan,https://github.com/pytorch/pytorch/issues/68110,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: special,"decide whether this is bug, use case, docs, or support",2026-01-20T13:56:44Z
issue,172803,[OpenReg] PrivateUse1 backend missing distributed backend registration causes KeyError,https://github.com/pytorch/pytorch/issues/172803,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: openreg, oncall: distributed","decide whether this is bug, use case, docs, or support",2026-01-20T09:50:34Z
issue,140458,Making nn.Module generic,https://github.com/pytorch/pytorch/issues/140458,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: nn, module: typing","decide whether this is bug, use case, docs, or support",2026-01-19T14:34:25Z
issue,170743,Reference count modifications led to performance drop,https://github.com/pytorch/pytorch/issues/170743,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: cpu, module: performance","decide whether this is bug, use case, docs, or support",2026-01-19T02:51:00Z
issue,76580,Random Generator for Dropout,https://github.com/pytorch/pytorch/issues/76580,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: random,"decide whether this is bug, use case, docs, or support",2026-01-19T01:30:06Z
issue,149605,torch.export does not support nested tensor operations,https://github.com/pytorch/pytorch/issues/149605,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: onnx, oncall: export, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-01-18T22:39:19Z
issue,172623,Nested Tensors should have their own type,https://github.com/pytorch/pytorch/issues/172623,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: nestedtensor, module: typing","decide whether this is bug, use case, docs, or support",2026-01-18T08:17:35Z
issue,172475,[Inductor] Track deterministic `alloc_id` assignment across cached compilation artifacts,https://github.com/pytorch/pytorch/issues/172475,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: determinism, module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-01-18T04:02:06Z
issue,172213,Eager buffer mutation after `torch.compile` + backward + `torch.export.export` hits `_map_assigned_buffer_to_proxy` `FakeTensor` assertion,https://github.com/pytorch/pytorch/issues/172213,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: aotdispatch, module: pt2-dispatcher, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-01-18T04:01:24Z
issue,172632,AArch64 GCC14 - inductor/test_mkldnn_pattern_matcher.py FAIL - internal compiler error: Segmentation fault,https://github.com/pytorch/pytorch/issues/172632,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: arm, module: build, module: third_party","decide whether this is bug, use case, docs, or support",2026-01-16T18:51:37Z
issue,172445,AArch64 GCC14 Issues Tracker,https://github.com/pytorch/pytorch/issues/172445,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: arm, module: build, module: third_party","decide whether this is bug, use case, docs, or support",2026-01-16T12:38:15Z
issue,164281,Document Symmetric Memory,https://github.com/pytorch/pytorch/issues/164281,needs maintainer decision,28,open issue has no stronger automated bucket,"decide whether this is bug, use case, docs, or support",2026-01-16T05:41:21Z
issue,172351,[RFC][Release-2.11] Dropping Volta support from CUDA-12.8,https://github.com/pytorch/pytorch/issues/172351,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: binaries, module: cuda, module: cudnn","decide whether this is bug, use case, docs, or support",2026-01-15T23:44:05Z
issue,172549,"AOTAutograd error when compiled graph has two DTensor outputs that alias each other, one differentiable one not",https://github.com/pytorch/pytorch/issues/172549,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: pt2-dispatcher,"decide whether this is bug, use case, docs, or support",2026-01-15T18:52:50Z
issue,169672,torch while_loop does not avoid CPU synchronization,https://github.com/pytorch/pytorch/issues/169672,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: higher order operators, module: pt2-dispatcher, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-01-15T18:18:52Z
issue,99082,Look into test coverage for `UntypedStorage`,https://github.com/pytorch/pytorch/issues/99082,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: tests, module: typing","decide whether this is bug, use case, docs, or support",2026-01-15T15:41:48Z
issue,165398,Skipping combo kernel for large pointwise nodes,https://github.com/pytorch/pytorch/issues/165398,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: inductor, module: vllm, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-01-15T15:12:19Z
issue,97892,A parameterized fill value for triu and tril functions,https://github.com/pytorch/pytorch/issues/97892,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: viewing and reshaping,"decide whether this is bug, use case, docs, or support",2026-01-15T06:38:24Z
issue,172522,Support out-of-graph mutations with side streams,https://github.com/pytorch/pytorch/issues/172522,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: oncall: pt2,"decide whether this is bug, use case, docs, or support",2026-01-15T01:05:18Z
issue,167881,[RFC] Modifying Getting Started Page for XPU support,https://github.com/pytorch/pytorch/issues/167881,needs maintainer decision,28,open issue has no stronger automated bucket,"decide whether this is bug, use case, docs, or support",2026-01-14T21:32:13Z
issue,172495,Add T5Gemma2 to TorchInductor benchmark HF suite,https://github.com/pytorch/pytorch/issues/172495,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: benchmark, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-01-14T21:13:59Z
issue,136532,Wrong results when sampling from the beta distribution for small alpha=beta,https://github.com/pytorch/pytorch/issues/136532,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: distributions, module: numerical-stability","decide whether this is bug, use case, docs, or support",2026-01-14T19:53:08Z
issue,146780,TypeError when using torch.compile with RegionViT under torch.inference_mode(),https://github.com/pytorch/pytorch/issues/146780,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-01-14T19:52:53Z
issue,172427,[CI] Permission denied while trying to connect to the docker API,https://github.com/pytorch/pytorch/issues/172427,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: ci, module: infra","decide whether this is bug, use case, docs, or support",2026-01-14T16:16:54Z
issue,172273,Numeric difference between div and foreach_div,https://github.com/pytorch/pytorch/issues/172273,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: mta,"decide whether this is bug, use case, docs, or support",2026-01-14T15:01:12Z
issue,172338,Fix: Support is_causal=True with need_weights=True in Multi-Head Attention,https://github.com/pytorch/pytorch/issues/172338,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: nn,"decide whether this is bug, use case, docs, or support",2026-01-14T14:55:14Z
issue,172383,torch.func.linearize produces NaNs with autocast + LayerNorm (jvp works),https://github.com/pytorch/pytorch/issues/172383,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: amp (automated mixed precision), module: functorch","decide whether this is bug, use case, docs, or support",2026-01-14T14:51:58Z
issue,171376,PyTorch ThroughputBenchmark.add_input crashes when split_sizes is incompatible with input tensor shape,https://github.com/pytorch/pytorch/issues/171376,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: benchmark,"decide whether this is bug, use case, docs, or support",2026-01-14T08:00:27Z
issue,38035,Quantile Regression Loss,https://github.com/pytorch/pytorch/issues/38035,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: nn,"decide whether this is bug, use case, docs, or support",2026-01-14T03:06:31Z
issue,171759,[ONNX] Create a tutorial,https://github.com/pytorch/pytorch/issues/171759,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: onnx,"decide whether this is bug, use case, docs, or support",2026-01-13T17:36:32Z
issue,172322,[OpenReg] Extend test_rng.py and test_misc.py coverage,https://github.com/pytorch/pytorch/issues/172322,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: openreg,"decide whether this is bug, use case, docs, or support",2026-01-13T15:08:06Z
issue,172319,[OpenReg] Add OpInfo-based operator tests using op_db,https://github.com/pytorch/pytorch/issues/172319,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: openreg,"decide whether this is bug, use case, docs, or support",2026-01-13T15:07:44Z
issue,172318,[OpenReg] Add DeviceTypeTestBase example with instantiate_device_type_tests,https://github.com/pytorch/pytorch/issues/172318,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: openreg,"decide whether this is bug, use case, docs, or support",2026-01-13T15:07:28Z
issue,169597,Standardize Testing in OpenReg,https://github.com/pytorch/pytorch/issues/169597,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: openreg, module: privateuse1","decide whether this is bug, use case, docs, or support",2026-01-13T08:44:19Z
issue,172059,Add Tweedie distribution to torch.distributions,https://github.com/pytorch/pytorch/issues/172059,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: distributions,"decide whether this is bug, use case, docs, or support",2026-01-12T22:30:18Z
issue,143471,NFS errors during DataLoader shutdown when num_workers > 1 when temporary directory is on NFS,https://github.com/pytorch/pytorch/issues/143471,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: data,"decide whether this is bug, use case, docs, or support",2026-01-12T21:42:08Z
issue,171194,CUDA Smoke tests: Extend smoke tests to include profiler coverage,https://github.com/pytorch/pytorch/issues/171194,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: binaries, module: ci, module: cuda","decide whether this is bug, use case, docs, or support",2026-01-12T20:07:09Z
issue,17023,Download speed issues with the pytorch conda channel,https://github.com/pytorch/pytorch/issues/17023,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: dependency bug,"decide whether this is bug, use case, docs, or support",2026-01-12T14:09:06Z
issue,171501,Several Windows-related GitHub Actions not running — are they intentionally disabled?,https://github.com/pytorch/pytorch/issues/171501,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: arm, module: ci, module: windows","decide whether this is bug, use case, docs, or support",2026-01-12T07:14:05Z
issue,164455,AMP + Torch Compile bug,https://github.com/pytorch/pytorch/issues/164455,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: amp (automated mixed precision), module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-01-11T18:29:11Z
issue,171866,Why are there three run_repro functions implemented in pytorch?,https://github.com/pytorch/pytorch/issues/171866,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: docs, module: minifier, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-01-10T19:34:32Z
issue,167062,How to use torch.compile on Windows GPU?,https://github.com/pytorch/pytorch/issues/167062,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: windows, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-01-10T18:07:57Z
issue,171796,Exception in subprocess causes infinite hang and zombie processes,https://github.com/pytorch/pytorch/issues/171796,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: multiprocessing, oncall: distributed","decide whether this is bug, use case, docs, or support",2026-01-10T11:50:51Z
pr,178513,Fix interpolate nearest mode with non-integer scale_factor,https://github.com/pytorch/pytorch/pull/178513,has linked issue,27,PR closes or references open issue numbers; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor,review when ready,2026-05-24T17:45:53Z
pr,181073,Validate norm_type in lp_pool1d/2d/3d,https://github.com/pytorch/pytorch/pull/181073,has linked issue,27,PR closes or references open issue numbers; PR has not updated for at least 45 days,review when ready,2026-05-20T21:44:52Z
pr,176893,Add save_state_dict and load_state_dict functions (fixes #150857),https://github.com/pytorch/pytorch/pull/176893,has linked issue,27,PR closes or references open issue numbers; PR has not updated for at least 45 days,review when ready,2026-05-20T13:58:06Z
pr,175537,Fix: Exclude third-party headers from PyTorch wheels,https://github.com/pytorch/pytorch/pull/175537,has linked issue,27,PR closes or references open issue numbers; PR has not updated for at least 45 days,review when ready,2026-05-20T13:56:32Z
pr,174882,Fix torch.histc float16 precision inconsistency,https://github.com/pytorch/pytorch/pull/174882,has linked issue,27,PR closes or references open issue numbers; PR has not updated for at least 45 days,review when ready,2026-05-20T13:55:21Z
pr,179936,[MPS] Fix avg_pool2d precision by using Metal kernel instead of MPSGraph,https://github.com/pytorch/pytorch/pull/179936,has linked issue,27,PR closes or references open issue numbers; PR has not updated for at least 45 days; CI lane labels: ciflow/mps,review when ready,2026-05-20T08:15:02Z
pr,184349,Improve complex input error for BatchNorm,https://github.com/pytorch/pytorch/pull/184349,has linked issue,27,PR closes or references open issue numbers; PR has not updated for at least 45 days,review when ready,2026-05-19T17:21:33Z
pr,184058,nn.functional: raise ValueError for tau <= 0 in gumbel_softmax,https://github.com/pytorch/pytorch/pull/184058,has linked issue,27,PR closes or references open issue numbers; PR has not updated for at least 45 days,review when ready,2026-05-18T21:44:52Z
pr,183522,Fix using `torch.cuda.amp. autocast` with JIT on Python 3.13,https://github.com/pytorch/pytorch/pull/183522,has linked issue,27,PR closes or references open issue numbers; PR has not updated for at least 45 days,review when ready,2026-05-18T21:32:58Z
pr,180420,linalg.qr: reject overlapping or aliased Q and R in out=,https://github.com/pytorch/pytorch/pull/180420,has linked issue,27,PR closes or references open issue numbers; PR has not updated for at least 45 days,review when ready,2026-05-18T17:34:49Z
pr,182348,[SymmMem] Implement StoreExchange::broadcast for multicast handle exchange,https://github.com/pytorch/pytorch/pull/182348,has linked issue,27,PR closes or references open issue numbers; PR has not updated for at least 45 days; CI lane labels: ciflow/h100-symm-mem,review when ready,2026-05-15T07:08:50Z
pr,183367,Avoid initializing CUDA in mem_get_info,https://github.com/pytorch/pytorch/pull/183367,has linked issue,27,PR closes or references open issue numbers; PR has not updated for at least 45 days,review when ready,2026-05-14T16:06:37Z
pr,182803,Fix torch.trapz documentation signature,https://github.com/pytorch/pytorch/pull/182803,has linked issue,27,PR closes or references open issue numbers; PR has not updated for at least 45 days,review when ready,2026-05-14T16:03:57Z
pr,182609,Fix inconsistent CPU/CUDA casting for fmax/fmin/maximum/minimum with narrower out dtype,https://github.com/pytorch/pytorch/pull/182609,has linked issue,27,PR closes or references open issue numbers; PR has not updated for at least 45 days,review when ready,2026-05-14T16:02:23Z
pr,180748,Fix NumPy 2 deprecation warning in __array_wrap__,https://github.com/pytorch/pytorch/pull/180748,has linked issue,27,PR closes or references open issue numbers; PR has not updated for at least 45 days,review when ready,2026-05-13T18:32:12Z
pr,183052,Fix OOM error suggesting expandable_segments when already enabled,https://github.com/pytorch/pytorch/pull/183052,has linked issue,27,PR closes or references open issue numbers; PR has not updated for at least 45 days,review when ready,2026-05-12T05:42:23Z
pr,181052,Add docs for torch version information,https://github.com/pytorch/pytorch/pull/181052,has linked issue,27,PR closes or references open issue numbers; PR has not updated for at least 45 days,review when ready,2026-05-11T21:39:44Z
pr,181367,[CPU] Fix fp16 matmul ~100x slower than fp32 on CPUs without AVX512-FP16,https://github.com/pytorch/pytorch/pull/181367,has linked issue,27,PR closes or references open issue numbers; PR has not updated for at least 45 days,review when ready,2026-05-11T20:54:39Z
pr,183039,[c10d] Remove false-positive DeviceMesh pickle warning in async checkpoint,https://github.com/pytorch/pytorch/pull/183039,has linked issue,27,PR closes or references open issue numbers; PR has not updated for at least 45 days; CI lane labels: ciflow/torchtitan,review when ready,2026-05-11T18:56:36Z
pr,159862,Fix skipIfXpu and skipIfHpu and similar skip decorators,https://github.com/pytorch/pytorch/pull/159862,has linked issue,27,PR closes or references open issue numbers; PR has not updated for at least 45 days,review when ready,2026-05-11T11:21:05Z
pr,167306,Fix integer overflow in (fake) quantization,https://github.com/pytorch/pytorch/pull/167306,has linked issue,27,PR closes or references open issue numbers; PR has not updated for at least 45 days,review when ready,2026-05-11T08:45:02Z
pr,152828,[MSVC] Remove redundant MEM_EFF_ATTENTION_CUDA_SOURCES definition as /Zc:lambda is set globally for C++20 by default,https://github.com/pytorch/pytorch/pull/152828,has linked issue,27,PR closes or references open issue numbers; PR has not updated for at least 45 days,review when ready,2026-05-10T18:50:12Z
pr,173644,[TORCH] [JIT] Change emitDefautlArgsWithOutArgs -> emitDefaultArgsWithOutArgs in export_module.cpp,https://github.com/pytorch/pytorch/pull/173644,has linked issue,27,PR closes or references open issue numbers; PR has not updated for at least 45 days,review when ready,2026-04-12T17:26:12Z
pr,127702,Define PY_SSIZE_T_CLEAN before #include <Python.h>,https://github.com/pytorch/pytorch/pull/127702,has linked issue,27,PR closes or references open issue numbers; PR has not updated for at least 45 days,review when ready,2026-04-07T17:02:40Z
pr,154395,Updated padding validation in max_pool functions to account for dilation,https://github.com/pytorch/pytorch/pull/154395,has linked issue,27,PR closes or references open issue numbers; PR has not updated for at least 45 days,review when ready,2025-12-05T16:55:38Z
pr,162130,Raise error when no record on `extra_files`,https://github.com/pytorch/pytorch/pull/162130,has linked issue,27,PR closes or references open issue numbers; PR has not updated for at least 45 days,review when ready,2025-11-17T15:00:50Z
pr,152104,Update _torch_docs.py to Fix torch.bernoulli(),https://github.com/pytorch/pytorch/pull/152104,has linked issue,27,PR closes or references open issue numbers; PR has not updated for at least 45 days,review when ready,2025-09-26T04:01:01Z
pr,142160,Introduce int32 index_fill and index_copy indices,https://github.com/pytorch/pytorch/pull/142160,has linked issue,27,PR closes or references open issue numbers; PR has not updated for at least 45 days,review when ready,2025-09-04T19:12:52Z
pr,102148,Add mps support for maxpool3d,https://github.com/pytorch/pytorch/pull/102148,has linked issue,27,PR closes or references open issue numbers; PR has not updated for at least 45 days; CI lane labels: ciflow/mps,review when ready,2025-07-30T18:42:18Z
pr,147365,Replace `fw_metadata` info with trace log hint in hint message,https://github.com/pytorch/pytorch/pull/147365,has linked issue,27,PR closes or references open issue numbers; PR has not updated for at least 45 days,review when ready,2025-07-28T17:34:49Z
pr,88998,Fix fake tensor propagation for nvprims,https://github.com/pytorch/pytorch/pull/88998,has linked issue,27,PR closes or references open issue numbers; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor,review when ready,2023-04-17T00:36:00Z
issue,158007,"`torch.compile` cannot compile a model with a basic `LSTM`, even on latest main",https://github.com/pytorch/pytorch/issues/158007,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, module: rnn, oncall: pt2",review linked PR state before touching issue,2026-06-15T17:05:44Z
issue,185650,[Bug] torch.kron fails with RuntimeError on non-contiguous inputs,https://github.com/pytorch/pytorch/issues/185650,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: linear algebra, module: viewing and reshaping",review linked PR state before touching issue,2026-06-06T16:22:24Z
issue,137440,"`torch.pow` throws with `RuntimeError: ""reciprocal_cuda"" not implemented for 'Char'`",https://github.com/pytorch/pytorch/issues/137440,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: core aten, module: cuda",review linked PR state before touching issue,2026-06-01T06:48:43Z
issue,184280,GroupNorm with num_groups=0 raises ZeroDivisionError instead of ValueError,https://github.com/pytorch/pytorch/issues/184280,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: error checking, module: nn, module: norms and normalization",review linked PR state before touching issue,2026-05-22T15:56:07Z
issue,177827,Internal Assert Failure in `torch.cuda.memory.caching_allocator_alloc` with negative allocation size,https://github.com/pytorch/pytorch/issues/177827,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: error checking",review linked PR state before touching issue,2026-03-23T17:31:25Z
issue,127111,[dynamo] Detect mkldnn_max_pool2d/mkldnn view errors at tracing time,https://github.com/pytorch/pytorch/issues/127111,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, module: faketensor, oncall: pt2",review linked PR state before touching issue,2026-03-11T21:57:17Z
issue,140171,Compiling `as_strided` call on a view input errors.,https://github.com/pytorch/pytorch/issues/140171,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: faketensor, module: pt2-dispatcher, oncall: pt2",review linked PR state before touching issue,2026-02-23T04:02:32Z
issue,171981,torch.fx.Tracer.trace fails on wrapped forward methods with arbitrary arguments. ValueError: code: co_varnames is too small,https://github.com/pytorch/pytorch/issues/171981,has linked PR,26,open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: fx,review linked PR state before touching issue,2026-01-08T17:09:11Z
issue,120375,Dynamo doesn't generate resume calls after graph breaking on log calls,https://github.com/pytorch/pytorch/issues/120375,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",review linked PR state before touching issue,2026-01-07T08:15:39Z
issue,158384,[dynamo] full internal stack trace being shown with `torch._dynamo.exc.BackendCompilerFailed` with suggestion to show more,https://github.com/pytorch/pytorch/issues/158384,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: compile ux, module: dynamo, oncall: pt2",review linked PR state before touching issue,2026-01-06T19:02:18Z
issue,171356,`torch.clip` has checks for float16 scalar overflow on CPU but not on GPU,https://github.com/pytorch/pytorch/issues/171356,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: edge cases, module: error checking",review linked PR state before touching issue,2026-01-05T17:23:02Z
issue,170781,meta_model.to_empty(...) equivalent for FakeTensor,https://github.com/pytorch/pytorch/issues/170781,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: faketensor, module: pt2-dispatcher, oncall: pt2",review linked PR state before touching issue,2025-12-29T05:02:38Z
issue,168974,Fuzzy mode for DebugMode.check_hash_mismatches,https://github.com/pytorch/pytorch/issues/168974,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: pt2-dispatcher, oncall: pt2",review linked PR state before touching issue,2025-12-28T22:07:02Z
issue,170770,FakeTensorMode induced Cannot swap t1 because it has weakref associated with it,https://github.com/pytorch/pytorch/issues/170770,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: faketensor, module: pt2-dispatcher, oncall: pt2",review linked PR state before touching issue,2025-12-28T04:48:47Z
issue,137068,UserWarning: Graph break due to unsupported builtin time.time,https://github.com/pytorch/pytorch/issues/137068,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",review linked PR state before touching issue,2025-12-28T04:34:13Z
issue,170834,Generated torch.Library autograd functions seem not have `setup_context` which breaks `torch.func.grad` usage,https://github.com/pytorch/pytorch/issues/170834,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: custom-operators, module: pt2-dispatcher, oncall: pt2",review linked PR state before touching issue,2025-12-23T16:07:09Z
issue,169769,pytorch while_loop / cond overly strict about memory aliasing,https://github.com/pytorch/pytorch/issues/169769,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: higher order operators, module: pt2-dispatcher, oncall: pt2",review linked PR state before touching issue,2025-12-23T12:59:16Z
issue,170955,`grouped_mm` does grouping on CPU on 5060 Ti (Windows),https://github.com/pytorch/pytorch/issues/170955,has linked PR,26,open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: cuda,review linked PR state before touching issue,2025-12-22T18:17:05Z
issue,170370,[MPS][CUDA] EmbeddingBag on CUDA/MPS does not check offsets[0] like CPU,https://github.com/pytorch/pytorch/issues/170370,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: cuda, module: docs, module: edge cases",review linked PR state before touching issue,2025-12-20T00:52:46Z
issue,169779,[Inductor] Inconsistent handling of negative indices in index_select: CPU/CUDA/MPS,https://github.com/pytorch/pytorch/issues/169779,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: decompositions, module: python frontend, oncall: pt2",review linked PR state before touching issue,2025-12-19T11:21:35Z
issue,32864,[docs] Missing docs for torch.__version__ and torch.version,https://github.com/pytorch/pytorch/issues/32864,has linked PR,26,open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: docs,review linked PR state before touching issue,2025-12-18T19:49:28Z
issue,170298,"`resize_` updates tensor shape metadata even when storage resize fails, creating corrupted ""Zombie"" tensors",https://github.com/pytorch/pytorch/issues/170298,has linked PR,26,open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: python frontend,review linked PR state before touching issue,2025-12-18T15:38:37Z
issue,136586,support FakeTensor input for torch.compile,https://github.com/pytorch/pytorch/issues/136586,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, module: faketensor, module: pt2-dispatcher, oncall: pt2",review linked PR state before touching issue,2025-12-18T15:12:09Z
issue,141790,AOTAutograd cache always misses even on simple program,https://github.com/pytorch/pytorch/issues/141790,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: aotdispatch, oncall: pt2",review linked PR state before touching issue,2025-12-09T18:57:48Z
issue,160739,ConvTranspose 3D with BF16 or FP16 types is not supported on MPS,https://github.com/pytorch/pytorch/issues/160739,has linked PR,26,open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: mps,review linked PR state before touching issue,2025-12-08T06:13:37Z
issue,169099,[RFC] Implement StoreExchange.broadcast,https://github.com/pytorch/pytorch/issues/169099,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: symm_mem, oncall: distributed",review linked PR state before touching issue,2025-12-05T07:03:40Z
issue,130182,autograd.Function HOP backward input names could be better,https://github.com/pytorch/pytorch/issues/130182,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: higher order operators, oncall: pt2",review linked PR state before touching issue,2025-12-05T01:39:54Z
issue,165385,[dynamo] Invalid setattr's are not handled properly in general,https://github.com/pytorch/pytorch/issues/165385,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",review linked PR state before touching issue,2025-12-03T01:53:30Z
issue,168307,torch.compile error when creating nested tensors dynamically in forward pass,https://github.com/pytorch/pytorch/issues/168307,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, module: nestedtensor, oncall: pt2",review linked PR state before touching issue,2025-12-02T23:57:02Z
issue,162415,[dynamo] Weird interaction between fullgraph=True and skipped frames,https://github.com/pytorch/pytorch/issues/162415,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: compile ux, module: dynamo, oncall: pt2",review linked PR state before touching issue,2025-12-02T23:48:04Z
issue,169208,Segmentation fault in torch.native_batch_norm with empty running_var tensor,https://github.com/pytorch/pytorch/issues/169208,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: empty tensor, module: error checking, module: norms and normalization",review linked PR state before touching issue,2025-12-02T14:42:24Z
issue,169150,[debug mode] do not hash collectives,https://github.com/pytorch/pytorch/issues/169150,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: pt2-dispatcher, oncall: pt2",review linked PR state before touching issue,2025-12-01T22:47:49Z
issue,135061,[export] non-strict export error on torch.distributions.Normal,https://github.com/pytorch/pytorch/issues/135061,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamic shapes, oncall: export, oncall: pt2",review linked PR state before touching issue,2025-12-01T19:36:49Z
issue,168975,No way to dump DebugMode hashes to file for further processing / diffing,https://github.com/pytorch/pytorch/issues/168975,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: pt2-dispatcher, oncall: pt2",review linked PR state before touching issue,2025-12-01T18:22:16Z
issue,169236,"[MPS] nn.ConvTranspose{1,2,3}d ignore validation check for output_padding < stride (Silent Failure)",https://github.com/pytorch/pytorch/issues/169236,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: error checking, module: mps",review linked PR state before touching issue,2025-11-30T03:51:39Z
issue,136642,Linespace does not support SymInt step. (but int),https://github.com/pytorch/pytorch/issues/136642,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamic shapes, module: faketensor, module: pt2-dispatcher, oncall: export",review linked PR state before touching issue,2025-11-26T16:12:21Z
issue,168566,Test: AOTInductorTestsTemplate.test_autotune_int64_user_defined_triton_kernel,https://github.com/pytorch/pytorch/issues/168566,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: aotinductor, module: rocm, oncall: export, oncall: pt2",review linked PR state before touching issue,2025-11-25T19:46:07Z
issue,168872,Test: TestSDPACudaOnly.test_cudnn_attention_broken_166211,https://github.com/pytorch/pytorch/issues/168872,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: rocm, module: sdpa",review linked PR state before touching issue,2025-11-25T19:13:07Z
issue,168868,Test: TestTorchDeviceType.test_pdist_norm_large,https://github.com/pytorch/pytorch/issues/168868,has linked PR,26,open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: rocm,review linked PR state before touching issue,2025-11-25T00:51:52Z
issue,168136,tracing torch _ops.py like viztracer and cpython settrace will case stuck in an infinite loop.,https://github.com/pytorch/pytorch/issues/168136,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: library, module: pt2-dispatcher, oncall: pt2",review linked PR state before touching issue,2025-11-24T14:17:52Z
issue,150319,[dynamo] Bad error message when trying to compile async functions due to bad bytecode jump targets,https://github.com/pytorch/pytorch/issues/150319,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: compile ux, module: dynamo, oncall: pt2",review linked PR state before touching issue,2025-11-21T20:58:40Z
issue,167596,[dynamo][feature] Guard on constants only if graph is specialized and not bytecode,https://github.com/pytorch/pytorch/issues/167596,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",review linked PR state before touching issue,2025-11-20T17:44:45Z
issue,167906,Avoid Exception Refcycle Problems,https://github.com/pytorch/pytorch/issues/167906,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: memory usage, module: python frontend",review linked PR state before touching issue,2025-11-18T22:15:10Z
issue,139092,`tensor` not a `FakeTensor` under `FakeTensorMode` and `device('meta')`,https://github.com/pytorch/pytorch/issues/139092,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: faketensor, module: pt2-dispatcher, oncall: pt2",review linked PR state before touching issue,2025-11-18T20:11:28Z
issue,167719,KeyError when exporting PyTorch model with dynamic string keys in ModuleDict using torch.export,https://github.com/pytorch/pytorch/issues/167719,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: oncall: export, oncall: pt2",review linked PR state before touching issue,2025-11-18T19:13:04Z
issue,129327,[export/dynamo] torch._check fails at compile time when the condition evaluates to False,https://github.com/pytorch/pytorch/issues/129327,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",review linked PR state before touching issue,2025-11-17T21:14:35Z
issue,167732,Dynamo should support backward() capture,https://github.com/pytorch/pytorch/issues/167732,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",review linked PR state before touching issue,2025-11-17T21:10:06Z
issue,167191,[dynamo] [dynamic] `nn.LazyLinear` reject SymInt when direct using `torch.compile()`,https://github.com/pytorch/pytorch/issues/167191,has linked PR,26,open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: oncall: pt2,review linked PR state before touching issue,2025-11-17T07:16:46Z
issue,158975,Handle matmuls gracefully in foreach_map,https://github.com/pytorch/pytorch/issues/158975,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: higher order operators, module: pt2-dispatcher, oncall: pt2",review linked PR state before touching issue,2025-11-15T04:47:06Z
issue,126024,Can not use torch.compile with dynamic=Ture when using multi-threads,https://github.com/pytorch/pytorch/issues/126024,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamic shapes, module: multithreading, oncall: pt2",review linked PR state before touching issue,2025-11-13T06:41:30Z
issue,167026,Support torch.cuda.use_mem_pool in torch.compile,https://github.com/pytorch/pytorch/issues/167026,has linked PR,26,open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: oncall: pt2,review linked PR state before touching issue,2025-11-11T18:40:53Z
issue,167117,`fx.GraphModule` serialization fails when type annotations are `str`s,https://github.com/pytorch/pytorch/issues/167117,has linked PR,26,open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: oncall: pt2,review linked PR state before touching issue,2025-11-11T18:19:48Z
issue,131794,[torch.compile] Unused differentiable outputs lead to unnecessary gradient computation when compared to eager,https://github.com/pytorch/pytorch/issues/131794,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: aotdispatch, module: pt2-dispatcher, oncall: pt2",review linked PR state before touching issue,2025-11-10T19:31:21Z
issue,123663,Improve Foreach OpInfo test infra to enable meta coverage,https://github.com/pytorch/pytorch/issues/123663,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: mta, module: testing, oncall: pt2",review linked PR state before touching issue,2025-11-07T18:00:34Z
issue,122200,"Dynamo compilation should always run with empty functorch interpreter stack, even if ambient context has functorch",https://github.com/pytorch/pytorch/issues/122200,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, module: functorch, module: pt2-dispatcher, oncall: pt2",review linked PR state before touching issue,2025-11-07T15:53:41Z
issue,128554,aot_export_module produces unserializable GraphModules,https://github.com/pytorch/pytorch/issues/128554,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: oncall: export, oncall: pt2",review linked PR state before touching issue,2025-11-06T22:18:19Z
issue,139077,Add support for binding tensor kwargs in max-autotune templates,https://github.com/pytorch/pytorch/issues/139077,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",review linked PR state before touching issue,2025-11-06T17:59:03Z
issue,137520,Support torch.cond with differing (but same dimensionality) sizes on true/false branches,https://github.com/pytorch/pytorch/issues/137520,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamic shapes, oncall: pt2",review linked PR state before touching issue,2025-11-06T02:45:10Z
issue,157926,torch.compile() disables torch.distributions parameter validation globally,https://github.com/pytorch/pytorch/issues/157926,has linked PR,26,open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: oncall: pt2,review linked PR state before touching issue,2025-11-05T22:09:23Z
issue,129845,vmap + autograd.Function [generate_vmap_rule=True] + torch.compile don't work together,https://github.com/pytorch/pytorch/issues/129845,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, module: pt2-dispatcher, oncall: pt2",review linked PR state before touching issue,2025-11-05T21:06:24Z
issue,128830,TORCHDYNAMO_REPRO_AFTER=aot produces invalid repro code,https://github.com/pytorch/pytorch/issues/128830,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: minifier, oncall: pt2",review linked PR state before touching issue,2025-11-05T18:58:55Z
issue,137875,TORCH_LOGS=+inductor spams too much information about cache,https://github.com/pytorch/pytorch/issues/137875,has linked PR,26,open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: oncall: pt2,review linked PR state before touching issue,2025-11-05T18:53:19Z
issue,166516,Performance issue of torch._higher_order_ops.scan,https://github.com/pytorch/pytorch/issues/166516,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: higher order operators, module: pt2-dispatcher, oncall: pt2",review linked PR state before touching issue,2025-11-04T21:32:28Z
issue,166905,Type Annotation for `named_modules()`,https://github.com/pytorch/pytorch/issues/166905,has linked PR,26,open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: typing,review linked PR state before touching issue,2025-11-04T19:25:57Z
issue,166525,Inductor Fails with `AttributeError: 'int' object has no attribute 'meta'` for `torch.while_loop` Containing Scalar Values in Loop State,https://github.com/pytorch/pytorch/issues/166525,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: higher order operators, module: pt2-dispatcher, oncall: pt2",review linked PR state before touching issue,2025-11-04T10:06:01Z
issue,161650,config.activation_memory_budget,https://github.com/pytorch/pytorch/issues/161650,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: functorch, oncall: pt2",review linked PR state before touching issue,2025-10-26T15:11:45Z
issue,133592,[torchbind x compile] Can't register a torchbind operator that mutates a tensor,https://github.com/pytorch/pytorch/issues/133592,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: pt2-dispatcher, module: torchbind, module: vllm, oncall: pt2",review linked PR state before touching issue,2025-10-24T16:39:56Z
issue,136404,Profiling Torch.compile (with_stack=True) leads to frozen program with massive memory.,https://github.com/pytorch/pytorch/issues/136404,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: vllm, oncall: profiler, oncall: pt2",review linked PR state before touching issue,2025-10-24T16:39:00Z
issue,152548,FakeTensorUpdater does not trace nodes correctly,https://github.com/pytorch/pytorch/issues/152548,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: inductor, module: vllm, oncall: pt2",review linked PR state before touching issue,2025-10-24T16:38:43Z
issue,137869,TORCH_COMPILE_CPROFILE does not work for python 3.12,https://github.com/pytorch/pytorch/issues/137869,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: vllm, oncall: pt2",review linked PR state before touching issue,2025-10-24T16:38:29Z
issue,165547,error message for deepcopy a cuda tensor under FakeTensorMode is pretty cryptic,https://github.com/pytorch/pytorch/issues/165547,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: faketensor, module: pt2-dispatcher, module: vllm, oncall: pt2",review linked PR state before touching issue,2025-10-24T16:38:12Z
issue,161119,[UED][Falcon-7B x vLLM] excessive shape recompilations (static specialization),https://github.com/pytorch/pytorch/issues/161119,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamic shapes, module: dynamo, module: vllm, oncall: pt2",review linked PR state before touching issue,2025-10-24T16:38:04Z
issue,161111,[UED][Falcon-7B x vLLM] errors tracing causal_conv_1d triton kernel grid fn,https://github.com/pytorch/pytorch/issues/161111,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, module: vllm, oncall: pt2",review linked PR state before touching issue,2025-10-24T16:38:02Z
issue,157776,Support for _Float16/C++23 std::float16_t,https://github.com/pytorch/pytorch/issues/157776,has linked PR,26,open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: cpp,review linked PR state before touching issue,2025-10-23T18:34:00Z
issue,165981,Fails to use side_effect op inside torch.cond branch,https://github.com/pytorch/pytorch/issues/165981,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: higher order operators, module: pt2-dispatcher, oncall: pt2",review linked PR state before touching issue,2025-10-22T23:47:49Z
issue,166009,[Inductor] DCE breaks with fallback operators that produce aliases + mutation,https://github.com/pytorch/pytorch/issues/166009,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",review linked PR state before touching issue,2025-10-21T18:45:41Z
issue,127153,"[ONNX] pad_sequence() is not exportable, with neither legacy onnx.export nor with dynamo_export",https://github.com/pytorch/pytorch/issues/127153,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: onnx, oncall: export, oncall: pt2",review linked PR state before touching issue,2025-10-18T14:10:49Z
issue,129418,functionalization should run underneath python decomps,https://github.com/pytorch/pytorch/issues/129418,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: functionalization, module: pt2-dispatcher, oncall: export, oncall: pt2",review linked PR state before touching issue,2025-10-16T22:17:40Z
issue,165097,Distributed checkpoint should not expect private use 1 to be cuda compatible,https://github.com/pytorch/pytorch/issues/165097,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: privateuse1, oncall: distributed, oncall: distributed checkpointing",review linked PR state before touching issue,2025-10-16T11:57:37Z
issue,137388,some guards are not shown in the code_parts,https://github.com/pytorch/pytorch/issues/137388,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, module: guards, oncall: pt2",review linked PR state before touching issue,2025-10-15T18:55:55Z
issue,93367,Aot accuracy minifier with dynamic shapes doesn't work,https://github.com/pytorch/pytorch/issues/93367,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: minifier, oncall: pt2",review linked PR state before touching issue,2025-10-15T18:53:02Z
issue,106135,Registering function that takes const std::vector<c10::SymInt>& to SymInt[] schema gives confusing error message,https://github.com/pytorch/pytorch/issues/106135,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: custom-operators, module: pt2-dispatcher, oncall: pt2",review linked PR state before touching issue,2025-10-15T18:37:18Z
issue,164922,`torch.compile` fails to trace `datetime.now()` with Dynamo guard check failure,https://github.com/pytorch/pytorch/issues/164922,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",review linked PR state before touching issue,2025-10-14T20:25:33Z
issue,150613,"Torch compile for `torch.searchsorted` failed when capturing scalar outputs, with scalar `values` taken from `sorted_sequence`",https://github.com/pytorch/pytorch/issues/150613,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: decompositions, module: dynamic shapes, oncall: pt2",review linked PR state before touching issue,2025-10-14T03:09:18Z
issue,164883,Wheels include protobuf headers (and headers for other projects),https://github.com/pytorch/pytorch/issues/164883,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: cpp",review linked PR state before touching issue,2025-10-13T17:19:00Z
issue,122773,torch.gather - Cannot call numel() on tensor with symbolic sizes/strides,https://github.com/pytorch/pytorch/issues/122773,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamic shapes, oncall: pt2",review linked PR state before touching issue,2025-10-10T16:02:55Z
issue,111385,Torch Compile Dynamic fails on sample on diffusers VAE,https://github.com/pytorch/pytorch/issues/111385,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",review linked PR state before touching issue,2025-10-10T05:45:19Z
issue,164559,fwd_rng_state show up in the aot_export_joint grpah input,https://github.com/pytorch/pytorch/issues/164559,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: aotdispatch, module: pt2-dispatcher, oncall: pt2",review linked PR state before touching issue,2025-10-06T19:10:58Z
issue,136628,[torch.export.load] failed while executing `pow_by_natural`,https://github.com/pytorch/pytorch/issues/136628,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: oncall: export, oncall: pt2",review linked PR state before touching issue,2025-10-05T22:48:02Z
issue,127163,BUG: `torch.pow(-0.0)` wrongly returns `-0.0` (should be +0.0),https://github.com/pytorch/pytorch/issues/127163,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: python frontend",review linked PR state before touching issue,2025-10-03T09:02:04Z
issue,164478,Out of date references in CONTRIBUTING.md and Wiki,https://github.com/pytorch/pytorch/issues/164478,has linked PR,26,open PR references this issue; old update age or low-urgency/stale label,review linked PR state before touching issue,2025-10-02T18:22:17Z
issue,133655,PythonMod vs Mod spookiness,https://github.com/pytorch/pytorch/issues/133655,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamic shapes, oncall: pt2",review linked PR state before touching issue,2025-10-01T03:25:38Z
issue,164124,"torch.compile compiles multiple Triton autotune kernels, but uses the wrong ones",https://github.com/pytorch/pytorch/issues/164124,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamic shapes, module: user triton, oncall: pt2",review linked PR state before touching issue,2025-09-29T16:50:17Z
issue,160388,Perf differences observed between AOT and JIT on RAFT model,https://github.com/pytorch/pytorch/issues/160388,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",review linked PR state before touching issue,2025-09-24T22:20:39Z
issue,163143,[torch.compile] Side-Effectful Tokens doesn't work well with torch.cond,https://github.com/pytorch/pytorch/issues/163143,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: higher order operators, module: pt2-dispatcher, oncall: pt2",review linked PR state before touching issue,2025-09-24T17:41:47Z
issue,163641,u0==u1 but u0//u1==1 throw DDE,https://github.com/pytorch/pytorch/issues/163641,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamic shapes, oncall: pt2",review linked PR state before touching issue,2025-09-23T19:14:12Z
issue,163037,[compile/inductor][functionalization] `prims.broadcast_in_dim` triggers alias-annotation error under `torch.compile` (eager OK),https://github.com/pytorch/pytorch/issues/163037,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: aotdispatch, module: pt2-dispatcher, oncall: pt2",review linked PR state before touching issue,2025-09-20T16:27:34Z
issue,162980,Adding custom function to _allowed_determinism_checks_to_fns from model code,https://github.com/pytorch/pytorch/issues/162980,has linked PR,26,open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: activation checkpointing,review linked PR state before touching issue,2025-09-19T21:10:55Z
issue,163040,[torch.export.load] Passing an `ExportedProgram` raises `AttributeError: 'ExportedProgram' object has no attribute 'tell'` instead of a clear `TypeError`,https://github.com/pytorch/pytorch/issues/163040,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: oncall: export, oncall: pt2",review linked PR state before touching issue,2025-09-18T16:29:49Z
issue,160016,Requesting improved torch.auto_grad.detect_anomaly() for NaN detection,https://github.com/pytorch/pytorch/issues/160016,has linked PR,26,open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: autograd,review linked PR state before touching issue,2025-09-18T15:02:47Z
issue,152178,Raise an Error when File Not Found in `torch.jit.load()`,https://github.com/pytorch/pytorch/issues/152178,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: error checking, oncall: jit",review linked PR state before touching issue,2025-09-18T14:26:46Z
issue,162973,[dynamo] GB_REGISTRY linter taking a long time,https://github.com/pytorch/pytorch/issues/162973,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",review linked PR state before touching issue,2025-09-15T20:08:39Z
issue,162687,custom op with torch.compile backward error when op has default arguments,https://github.com/pytorch/pytorch/issues/162687,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: custom-operators, module: pt2-dispatcher, oncall: pt2",review linked PR state before touching issue,2025-09-15T18:48:09Z
issue,162151,`permute` + `add` + `transpose` plays wrong on tensor slice,https://github.com/pytorch/pytorch/issues/162151,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: pt2-dispatcher, oncall: pt2",review linked PR state before touching issue,2025-09-15T17:43:18Z
issue,132923,"If you compile a function/module, we should definitely attempt to compile it (don't try to skip it)",https://github.com/pytorch/pytorch/issues/132923,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: compile ux, module: dynamo, oncall: pt2",review linked PR state before touching issue,2025-09-12T21:20:36Z
issue,142358,`torch.compiler.disable()` on module hooks will disable `module.compile()`,https://github.com/pytorch/pytorch/issues/142358,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: compile ux, module: dynamo, module: nn, oncall: pt2",review linked PR state before touching issue,2025-09-12T21:15:15Z
issue,143308,"[dynamo, guards] Implement FrameLocalsMapping version of check_verbose_nopybind",https://github.com/pytorch/pytorch/issues/143308,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, module: guards, oncall: pt2",review linked PR state before touching issue,2025-09-12T21:14:27Z
issue,144360,Skip empty frames recursively when top-level is empty,https://github.com/pytorch/pytorch/issues/144360,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: compile ux, module: dynamo, oncall: pt2",review linked PR state before touching issue,2025-09-12T21:13:43Z
issue,149565,Should (eventually) be recommending nonstrict_trace instead of allow_in_graph in error messages:,https://github.com/pytorch/pytorch/issues/149565,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: compile ux, module: dynamo, oncall: pt2",review linked PR state before touching issue,2025-09-12T21:07:57Z
issue,144820,[dynamo] Do not always skip code objects unconditionally,https://github.com/pytorch/pytorch/issues/144820,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: compile ux, module: dynamo, oncall: pt2",review linked PR state before touching issue,2025-09-12T21:07:06Z
issue,150915,`module.compile()` behaves differently from `torch.compile(module)`,https://github.com/pytorch/pytorch/issues/150915,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: compile ux, module: dynamo, oncall: pt2",review linked PR state before touching issue,2025-09-12T21:06:40Z
issue,161796,[dynamo] Graph break logspam,https://github.com/pytorch/pytorch/issues/161796,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: compile ux, module: dynamo, module: logging, oncall: pt2",review linked PR state before touching issue,2025-09-12T19:57:00Z
issue,161705,`aten.slice` adds guard on backed symint which causes inconsistent behavior between eager and trace modes,https://github.com/pytorch/pytorch/issues/161705,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamic shapes, oncall: export, oncall: pt2",review linked PR state before touching issue,2025-09-11T03:42:39Z
issue,162287,using torch.sym_not cause Z3Exception: Symbolic expressions cannot be cast to concrete Boolean values.,https://github.com/pytorch/pytorch/issues/162287,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamic shapes, oncall: pt2",review linked PR state before touching issue,2025-09-10T06:56:17Z
issue,162350,Inplace modification error with softmax when combining forward and backward AD,https://github.com/pytorch/pytorch/issues/162350,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: forward ad",review linked PR state before touching issue,2025-09-08T23:29:26Z
issue,161764,"The ""max-autotune"" mode in torch.compile can lead to longer execution times relative to eager execution.",https://github.com/pytorch/pytorch/issues/161764,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: performance, oncall: pt2",review linked PR state before touching issue,2025-09-08T17:49:39Z
issue,162373,Pylint E1102: torch.fft.rfft/istft/irfft and more is not callable,https://github.com/pytorch/pytorch/issues/162373,has linked PR,26,open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: typing,review linked PR state before touching issue,2025-09-08T15:50:36Z
issue,161456,[dynamo][hop] speculate_subgraph doesn't work with subclass inputs,https://github.com/pytorch/pytorch/issues/161456,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, module: higher order operators, module: pt2-dispatcher, oncall: pt2",review linked PR state before touching issue,2025-09-05T19:26:36Z
issue,155238,Dynamo Fails on pad_packed_sequence Due to Fake Tensor Allocation Issue During FX Graph Tracing,https://github.com/pytorch/pytorch/issues/155238,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, module: pt2-dispatcher, module: rnn, oncall: pt2",review linked PR state before touching issue,2025-08-30T18:12:42Z
issue,161088,torch.where should throw an exception when invoked with non-bool tensor,https://github.com/pytorch/pytorch/issues/161088,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: oncall: export, oncall: pt2",review linked PR state before touching issue,2025-08-26T18:12:22Z
issue,132635,[Dynamo] support logging via reorderable_logging_functions,https://github.com/pytorch/pytorch/issues/132635,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",review linked PR state before touching issue,2025-08-25T23:51:15Z
issue,161067,torch.compile not eliminating common quantization nodes,https://github.com/pytorch/pytorch/issues/161067,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: aotdispatch, module: pt2-dispatcher, oncall: pt2",review linked PR state before touching issue,2025-08-25T17:59:40Z
issue,160757,Graph break when torch.compiling an autograd.Function,https://github.com/pytorch/pytorch/issues/160757,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",review linked PR state before touching issue,2025-08-23T16:47:47Z
issue,161138,Fuse 3 elementwise ops and reduction into 1 hbm reads,https://github.com/pytorch/pytorch/issues/161138,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",review linked PR state before touching issue,2025-08-23T13:17:21Z
issue,161289,automatically updating ci benchmark expected csv fails,https://github.com/pytorch/pytorch/issues/161289,has linked PR,26,open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: oncall: pt2,review linked PR state before touching issue,2025-08-23T13:10:57Z
issue,129319,`int` type for `dims` of `tile()` without `dims=` works with a tensor against the doc,https://github.com/pytorch/pytorch/issues/129319,has linked PR,26,open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: docs,review linked PR state before touching issue,2025-08-21T12:50:54Z
issue,117016,There is a performance drop because we have not yet implemented the batching rule for aten::_scaled_dot_product_efficient_attention_backward,https://github.com/pytorch/pytorch/issues/117016,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: sdpa, module: vmap",review linked PR state before touching issue,2025-08-21T05:42:19Z
issue,159843,[DTensor] nn.Embedding Compile Failure when Creating FX Graph,https://github.com/pytorch/pytorch/issues/159843,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dtensor, module: dynamo, oncall: pt2",review linked PR state before touching issue,2025-08-18T22:22:46Z
issue,155421,Backward fails with compiled attention on nested tensors on RTX6000,https://github.com/pytorch/pytorch/issues/155421,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: nestedtensor, oncall: pt2",review linked PR state before touching issue,2025-08-18T17:16:29Z
issue,160544,Error with torch.ops.higher_order.scan + torch.compile [KeyError: combine_fn],https://github.com/pytorch/pytorch/issues/160544,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, module: higher order operators, module: pt2-dispatcher, oncall: pt2",review linked PR state before touching issue,2025-08-15T22:48:59Z
issue,156075,`torch.logsumexp`: support `dim=None`,https://github.com/pytorch/pytorch/issues/156075,has linked PR,26,open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: python frontend,review linked PR state before touching issue,2025-08-13T22:50:29Z
issue,160247,wan2.1 vae take more gpu memory after compile,https://github.com/pytorch/pytorch/issues/160247,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, module: memory usage, oncall: pt2",review linked PR state before touching issue,2025-08-11T20:06:56Z
issue,160083,compile regression on forward pre hook with prepend=True,https://github.com/pytorch/pytorch/issues/160083,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",review linked PR state before touching issue,2025-08-11T18:25:29Z
issue,160271,Error when using torch.compile with dynamic=True: tensor 'guidance' stride mismatch at index 0,https://github.com/pytorch/pytorch/issues/160271,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamic shapes, oncall: pt2",review linked PR state before touching issue,2025-08-11T17:20:32Z
issue,159831,ModuleDict subscription no longer works after compile().,https://github.com/pytorch/pytorch/issues/159831,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",review linked PR state before touching issue,2025-08-05T21:24:40Z
issue,159550,_adaptive_avg_pool2d does not support dynamic shapes？,https://github.com/pytorch/pytorch/issues/159550,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamic shapes, oncall: pt2",review linked PR state before touching issue,2025-08-04T20:14:01Z
issue,129648,`torch.compile` mode and eager mode are different in generating tensor API behavior,https://github.com/pytorch/pytorch/issues/129648,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",review linked PR state before touching issue,2025-08-04T18:58:24Z
issue,128531,test/dynamo/test_torchrec.py doesn't run in OSS CI (and won't run locally),https://github.com/pytorch/pytorch/issues/128531,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",review linked PR state before touching issue,2025-08-04T18:35:30Z
issue,159594,"During CUDA graph capture, CUDACachingAllocator could potentially trigger cudaFree()",https://github.com/pytorch/pytorch/issues/159594,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: cuda graphs, module: cudacachingallocator",review linked PR state before touching issue,2025-08-04T17:37:41Z
issue,158970,Add option to assert if kernel is not fully fused in foreach_map,https://github.com/pytorch/pytorch/issues/158970,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: higher order operators, module: pt2-dispatcher, oncall: pt2",review linked PR state before touching issue,2025-08-04T17:06:47Z
issue,159613,[Torch FX] subgraph_rewriter.replace_pattern does not handle back-to-back matches correctly.,https://github.com/pytorch/pytorch/issues/159613,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: oncall: export, oncall: pt2",review linked PR state before touching issue,2025-08-04T16:19:01Z
issue,75862,LayerNorm and GroupNorm with num_groups=1 not equivalent,https://github.com/pytorch/pytorch/issues/75862,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: nn, module: norms and normalization",review linked PR state before touching issue,2025-08-03T04:11:10Z
issue,124747,TestAOTAutograd.test_mem_leak_from_save_for_bw fails locally but not on CI when run with dynamo,https://github.com/pytorch/pytorch/issues/124747,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: ci, module: dynamo, oncall: pt2",review linked PR state before touching issue,2025-08-01T23:33:40Z
issue,126834,[dynamo] Check for invalid random_ arguments while tracing,https://github.com/pytorch/pytorch/issues/126834,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",review linked PR state before touching issue,2025-08-01T23:14:07Z
issue,159457,`@assume_constant_result` Causes `NameError` When Method Contains Tensor Operations and Dynamic Attribute Assignment During Compilation,https://github.com/pytorch/pytorch/issues/159457,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",review linked PR state before touching issue,2025-07-30T17:27:45Z
issue,93437,Accuracy minifier can find spurious accuracy failures involving uninitialized memory,https://github.com/pytorch/pytorch/issues/93437,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: minifier, oncall: pt2",review linked PR state before touching issue,2025-07-29T14:19:18Z
issue,118739,"Getting ""RuntimeError: Trying to backward through the graph a second time"" when calling backward on compiled unbind / split ops",https://github.com/pytorch/pytorch/issues/118739,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: aotdispatch, module: pt2-dispatcher, oncall: pt2",review linked PR state before touching issue,2025-07-29T14:18:43Z
issue,106704,Dataloader extremely slow on in-memory datasets,https://github.com/pytorch/pytorch/issues/106704,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: data, module: dataloader",review linked PR state before touching issue,2025-07-29T10:57:00Z
issue,159259,Use aten/src/ATen/native/native_functions.yaml to automatically generate tests for type hints,https://github.com/pytorch/pytorch/issues/159259,has linked PR,26,open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: typing,review linked PR state before touching issue,2025-07-28T22:02:33Z
issue,159192,Invalid meta kernel for _fused_moving_avg_obs_fq_helper_functional,https://github.com/pytorch/pytorch/issues/159192,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",review linked PR state before touching issue,2025-07-28T14:31:39Z
issue,159160,"[dynamo, test] Logging error spam at the end of pytest",https://github.com/pytorch/pytorch/issues/159160,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",review linked PR state before touching issue,2025-07-25T15:20:36Z
issue,158561,`torch.compile` In-place operation raises Autograd error even when gradients are not required,https://github.com/pytorch/pytorch/issues/158561,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: pt2-dispatcher, oncall: pt2",review linked PR state before touching issue,2025-07-22T14:44:13Z
issue,158521,[static cuda launcher] scalar float support,https://github.com/pytorch/pytorch/issues/158521,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",review linked PR state before touching issue,2025-07-22T14:36:30Z
issue,154282,LazyGraphModule causes graph breaks,https://github.com/pytorch/pytorch/issues/154282,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",review linked PR state before touching issue,2025-07-21T19:05:28Z
issue,134784,Dynamo x autograd.Function x funtorch transform: intermediate sliced tensor doesn't require grad caused incorrect result,https://github.com/pytorch/pytorch/issues/134784,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, module: functorch, oncall: pt2",review linked PR state before touching issue,2025-07-17T23:36:14Z
issue,117584,[pt2d] module register_pre_forward_hook and register_forward_hook triggered graph break when it's root module,https://github.com/pytorch/pytorch/issues/117584,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: distributed, oncall: pt2",review linked PR state before touching issue,2025-07-17T22:34:37Z
issue,105768,torch.compile uses more memory when using less parameters,https://github.com/pytorch/pytorch/issues/105768,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: memory usage, oncall: pt2",review linked PR state before touching issue,2025-07-17T15:57:13Z
issue,141005,OptimizedModule _parameters attr getting and setting are not working as expected,https://github.com/pytorch/pytorch/issues/141005,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",review linked PR state before touching issue,2025-07-16T23:19:26Z
issue,138961,CompiledFxGraph.current_callable is not thread-safe,https://github.com/pytorch/pytorch/issues/138961,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: multithreading, oncall: pt2",review linked PR state before touching issue,2025-07-15T18:32:51Z
issue,158088,`torch.compile` doesn't properly raise eager fake tensor exception,https://github.com/pytorch/pytorch/issues/158088,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, module: faketensor, module: pt2-dispatcher, oncall: pt2",review linked PR state before touching issue,2025-07-15T17:51:31Z
issue,158087,`torch.compile` ignores nan check in `int(torch.tensor(torch.nan))`,https://github.com/pytorch/pytorch/issues/158087,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, module: nans and infs, oncall: pt2",review linked PR state before touching issue,2025-07-14T20:19:34Z
issue,39383,Pickling a Tensor or a Storage is not deterministic,https://github.com/pytorch/pytorch/issues/39383,has linked PR,26,open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: pickle,review linked PR state before touching issue,2025-07-13T23:37:04Z
issue,122512,Higher peak memory with torch.compile,https://github.com/pytorch/pytorch/issues/122512,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: aotdispatch, module: pt2-dispatcher, oncall: pt2",review linked PR state before touching issue,2025-07-11T18:31:58Z
issue,114296,Minifier doesn't work with dynamic shapes,https://github.com/pytorch/pytorch/issues/114296,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamic shapes, module: minifier, oncall: pt2",review linked PR state before touching issue,2025-07-11T18:17:52Z
issue,125077,[Inductor] Generate triton block pointers for discontiguous strided tensors,https://github.com/pytorch/pytorch/issues/125077,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",review linked PR state before touching issue,2025-07-11T18:04:25Z
issue,122840,a log_softmax kernel get much worse perf with padding,https://github.com/pytorch/pytorch/issues/122840,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",review linked PR state before touching issue,2025-07-11T17:35:22Z
issue,123470,torch.compile x custom ops: op that accepts float also accepts Tensor in eager-mode,https://github.com/pytorch/pytorch/issues/123470,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",review linked PR state before touching issue,2025-07-10T23:58:22Z
issue,125438,Fix broken linalg unittests on ARM platform,https://github.com/pytorch/pytorch/issues/125438,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: arm, module: linear algebra, module: tests",review linked PR state before touching issue,2025-07-10T16:37:06Z
issue,157832,[FSDP2] should pass args as is instead of creating new ones,https://github.com/pytorch/pytorch/issues/157832,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: fsdp, oncall: distributed",review linked PR state before touching issue,2025-07-08T19:31:15Z
issue,157610,`torch.compile` fails on `prims.broadcast_in_dim` with alias annotation error,https://github.com/pytorch/pytorch/issues/157610,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: pt2-dispatcher, oncall: pt2",review linked PR state before touching issue,2025-07-07T22:51:34Z
issue,155584,[Upstream Triton] Handle user-specified triton.set_allocator function,https://github.com/pytorch/pytorch/issues/155584,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",review linked PR state before touching issue,2025-07-07T15:12:29Z
issue,157217,[dynamo] dynamo is unable to enter `except RuntimeError` while eager can,https://github.com/pytorch/pytorch/issues/157217,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",review linked PR state before touching issue,2025-07-04T14:18:30Z
issue,149982,request for faster inductor kernels for blockwise reduction across dim1 -> write,https://github.com/pytorch/pytorch/issues/149982,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",review linked PR state before touching issue,2025-06-26T12:49:49Z
issue,156614,[dynamo] `torch.Tensor.random_` causes graph break,https://github.com/pytorch/pytorch/issues/156614,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",review linked PR state before touching issue,2025-06-24T01:48:16Z
issue,156640,[dynamo] Wrap asserts (and other Dynamo errors) more clearly,https://github.com/pytorch/pytorch/issues/156640,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: compile ux, module: dynamo, oncall: pt2",review linked PR state before touching issue,2025-06-23T21:34:56Z
issue,156162,torch2.7.1 issue for torch.compile numpy,https://github.com/pytorch/pytorch/issues/156162,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, module: numpy, oncall: pt2",review linked PR state before touching issue,2025-06-20T15:57:49Z
issue,156191,Dynamo does not know how to trace method `__len__` of class `<unknown type>` with torch.logging calls,https://github.com/pytorch/pytorch/issues/156191,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",review linked PR state before touching issue,2025-06-17T19:27:01Z
issue,156127,[dynamo] Show carets in graph break stack traces,https://github.com/pytorch/pytorch/issues/156127,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: compile ux, module: dynamo, oncall: pt2",review linked PR state before touching issue,2025-06-16T22:26:40Z
issue,155800,torch.distributions.kl_divergence Fails with MultivariateNormal in Dynamo Due to _infer_size Type Error,https://github.com/pytorch/pytorch/issues/155800,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",review linked PR state before touching issue,2025-06-16T19:03:07Z
issue,156059,Dynamo trace an incorrect result on torch._C._storage_Use_Count,https://github.com/pytorch/pytorch/issues/156059,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: dynamo, oncall: pt2",review linked PR state before touching issue,2025-06-16T18:25:16Z
issue,153227,eval should handle (unhinted: (s77 > 3) | (u0 > 200)) when s77 has hint =5,https://github.com/pytorch/pytorch/issues/153227,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamic shapes, oncall: pt2",review linked PR state before touching issue,2025-06-14T02:56:33Z
issue,155856,[user triton][dynamo Fix TMA Descriptor reconstruction,https://github.com/pytorch/pytorch/issues/155856,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",review linked PR state before touching issue,2025-06-13T21:14:48Z
issue,153204,mark_unbacked for strides.,https://github.com/pytorch/pytorch/issues/153204,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamic shapes, oncall: pt2",review linked PR state before touching issue,2025-06-13T03:55:43Z
issue,146274,torch.nn.functional.one_hot has inconsistent behavior between eager and torch.compile when num_classes=0,https://github.com/pytorch/pytorch/issues/146274,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: faketensor, module: pt2-dispatcher, oncall: pt2",review linked PR state before touching issue,2025-06-12T17:48:58Z
issue,154807,FP8 scaled mm lowering ignores scale_result argument,https://github.com/pytorch/pytorch/issues/154807,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",review linked PR state before touching issue,2025-06-06T19:07:31Z
issue,154161,PGO errors out in dynamo main path for sparse tensors,https://github.com/pytorch/pytorch/issues/154161,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamic shapes, oncall: pt2",review linked PR state before touching issue,2025-06-06T01:40:07Z
issue,132929,Compile + torch.autograd.grad returns no gradients,https://github.com/pytorch/pytorch/issues/132929,has linked PR,26,open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: oncall: pt2,review linked PR state before touching issue,2025-06-06T01:38:51Z
issue,154851,Improve warning when specializations happen due to operator,https://github.com/pytorch/pytorch/issues/154851,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamic shapes, oncall: pt2",review linked PR state before touching issue,2025-06-04T23:04:04Z
issue,154284,Using `compile` on `hessian(hessian)`,https://github.com/pytorch/pytorch/issues/154284,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: functorch, oncall: pt2",review linked PR state before touching issue,2025-06-04T22:44:23Z
issue,154592,inductor benchmark_compiled_module's random initialization will cause crash for data dependent indexing,https://github.com/pytorch/pytorch/issues/154592,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",review linked PR state before touching issue,2025-06-02T01:41:22Z
issue,101192,AOTAutograd export path does not support training graphs with parameters that do not receive gradients.,https://github.com/pytorch/pytorch/issues/101192,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: aotdispatch, module: pt2-dispatcher, oncall: pt2",review linked PR state before touching issue,2025-05-30T21:45:44Z
issue,154454,[Dynamo] Confusing re-raise exception handling graph break message,https://github.com/pytorch/pytorch/issues/154454,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: compile ux, module: dynamo, oncall: pt2",review linked PR state before touching issue,2025-05-27T21:58:10Z
issue,154262,max_pool2d padding assert incorrect with dilation,https://github.com/pytorch/pytorch/issues/154262,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: edge cases, module: error checking, module: nn, module: padding",review linked PR state before touching issue,2025-05-27T16:34:34Z
issue,152750,Error on padding 0-sized tensors,https://github.com/pytorch/pytorch/issues/152750,has linked PR,26,open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: python frontend,review linked PR state before touching issue,2025-05-23T22:12:17Z
issue,154259,torch.compile regression blocks torchvision/torchbench pin upgrade,https://github.com/pytorch/pytorch/issues/154259,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",review linked PR state before touching issue,2025-05-23T22:07:28Z
issue,112586,"Fix docstring errors in _guards.py, _ops.py, _jit_internal.py, functional.py, _tensor_str.py, library.py",https://github.com/pytorch/pytorch/issues/112586,has linked PR,26,open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: docs,review linked PR state before touching issue,2025-05-22T17:10:22Z
issue,109652,InstanceNorm does not catch dim mismatch,https://github.com/pytorch/pytorch/issues/109652,has linked PR,26,open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: nn,review linked PR state before touching issue,2025-05-21T00:38:35Z
issue,153149,Behavior of kernel_size parameter of torch.nn.functional.avg_pool2d does not match with documentation,https://github.com/pytorch/pytorch/issues/153149,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: docs, module: nn, module: pooling, oncall: export",review linked PR state before touching issue,2025-05-14T18:20:04Z
issue,128649,[Nested Tensor / subclasses] view(-1) and splitting dimensions with view() failures,https://github.com/pytorch/pytorch/issues/128649,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamic shapes, module: nestedtensor, oncall: pt2",review linked PR state before touching issue,2025-05-14T17:37:07Z
issue,153175,prefer_deferred_runtime_asserts_over_guards does not work correctly e2e with compile. insert_deferred_runtime_asserts and inductor codegen skips them,https://github.com/pytorch/pytorch/issues/153175,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamic shapes, oncall: pt2",review linked PR state before touching issue,2025-05-08T20:31:08Z
issue,152346,compile generates inefficient code for mutations on small slices of inputs,https://github.com/pytorch/pytorch/issues/152346,has linked PR,26,open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: oncall: pt2,review linked PR state before touching issue,2025-05-06T15:16:04Z
issue,92600,Update quantization to make source files complient with /Zc:lambda,https://github.com/pytorch/pytorch/issues/92600,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: windows, oncall: quantization",review linked PR state before touching issue,2025-05-05T19:54:21Z
issue,101249,`einsum` is about 40x slower on CUDA than manually multiplying and summing,https://github.com/pytorch/pytorch/issues/101249,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: linear algebra, module: performance",review linked PR state before touching issue,2025-05-02T22:26:18Z
issue,152343,cudagraphs: `static_input_indices` incorrectly including SymInt graph args when using tensor subclasses + dynamic shapes,https://github.com/pytorch/pytorch/issues/152343,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: __torch_dispatch__, module: cuda graphs, module: dynamic shapes, oncall: pt2",review linked PR state before touching issue,2025-04-28T18:29:03Z
issue,152183,write a custom ViewAndMutationmeta.__repr__,https://github.com/pytorch/pytorch/issues/152183,has linked PR,26,open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: oncall: pt2,review linked PR state before touching issue,2025-04-28T13:11:29Z
issue,152095,To fix inconsistency between signature and doc on `torch.bernoulli()`,https://github.com/pytorch/pytorch/issues/152095,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: distributions, module: docs",review linked PR state before touching issue,2025-04-24T15:08:45Z
issue,149989,[dynamo][higher order ops] Make support_aliasing and support_input_mutation default to False,https://github.com/pytorch/pytorch/issues/149989,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",review linked PR state before touching issue,2025-04-24T05:48:24Z
issue,134964,Add the pages of `torch.e` and `torch.pi` to PyTorch doc,https://github.com/pytorch/pytorch/issues/134964,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: docs, module: python frontend",review linked PR state before touching issue,2025-04-23T02:51:49Z
issue,151558,2.6: SDPA disallows masked attention under vmap,https://github.com/pytorch/pytorch/issues/151558,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: functorch, module: sdpa, module: vmap",review linked PR state before touching issue,2025-04-17T23:53:36Z
issue,139756,Bug with torch.compile + fwAD,https://github.com/pytorch/pytorch/issues/139756,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: forward ad, oncall: pt2",review linked PR state before touching issue,2025-04-15T21:32:31Z
issue,151196,compile of vmap of jacfwd fails,https://github.com/pytorch/pytorch/issues/151196,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: functorch, oncall: pt2",review linked PR state before touching issue,2025-04-14T22:29:13Z
issue,149097,Aten arange behavior when dtype is int64 and step size is greater than range,https://github.com/pytorch/pytorch/issues/149097,has linked PR,26,open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: python frontend,review linked PR state before touching issue,2025-04-14T04:14:02Z
issue,150857,No way to save/load model weights between PyTorch and LibTorch (C++) in a compatible way,https://github.com/pytorch/pytorch/issues/150857,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: serialization",review linked PR state before touching issue,2025-04-09T09:11:52Z
issue,150621,"2.8.0 Nightly - ""Feature 'cvt with .bf16.f16' requires .target sm_90 or higher""",https://github.com/pytorch/pytorch/issues/150621,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",review linked PR state before touching issue,2025-04-08T20:40:14Z
issue,149894,[ued][f5-tts][dynamo] `torch.compile` changes state dict,https://github.com/pytorch/pytorch/issues/149894,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",review linked PR state before touching issue,2025-04-07T17:49:02Z
issue,142321,torch.compile 路径处理错误导致 C++ 编译失败,https://github.com/pytorch/pytorch/issues/142321,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: windows, oncall: pt2",review linked PR state before touching issue,2025-04-07T09:34:26Z
issue,150465,`torch._dynamo.nonstrict_trace` has confusing user code stacktrace,https://github.com/pytorch/pytorch/issues/150465,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",review linked PR state before touching issue,2025-04-01T23:30:09Z
issue,150321,Masking out Loads from mask_modded out regions,https://github.com/pytorch/pytorch/issues/150321,has linked PR,26,open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: flex attention,review linked PR state before touching issue,2025-03-31T19:24:02Z
issue,149857,SDPA (`EFFICIENT_ATTENTION`) slower than torch.compile decomposition on `tf32`,https://github.com/pytorch/pytorch/issues/149857,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: sdpa, oncall: pt2",review linked PR state before touching issue,2025-03-31T16:19:09Z
issue,149921,[ued][whisper][dynamo] Graph break - setattr on class object,https://github.com/pytorch/pytorch/issues/149921,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",review linked PR state before touching issue,2025-03-29T15:38:35Z
issue,149911,[dynamo] `torch.compile` doesn't respect `GradTrackingTensor`'s data attribute mutation check,https://github.com/pytorch/pytorch/issues/149911,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",review linked PR state before touching issue,2025-03-29T15:38:34Z
issue,149909,[ued][kokoro] RNN/LSTMS do not work with torch.compile,https://github.com/pytorch/pytorch/issues/149909,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",review linked PR state before touching issue,2025-03-29T15:38:33Z
issue,149968,[ued][qwen][dynamo] inspect.signature with non-constant function,https://github.com/pytorch/pytorch/issues/149968,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",review linked PR state before touching issue,2025-03-29T15:35:23Z
issue,149895,[Dynamo] Cannot instantiate class if `__getattribute__` is defined,https://github.com/pytorch/pytorch/issues/149895,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",review linked PR state before touching issue,2025-03-29T15:00:26Z
issue,137275,Device transfer for NJT within torch.compile allocates a new nested int,https://github.com/pytorch/pytorch/issues/137275,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: nestedtensor, oncall: pt2",review linked PR state before touching issue,2025-03-28T18:54:25Z
issue,128063,[Inductor] Fusion of Tiled Point-Wise and Reduction Operators,https://github.com/pytorch/pytorch/issues/128063,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: floatx (formerly float8), module: inductor, oncall: pt2",review linked PR state before touching issue,2025-03-27T15:20:10Z
issue,133734,PyTorch `torch.compile` inductor backend shows enormous compile-time regressions.,https://github.com/pytorch/pytorch/issues/133734,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: compile-time, module: dynamic shapes, oncall: pt2",review linked PR state before touching issue,2025-03-27T02:11:58Z
issue,133924,AOTAutograd not tracing AMP correctly when using PyTorch/XLA.,https://github.com/pytorch/pytorch/issues/133924,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision), module: aotdispatch, module: pt2-dispatcher, module: xla",review linked PR state before touching issue,2025-03-26T12:16:01Z
issue,134533,torch._assert does not work with torch.export,https://github.com/pytorch/pytorch/issues/134533,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: oncall: export, oncall: pt2",review linked PR state before touching issue,2025-03-26T00:13:49Z
issue,149556,[ued][f5-tts][dynamo] dont graph break on `torch.jit.isinstance`,https://github.com/pytorch/pytorch/issues/149556,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",review linked PR state before touching issue,2025-03-25T00:29:11Z
issue,149566,User-torch._dynamo.disable annotations may be misleading,https://github.com/pytorch/pytorch/issues/149566,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",review linked PR state before touching issue,2025-03-24T10:35:36Z
issue,149583,Easydict support,https://github.com/pytorch/pytorch/issues/149583,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",review linked PR state before touching issue,2025-03-20T07:16:21Z
issue,149047,Unsupported: call_method NNModuleVariable() register_forward_hook [NestedUserFunctionVariable()] {},https://github.com/pytorch/pytorch/issues/149047,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",review linked PR state before touching issue,2025-03-19T01:18:07Z
issue,148842,We should use max size instead of hint size when autotuning,https://github.com/pytorch/pytorch/issues/148842,has linked PR,26,open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: oncall: pt2,review linked PR state before touching issue,2025-03-18T13:18:55Z
issue,143495,"infer_size(a, b) fails when it could return a value",https://github.com/pytorch/pytorch/issues/143495,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: faketensor, oncall: pt2",review linked PR state before touching issue,2025-03-14T14:48:03Z
issue,149037,(Will PR if ok) Support generator returning values,https://github.com/pytorch/pytorch/issues/149037,has linked PR,26,open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: oncall: pt2,review linked PR state before touching issue,2025-03-13T06:30:28Z
issue,149010,[XFORMERS] torch._dynamo.exc.Unsupported,https://github.com/pytorch/pytorch/issues/149010,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",review linked PR state before touching issue,2025-03-13T05:38:42Z
issue,148859,[Inductor] Output mismatch shape after compilation,https://github.com/pytorch/pytorch/issues/148859,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: pt2 accuracy, oncall: pt2",review linked PR state before touching issue,2025-03-10T22:25:53Z
issue,148779,[Export Benchmark] non-strict export doesn't work with .numpy on Kokoro model,https://github.com/pytorch/pytorch/issues/148779,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: oncall: export, oncall: pt2",review linked PR state before touching issue,2025-03-10T20:59:21Z
issue,147991,Verifier (in torch.export.export) does not make use of if-condition inside branches,https://github.com/pytorch/pytorch/issues/147991,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: oncall: export, oncall: pt2",review linked PR state before touching issue,2025-03-10T17:27:31Z
issue,140960,Better Error Message for SymIntArrayRef expected to contain only concrete integers,https://github.com/pytorch/pytorch/issues/140960,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamic shapes, oncall: pt2",review linked PR state before touching issue,2025-03-08T08:25:28Z
issue,148475,Dynamo replaces exception by hard error in `run_node`,https://github.com/pytorch/pytorch/issues/148475,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",review linked PR state before touching issue,2025-03-06T00:53:51Z
issue,148199,[Inductor-CPU] qlinear_binary output may have undefined strides with dynamic shape support,https://github.com/pytorch/pytorch/issues/148199,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamic shapes, module: dynamo, oncall: cpu inductor, oncall: pt2",review linked PR state before touching issue,2025-03-04T15:38:40Z
issue,147995,Checkpoint doesn't work with torch_function if torch_function change tensor metadata,https://github.com/pytorch/pytorch/issues/147995,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: __torch_function__, module: activation checkpointing",review linked PR state before touching issue,2025-02-27T16:15:48Z
issue,139603,`torch.compile` doesn't support `set` arguments,https://github.com/pytorch/pytorch/issues/139603,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",review linked PR state before touching issue,2025-02-27T00:25:25Z
issue,134950,"Bert model training is not running on cpu (X86/ARM) with torch.compile() mode, generating value error.",https://github.com/pytorch/pytorch/issues/134950,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",review linked PR state before touching issue,2025-02-26T22:55:23Z
issue,145445,[dynamo] `random.Random` gives wrong result on second call,https://github.com/pytorch/pytorch/issues/145445,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",review linked PR state before touching issue,2025-02-26T22:16:19Z
issue,147843,The opt_output in `fx_graph_runnable.py` is inconsistent with the actual output when testing run_repro(acc=True).,https://github.com/pytorch/pytorch/issues/147843,has linked PR,26,open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: oncall: pt2,review linked PR state before touching issue,2025-02-25T14:06:40Z
issue,122729,torch.func transforms don't yet support saved tensor hooks.,https://github.com/pytorch/pytorch/issues/122729,has linked PR,26,open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: vmap,review linked PR state before touching issue,2025-02-25T11:36:45Z
issue,147530,[fx][dynamo][be] Don't allow arbitrary dataclass in the graph,https://github.com/pytorch/pytorch/issues/147530,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",review linked PR state before touching issue,2025-02-25T06:29:37Z
issue,127863,Potential 7.5x compilation time improvement for foreach kernel,https://github.com/pytorch/pytorch/issues/127863,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: inductor, module: pt2 optimizer, oncall: pt2",review linked PR state before touching issue,2025-02-24T21:44:13Z
issue,147486,remote_cache.py loggings cause a logging error after pytest exit,https://github.com/pytorch/pytorch/issues/147486,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: logging, module: testing, oncall: pt2",review linked PR state before touching issue,2025-02-24T19:00:20Z
issue,147212,dict_tag optimization leads to wrong results with relational guards,https://github.com/pytorch/pytorch/issues/147212,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",review linked PR state before touching issue,2025-02-24T15:14:03Z
issue,129456,[dynamic][inline-inbuilt-nn-modules] Extra compilation - pytest test/onnx/dynamo/test_dynamo_with_onnxruntime_backend.py::TestDynamoWithONNXRuntime::test_llama_attention_with_local_backend_0,https://github.com/pytorch/pytorch/issues/129456,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamic shapes, module: onnx, oncall: pt2",review linked PR state before touching issue,2025-02-19T17:15:15Z
issue,128556,Dynamo graph breaks when passing a constant value to a nn module initialized within a function,https://github.com/pytorch/pytorch/issues/128556,has linked PR,26,open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: oncall: pt2,review linked PR state before touching issue,2025-02-18T23:29:31Z
issue,133462,PT2 inference slowdown after updating Huggingface pin on CI,https://github.com/pytorch/pytorch/issues/133462,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: inductor, module: performance, oncall: pt2",review linked PR state before touching issue,2025-02-14T22:38:32Z
issue,146951,[dynamo] `x is x` gets incorrectly interpreted to `False`,https://github.com/pytorch/pytorch/issues/146951,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",review linked PR state before touching issue,2025-02-13T06:11:05Z
issue,146673,"Dynamo should handle `class.method_descriptor(instance, *args, **kwargs)` generically",https://github.com/pytorch/pytorch/issues/146673,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",review linked PR state before touching issue,2025-02-08T01:28:56Z
issue,146628,Dynamo should consider tensor mutation when reconstructing generator,https://github.com/pytorch/pytorch/issues/146628,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",review linked PR state before touching issue,2025-02-07T02:52:01Z
issue,145596,[Dynamo] compile torch.logit with different data types,https://github.com/pytorch/pytorch/issues/145596,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamic shapes, oncall: pt2",review linked PR state before touching issue,2025-01-27T18:11:54Z
issue,113490,/nodefaultlib:vcomp doesn't seem to be set when compiling with Intel OpenMP on Windows,https://github.com/pytorch/pytorch/issues/113490,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: build, module: windows",review linked PR state before touching issue,2025-01-27T12:22:51Z
issue,145529,Module.to() fail in dynamo when swap_module_params_on_conversion is true,https://github.com/pytorch/pytorch/issues/145529,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",review linked PR state before touching issue,2025-01-24T06:47:14Z
issue,145231,Flaky Dynamo test: TestAutograd.test_gradcheck_nondeterministic,https://github.com/pytorch/pytorch/issues/145231,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: testing, oncall: pt2",review linked PR state before touching issue,2025-01-21T22:37:42Z
issue,143463,unreasonable ConstraintViolationError when using torch dynamo to compile torch model,https://github.com/pytorch/pytorch/issues/143463,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamic shapes, module: dynamo, oncall: export, oncall: pt2",review linked PR state before touching issue,2025-01-07T19:13:58Z
issue,132133,Adding betainc with backward operation,https://github.com/pytorch/pytorch/issues/132133,has linked PR,26,open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: special,review linked PR state before touching issue,2025-01-02T13:43:34Z
issue,142489,Stop importing arbitrary third-party libraries in Dynamo,https://github.com/pytorch/pytorch/issues/142489,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",review linked PR state before touching issue,2024-12-27T23:50:53Z
issue,135759,core_aten_decompositions table have non-functional or CIA entries.,https://github.com/pytorch/pytorch/issues/135759,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: pt2-dispatcher, oncall: export, oncall: pt2",review linked PR state before touching issue,2024-12-17T22:09:34Z
issue,142378,Support int step for nonfused optimizer implementations,https://github.com/pytorch/pytorch/issues/142378,has linked PR,26,open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: optimizer,review linked PR state before touching issue,2024-12-09T16:30:25Z
issue,141916,rms_norm forward performance analysis,https://github.com/pytorch/pytorch/issues/141916,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",review linked PR state before touching issue,2024-12-04T04:26:20Z
issue,137996,cprofile on compiler should trigger on backwards,https://github.com/pytorch/pytorch/issues/137996,has linked PR,26,open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: oncall: pt2,review linked PR state before touching issue,2024-12-02T17:55:39Z
issue,141744,"""RuntimeError: Triton Error [CUDA]: device kernel image is invalid"" when running generated Triton code",https://github.com/pytorch/pytorch/issues/141744,has linked PR,26,open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: inductor,review linked PR state before touching issue,2024-12-02T16:43:56Z
issue,113809,"Error during DDP, torch.compile, and cudagraph_trees",https://github.com/pytorch/pytorch/issues/113809,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: cuda graphs, module: ddp, module: inductor, oncall: distributed",review linked PR state before touching issue,2024-11-26T23:51:38Z
issue,141473,How to use torch.compile + HF model?,https://github.com/pytorch/pytorch/issues/141473,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",review linked PR state before touching issue,2024-11-26T04:21:05Z
issue,140909,[2.5.0+] Running a compiled model under FlopCounterMode regresses its performance,https://github.com/pytorch/pytorch/issues/140909,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: __torch_dispatch__, module: dynamo, oncall: pt2",review linked PR state before touching issue,2024-11-22T22:19:48Z
issue,141162,Dynamo should reorder warnings (and other logging) by default if they don't involve tensor values,https://github.com/pytorch/pytorch/issues/141162,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",review linked PR state before touching issue,2024-11-22T03:14:59Z
issue,141017,multilabel_margin_loss gives incorrect results in torch.compile,https://github.com/pytorch/pytorch/issues/141017,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: pt2 accuracy, oncall: pt2",review linked PR state before touching issue,2024-11-22T02:54:54Z
issue,141258,`torch.compile` fails with runtime error on Numpy scalar operations,https://github.com/pytorch/pytorch/issues/141258,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: numpy, oncall: pt2",review linked PR state before touching issue,2024-11-21T23:17:20Z
issue,123972,Dynamo export: limited support in Torch.cond,https://github.com/pytorch/pytorch/issues/123972,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: higher order operators, module: onnx, oncall: pt2",review linked PR state before touching issue,2024-11-21T19:46:00Z
issue,140845,compiled LSTM performance is worse than not compiled,https://github.com/pytorch/pytorch/issues/140845,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: decompositions, oncall: pt2",review linked PR state before touching issue,2024-11-19T14:59:07Z
issue,140884,`torch.compile` tries to invoke a torch op on numpy.ndarray in nightly release,https://github.com/pytorch/pytorch/issues/140884,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",review linked PR state before touching issue,2024-11-18T20:16:01Z
issue,140795,Memory usage increase post compilation for torch.compile ViT-H-14-quickgelu_dfn5b,https://github.com/pytorch/pytorch/issues/140795,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",review linked PR state before touching issue,2024-11-18T18:36:09Z
issue,139629,Inductor should run fallbacks without Autograd-related dispatch keys,https://github.com/pytorch/pytorch/issues/139629,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",review linked PR state before touching issue,2024-11-08T19:45:53Z
issue,139232,TORCH_COMPILE_CPROFILE=1 doesn't work on python 3.12,https://github.com/pytorch/pytorch/issues/139232,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",review linked PR state before touching issue,2024-10-31T00:18:13Z
issue,139080,[dynamo] `dynamo_expected_failures` is silent on certain tests that ended up passing,https://github.com/pytorch/pytorch/issues/139080,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, module: tests, oncall: pt2",review linked PR state before touching issue,2024-10-30T23:44:17Z
issue,138926,torch.cond complains two tensors have different metadata,https://github.com/pytorch/pytorch/issues/138926,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: oncall: export, oncall: pt2",review linked PR state before touching issue,2024-10-29T15:11:32Z
issue,41245,torch.sign is divergent from numpy.sign on NaN,https://github.com/pytorch/pytorch/issues/41245,has linked PR,26,open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: numpy,review linked PR state before touching issue,2024-10-19T11:06:27Z
issue,105839,"MPS memory issue, MPS backend out of memory, but works if I empty the MPS cache",https://github.com/pytorch/pytorch/issues/105839,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: memory usage, module: mps",review linked PR state before touching issue,2024-10-17T12:02:04Z
issue,126474,[dynamo] Handle inplace op aliasing errors,https://github.com/pytorch/pytorch/issues/126474,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: dynamo, oncall: pt2",review linked PR state before touching issue,2024-10-16T18:26:21Z
issue,137546,`torch.linspace` doesn't agree with decomposition.,https://github.com/pytorch/pytorch/issues/137546,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: decompositions, module: linear algebra",review linked PR state before touching issue,2024-10-09T13:31:24Z
issue,137468,overflow in `torch.quantize_per_tensor`,https://github.com/pytorch/pytorch/issues/137468,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: error checking, oncall: quantization",review linked PR state before touching issue,2024-10-08T14:52:01Z
issue,93501,_pack_padded_sequence fails in dynamo due to requiring a non-fake 2nd argument,https://github.com/pytorch/pytorch/issues/93501,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, module: performance, oncall: pt2",review linked PR state before touching issue,2024-09-28T21:11:26Z
issue,136271,[triton x pt2] Dynamo should trace data_ptr accesses,https://github.com/pytorch/pytorch/issues/136271,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, module: user triton, oncall: pt2",review linked PR state before touching issue,2024-09-27T23:37:37Z
issue,136258,`to()` Method Does Not Move Internal Components of Sparse Tensors,https://github.com/pytorch/pytorch/issues/136258,has linked PR,26,open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: sparse,review linked PR state before touching issue,2024-09-26T09:52:58Z
issue,120381,torch.compile has significantly higher peak memory for a simple checkpointed layer of matmul + layernorm + gelu,https://github.com/pytorch/pytorch/issues/120381,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",review linked PR state before touching issue,2024-09-13T10:22:55Z
issue,121318,torch/export/_trace.py and related non-strict entrypoints do not initialize TORCH_LOGS/TORCH_TRACE,https://github.com/pytorch/pytorch/issues/121318,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: logging, oncall: export, oncall: pt2",review linked PR state before touching issue,2024-09-13T03:28:25Z
issue,120911,Symbolic shape tests don't actually test that arguments aren't specialized,https://github.com/pytorch/pytorch/issues/120911,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamic shapes, module: tests, oncall: pt2",review linked PR state before touching issue,2024-09-13T03:16:40Z
issue,122294,AOIInductor: Dynamic Shapes Specificiaton fails for SAM,https://github.com/pytorch/pytorch/issues/122294,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamic shapes, oncall: export, oncall: pt2",review linked PR state before touching issue,2024-09-13T03:06:30Z
issue,118747,decorate_context results in uninformative backtraces,https://github.com/pytorch/pytorch/issues/118747,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: logging, oncall: pt2",review linked PR state before touching issue,2024-09-13T03:05:17Z
issue,118332,Further simplifying guards,https://github.com/pytorch/pytorch/issues/118332,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamic shapes, oncall: pt2",review linked PR state before touching issue,2024-09-12T18:17:24Z
issue,135492,inductor guards on 32 bit indexing even when generated code would be the same,https://github.com/pytorch/pytorch/issues/135492,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",review linked PR state before touching issue,2024-09-09T19:13:16Z
issue,123651,Divisibility substitution does the wrong thing with non-trivial max that would be < 2 after division,https://github.com/pytorch/pytorch/issues/123651,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamic shapes, oncall: pt2",review linked PR state before touching issue,2024-09-06T22:37:10Z
issue,134841,LPPool2d lacks a check for the validity of norm_type,https://github.com/pytorch/pytorch/issues/134841,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: error checking, module: norms and normalization",review linked PR state before touching issue,2024-08-30T15:28:39Z
issue,134363,Improving inductor pattern matcher's replacement graph assumptions,https://github.com/pytorch/pytorch/issues/134363,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",review linked PR state before touching issue,2024-08-26T19:04:44Z
issue,94336,`scatter` fails the gradient computation in reverse mode for `src` when `index` is empty,https://github.com/pytorch/pytorch/issues/94336,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: scatter & gather ops",review linked PR state before touching issue,2024-08-21T09:46:51Z
issue,133493,vmap + autograd.Function [generate_vmap_rule=False] + torch.compile don't work together,https://github.com/pytorch/pytorch/issues/133493,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, module: functorch, oncall: pt2",review linked PR state before touching issue,2024-08-19T19:15:26Z
issue,118334,Using autograd.Functions defined in torch/ cause graph breaks,https://github.com/pytorch/pytorch/issues/118334,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",review linked PR state before touching issue,2024-08-14T19:32:59Z
issue,133078,"`torch.compile` support for `CXX=""zig c++""`",https://github.com/pytorch/pytorch/issues/133078,has linked PR,26,open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: oncall: pt2,review linked PR state before touching issue,2024-08-10T00:44:58Z
issue,132368,Switch to new name `oneDNN` instead of `MKLDNN`,https://github.com/pytorch/pytorch/issues/132368,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: intel, module: mkldnn",review linked PR state before touching issue,2024-08-09T07:17:09Z
issue,131280,The shape of weights in lstm seems to be wrong.,https://github.com/pytorch/pytorch/issues/131280,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: docs, module: rnn",review linked PR state before touching issue,2024-08-06T03:46:44Z
issue,132020,`torch.Tensor.to` ignores `memory_format` kwarg,https://github.com/pytorch/pytorch/issues/132020,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: docs, module: memory format, module: python frontend",review linked PR state before touching issue,2024-07-29T18:38:25Z
issue,126673,torch.nn.AdaptiveAvgPool2d lacks checking of input dimension,https://github.com/pytorch/pytorch/issues/126673,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: error checking, module: nn",review linked PR state before touching issue,2024-07-29T07:28:29Z
issue,103417,"Torch model compile error ""/usr/bin/ld: cannot find -lcuda"" though cuda is installed via run file",https://github.com/pytorch/pytorch/issues/103417,has linked PR,26,open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: oncall: pt2,review linked PR state before touching issue,2024-07-27T17:51:11Z
issue,130641,`dims` of `flip()` without `list` or `tuple` works with a tensor agaist the doc,https://github.com/pytorch/pytorch/issues/130641,has linked PR,26,open PR references this issue; old update age or low-urgency/stale label,review linked PR state before touching issue,2024-07-13T01:44:57Z
issue,129936,[symbolic-shapes] Guard validation in Dynamo changes the shape_env,https://github.com/pytorch/pytorch/issues/129936,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamic shapes, oncall: pt2",review linked PR state before touching issue,2024-07-09T16:33:53Z
issue,129970,Setting the 0D tensor of only one element to `fill_value` of `full()`/`full_like()` works against error message,https://github.com/pytorch/pytorch/issues/129970,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: docs, module: error checking, module: python frontend",review linked PR state before touching issue,2024-07-08T13:15:04Z
issue,111320,test_max_pool1d reliably OOMs after https://github.com/pytorch/pytorch/pull/111216/,https://github.com/pytorch/pytorch/issues/111320,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: memory usage, oncall: pt2",review linked PR state before touching issue,2024-06-30T11:49:05Z
issue,112788,async_compile.triton is unable to cache generated triton kernels due to meta containing kernel names,https://github.com/pytorch/pytorch/issues/112788,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",review linked PR state before touching issue,2024-06-28T19:01:23Z
issue,129338,`linspace()` can also use `complex` and `bool` type for `start` and `end` argument against the doc,https://github.com/pytorch/pytorch/issues/129338,has linked PR,26,open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: docs,review linked PR state before touching issue,2024-06-24T15:51:03Z
issue,102839,Dynamo should only unroll loops by a preset factor (unless otherwise explicitly instructed),https://github.com/pytorch/pytorch/issues/102839,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: compile-time, module: dynamo, oncall: pt2",review linked PR state before touching issue,2024-06-18T18:49:39Z
issue,117265,[dynamo][inline-inbuilt-nn-modules]torch.compile silently incorrect with full_backward_pre_hook,https://github.com/pytorch/pytorch/issues/117265,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",review linked PR state before touching issue,2024-06-18T16:56:47Z
issue,124163,Support `dynamic=True` in torch._dynamo.explain,https://github.com/pytorch/pytorch/issues/124163,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",review linked PR state before touching issue,2024-06-17T23:52:44Z
issue,123411,dynamo/fx doesn't honor 'non-persistent' buffers,https://github.com/pytorch/pytorch/issues/123411,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",review linked PR state before touching issue,2024-06-17T23:44:44Z
issue,123374,[dynamo] incorrect error traceback for runtime errors when executing dynamo codegen,https://github.com/pytorch/pytorch/issues/123374,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",review linked PR state before touching issue,2024-06-17T23:43:29Z
issue,122578,Recompiles and cache_size_limit from detectron2 CycleBatchNormList,https://github.com/pytorch/pytorch/issues/122578,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",review linked PR state before touching issue,2024-06-17T23:40:15Z
issue,126642,"[optim] add missing 'maximize' parameter to LBFGS, NAdam and RAdam optimizers",https://github.com/pytorch/pytorch/issues/126642,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: loss, module: optimizer",review linked PR state before touching issue,2024-06-03T15:17:58Z
issue,38019,"torch.cuda.nccl.init_rank does not handle ""uid"" properly, causing runtime error",https://github.com/pytorch/pytorch/issues/38019,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: nccl",review linked PR state before touching issue,2024-06-02T10:21:34Z
issue,121526,Compiling RMSNorm Triton Kernal gives error,https://github.com/pytorch/pytorch/issues/121526,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: user triton, oncall: pt2",review linked PR state before touching issue,2024-05-22T06:54:40Z
issue,126514,"Weights become NaN with torch.compile optimizer capturable=True, lr=0.0, nn.Embedding",https://github.com/pytorch/pytorch/issues/126514,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: optimizer, oncall: pt2",review linked PR state before touching issue,2024-05-20T22:09:44Z
issue,125277,H100 Max Autotune Compilation Significantly Slower than A100,https://github.com/pytorch/pytorch/issues/125277,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",review linked PR state before touching issue,2024-05-14T18:43:02Z
issue,125236,torch.no_grad() is not working for dynamo inductor backend,https://github.com/pytorch/pytorch/issues/125236,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",review linked PR state before touching issue,2024-05-06T18:36:37Z
issue,125388,torch.multinomial raises no error when sampling from zero weight and replacement=False.,https://github.com/pytorch/pytorch/issues/125388,has linked PR,26,open PR references this issue; old update age or low-urgency/stale label,review linked PR state before touching issue,2024-05-03T00:40:42Z
issue,124110,Equality between SymBool is extremely busted,https://github.com/pytorch/pytorch/issues/124110,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamic shapes, oncall: pt2",review linked PR state before touching issue,2024-04-18T16:29:35Z
issue,105560,aten._trilinear,https://github.com/pytorch/pytorch/issues/105560,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",review linked PR state before touching issue,2024-03-28T23:16:52Z
issue,122191,torch.std_mean slower than separate torch.mean and torch.std calls on CPU,https://github.com/pytorch/pytorch/issues/122191,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: performance",review linked PR state before touching issue,2024-03-20T07:52:28Z
issue,120819,Support a context manager to set torch dtype?,https://github.com/pytorch/pytorch/issues/120819,has linked PR,26,open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: python frontend,review linked PR state before touching issue,2024-03-06T18:36:46Z
issue,119883,ModularIndexing does not have well-defined semantics,https://github.com/pytorch/pytorch/issues/119883,has linked PR,26,open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: inductor,review linked PR state before touching issue,2024-02-14T14:19:16Z
issue,119300,Optimizer step pre/post hook tests should try harder to ascertain the hook happens before/after the step call,https://github.com/pytorch/pytorch/issues/119300,has linked PR,26,open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: optimizer,review linked PR state before touching issue,2024-02-06T19:16:08Z
issue,111471,test_learnable_forward_per_channel fails due to integer overflow,https://github.com/pytorch/pytorch/issues/111471,has linked PR,26,open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,review linked PR state before touching issue,2023-12-21T21:09:00Z
issue,104860,torch.nn.Conv2d/Conv1d's padding mode circular cannot accept 3-dim input,https://github.com/pytorch/pytorch/issues/104860,has linked PR,26,open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: nn,review linked PR state before touching issue,2023-12-11T08:53:04Z
issue,93617,[Inductor] Input Buffers Should Be Representable As Storage And Layout,https://github.com/pytorch/pytorch/issues/93617,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",review linked PR state before touching issue,2023-11-30T04:07:16Z
issue,78618,Module parameters/submodules can be shadowed by class attributes silently,https://github.com/pytorch/pytorch/issues/78618,has linked PR,26,open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: nn,review linked PR state before touching issue,2023-11-07T10:33:10Z
issue,110865,BCEWithLogitsLoss: Check if labels / targets are within zero and one,https://github.com/pytorch/pytorch/issues/110865,has linked PR,26,open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: nn,review linked PR state before touching issue,2023-10-26T15:55:52Z
issue,105318,[docs] torch.sigmoid to make clear equivalence relations to other sigmoid functions,https://github.com/pytorch/pytorch/issues/105318,has linked PR,26,open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: docs,review linked PR state before touching issue,2023-07-24T18:36:15Z
issue,103584,torch.cuda.mem_get_info to return 0 if CUDA context isn't initialized,https://github.com/pytorch/pytorch/issues/103584,has linked PR,26,open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: cuda,review linked PR state before touching issue,2023-06-20T20:22:07Z
issue,99421,Question about GRU(RNN/LSTM) outputs shape,https://github.com/pytorch/pytorch/issues/99421,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: docs, module: nn, module: rnn",review linked PR state before touching issue,2023-04-21T19:13:49Z
issue,97750,irrelevant error output for Minified repro,https://github.com/pytorch/pytorch/issues/97750,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: minifier, oncall: pt2",review linked PR state before touching issue,2023-03-30T21:14:22Z
issue,97310,MPS: `unique` and `unique_consecutive` extremely slow when `return_counts=True`,https://github.com/pytorch/pytorch/issues/97310,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: mps, module: performance",review linked PR state before touching issue,2023-03-23T18:06:57Z
issue,70925,allow `dim=None` in `concat`,https://github.com/pytorch/pytorch/issues/70925,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: python array api, module: viewing and reshaping",review linked PR state before touching issue,2023-03-07T09:33:27Z
issue,70915,`arange` should return empty array if bounds are inconsistent with step sign,https://github.com/pytorch/pytorch/issues/70915,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: python array api, module: tensor creation",review linked PR state before touching issue,2023-03-07T09:32:09Z
issue,86558,Data conversion ops ignore `memory_format=torch.contiguous_format`,https://github.com/pytorch/pytorch/issues/86558,has linked PR,26,open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: primtorch,review linked PR state before touching issue,2022-10-11T05:49:35Z
issue,85960,Discrepancy in output shape for batch_norm inference mode between CUDA and CPU,https://github.com/pytorch/pytorch/issues/85960,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: nn",review linked PR state before touching issue,2022-10-04T17:13:57Z
issue,80301,"Need ""valid"" and ""same"" padding mode for convTranspose2d",https://github.com/pytorch/pytorch/issues/80301,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: nn, module: padding",review linked PR state before touching issue,2022-06-28T15:58:26Z
issue,51455,"[docs] Improve documentation for LayerNorm, GroupNorm, etc (+ add python reference impl)",https://github.com/pytorch/pytorch/issues/51455,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: docs, module: nn, module: norms and normalization",review linked PR state before touching issue,2022-05-25T18:34:20Z
issue,77170,Move the MPSGuardImpl to inherit from NoOpDeviceGuardImpl,https://github.com/pytorch/pytorch/issues/77170,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: backend, module: mps",review linked PR state before touching issue,2022-05-11T14:24:09Z
issue,71392,Error in `torch.trapz` documentation,https://github.com/pytorch/pytorch/issues/71392,has linked PR,26,open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: docs,review linked PR state before touching issue,2022-01-27T16:05:20Z
issue,67969,"[docs] F.multilabel_soft_margin_loss does not list arg reduction, but lists outdated size_average/reduce (true for other losses as well)",https://github.com/pytorch/pytorch/issues/67969,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: docs, module: nn",review linked PR state before touching issue,2021-11-18T17:37:11Z
issue,67887,Add more explanation on multithreaded graph building of Autograd,https://github.com/pytorch/pytorch/issues/67887,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: docs, module: multiprocessing, module: multithreading",review linked PR state before touching issue,2021-11-10T10:39:30Z
pr,186435,Fix mkldnn FakeTensor error handling,https://github.com/pytorch/pytorch/pull/186435,draft/noise,23,PR is draft; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,wait for author to mark ready,2026-07-08T11:53:23Z
pr,189006,Ensure correct sign of input for test_laplace in test_distributions.py,https://github.com/pytorch/pytorch/pull/189006,draft/noise,23,PR is draft; PR closes or references open issue numbers,wait for author to mark ready,2026-07-06T09:20:20Z
pr,186778,Reduce fp16 batch-1 Conv2d forward divergence on XPU,https://github.com/pytorch/pytorch/pull/186778,draft/noise,23,PR is draft; PR closes or references open issue numbers,wait for author to mark ready,2026-07-06T07:48:06Z
pr,180857,Add variable-length support for Flash Attention in XPU,https://github.com/pytorch/pytorch/pull/180857,draft/noise,23,PR is draft; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,wait for author to mark ready,2026-06-30T11:20:01Z
pr,188143,Add mixed Tensor and Scalar bound support for torch.clamp and torch.clip,https://github.com/pytorch/pytorch/pull/188143,draft/noise,23,PR is draft; PR closes or references open issue numbers,wait for author to mark ready,2026-06-29T14:05:29Z
pr,187930,Fix torch.compile validation for random ops,https://github.com/pytorch/pytorch/pull/187930,draft/noise,23,PR is draft; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,wait for author to mark ready,2026-06-28T02:34:01Z
pr,187409,[distributed] Add NCCL M2N-backed symm_mem reshard,https://github.com/pytorch/pytorch/pull/187409,draft/noise,23,"PR is draft; PR closes or references open issue numbers; CI lane labels: ciflow/h100-symm-mem, ciflow/torchtitan",wait for author to mark ready,2026-06-24T23:55:59Z
pr,186089,[XPU] Align XPU unit tests decorators and overrides.,https://github.com/pytorch/pytorch/pull/186089,draft/noise,23,PR is draft; PR closes or references open issue numbers; CI lane labels: ciflow/trunk,wait for author to mark ready,2026-06-24T11:40:04Z
pr,187896,Re-land chain-based dispatch for custom_op fast path with override detection,https://github.com/pytorch/pytorch/pull/187896,draft/noise,23,PR is draft; PR closes or references open issue numbers,wait for author to mark ready,2026-06-23T01:36:15Z
pr,187554,Fix flaky gradcheck in _gradcheck_log_prob for simplex distributions.,https://github.com/pytorch/pytorch/pull/187554,draft/noise,23,PR is draft; PR closes or references open issue numbers,wait for author to mark ready,2026-06-17T12:08:55Z
pr,183530,feat: add XPUCodeCache tests for compilation and execution,https://github.com/pytorch/pytorch/pull/183530,draft/noise,23,"PR is draft; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for author to mark ready,2026-06-17T12:02:41Z
pr,183366,[FSDP] Support replicated shard placement params,https://github.com/pytorch/pytorch/pull/183366,draft/noise,23,"PR is draft; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for author to mark ready,2026-06-16T05:40:51Z
pr,182147,Support XPU and other accelerators in RNG state handling,https://github.com/pytorch/pytorch/pull/182147,draft/noise,23,"PR is draft; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/trunk",wait for author to mark ready,2026-06-08T11:40:17Z
pr,186068,Add skipIfXpu decorator to tests for semi-structured sparsity support,https://github.com/pytorch/pytorch/pull/186068,draft/noise,23,PR is draft; PR closes or references open issue numbers,wait for author to mark ready,2026-06-03T10:46:23Z
pr,183875,Add claude skill for PrivateUse1 (PU1) third-party device integration,https://github.com/pytorch/pytorch/pull/183875,draft/noise,23,PR is draft; review decision requests changes; PR has not updated for at least 45 days,wait for contributor update,2026-05-20T02:07:32Z
pr,179565,[WIP][CI] Add RHEL 9.6 build and test support,https://github.com/pytorch/pytorch/pull/179565,draft/noise,23,"PR is draft; review decision requests changes; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-05-15T05:47:27Z
pr,180225,Enable ACL operator fusions for aarch64,https://github.com/pytorch/pytorch/pull/180225,draft/noise,23,"PR is draft; review decision requests changes; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/linux-aarch64",wait for contributor update,2026-05-11T09:07:09Z
pr,97825,Support sparse COO/CSR/CSC/BSR/BSC return values in gradcheck input function,https://github.com/pytorch/pytorch/pull/97825,draft/noise,23,PR is draft; review decision requests changes; PR has not updated for at least 45 days,wait for contributor update,2025-06-19T08:46:33Z
pr,140372,[Intel GPU] Enable mkldnn::_convolution.pointwise at XPU backend,https://github.com/pytorch/pytorch/pull/140372,draft/noise,23,PR is draft; review decision requests changes; PR has not updated for at least 45 days; CI lane labels: ciflow/xpu,wait for contributor update,2025-05-24T06:21:32Z
issue,93419,"torchdynamo is not properly setting up input tracking (e.g., for symbolic shape guards) for view bases",https://github.com/pytorch/pytorch/issues/93419,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: dynamic shapes, oncall: pt2",route to design/research decision before patch planning,2026-03-31T23:18:40Z
issue,50112,need a clear guide for when and how to use torch.cuda.set_device(),https://github.com/pytorch/pytorch/issues/50112,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: docs",route to design/research decision before patch planning,2025-12-26T12:57:46Z
issue,170944,Indexing a tensor with numpy boolean array is very slow,https://github.com/pytorch/pytorch/issues/170944,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label,route to design/research decision before patch planning,2025-12-22T18:24:45Z
issue,83529,Adding Levenberg-marquardt optimizer in PyTorch,https://github.com/pytorch/pytorch/issues/83529,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: optimizer,route to design/research decision before patch planning,2025-12-08T14:18:29Z
issue,160553,[MPS] Out of bounds indexing inconsistency between MPS and CPU/CUDA,https://github.com/pytorch/pytorch/issues/160553,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: error checking, module: mps",route to design/research decision before patch planning,2025-12-02T09:28:54Z
issue,156424,research adding cuda-bindings to core,https://github.com/pytorch/pytorch/issues/156424,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda,route to design/research decision before patch planning,2025-11-18T18:31:46Z
issue,130451,vmap over torch.allclose doesn't work,https://github.com/pytorch/pytorch/issues/130451,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: functorch,route to design/research decision before patch planning,2025-11-17T13:31:48Z
issue,158043,Add GoLU Activation Function 🚀,https://github.com/pytorch/pytorch/issues/158043,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: nn,route to design/research decision before patch planning,2025-11-06T14:41:14Z
issue,99873,Dynamo config patching in our code is brittle,https://github.com/pytorch/pytorch/issues/99873,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",route to design/research decision before patch planning,2025-11-05T19:40:03Z
issue,166826,PyTorch Layer for Pre-Summation Nonlinear Feature Transformation,https://github.com/pytorch/pytorch/issues/166826,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: nn, module: python frontend",route to design/research decision before patch planning,2025-11-04T19:29:44Z
issue,6662,Autogenerate code example / tutorial outputs in documentation,https://github.com/pytorch/pytorch/issues/6662,good volunteer slice,20,"label says issue is suitable for a scoped external contribution; old update age or low-urgency/stale label; component/oncall labels: module: doc infra, module: docs",scan when time allows,2025-10-27T15:13:59Z
issue,166208,OptimizerDict - design and write an abstraction for a single Optim that'd control multiple optimizers,https://github.com/pytorch/pytorch/issues/166208,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: optimizer,route to design/research decision before patch planning,2025-10-27T14:09:30Z
issue,165577,CI: What is the purpose of `slow.yml`,https://github.com/pytorch/pytorch/issues/165577,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: ci,route to design/research decision before patch planning,2025-10-16T09:24:45Z
issue,83264,Emulating FP64 and increased precisions on Apple silicon,https://github.com/pytorch/pytorch/issues/83264,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: mps,route to design/research decision before patch planning,2025-10-04T13:07:30Z
issue,24826,Transformer Lack of Embedding Layer and Positional Encodings,https://github.com/pytorch/pytorch/issues/24826,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: docs, module: nn",route to design/research decision before patch planning,2025-09-29T23:14:42Z
issue,65392,torch.sparse.sum result has wrong dtype when reducing over all dimensions,https://github.com/pytorch/pytorch/issues/65392,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: sparse",route to design/research decision before patch planning,2025-09-27T03:36:32Z
issue,55152,Better handling of OpInfo.sample_inputs,https://github.com/pytorch/pytorch/issues/55152,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: tests,route to design/research decision before patch planning,2025-09-26T23:51:21Z
issue,84347,Lack of newly raised optimizers,https://github.com/pytorch/pytorch/issues/84347,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: optimizer,route to design/research decision before patch planning,2025-09-26T23:37:46Z
issue,68332,[feature request] Provide functional form of scheduler formulas (and reconsider older decisions of not doing it),https://github.com/pytorch/pytorch/issues/68332,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: lrscheduler, module: optimizer",route to design/research decision before patch planning,2025-09-19T01:43:33Z
issue,160637,Support optimisation of sparse parameters via LBFGS,https://github.com/pytorch/pytorch/issues/160637,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: optimizer, module: sparse",route to design/research decision before patch planning,2025-09-17T15:58:39Z
issue,141994,Add torch.set_default_int_dtype() / extend set_default_dtype() to allow setting default signed integer dtype,https://github.com/pytorch/pytorch/issues/141994,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: python frontend,route to design/research decision before patch planning,2025-09-16T18:25:56Z
issue,153669,Make tlparse able to show a summary of distinct graph breaks,https://github.com/pytorch/pytorch/issues/153669,good volunteer slice,20,"label says issue is suitable for a scoped external contribution; old update age or low-urgency/stale label; component/oncall labels: module: compile ux, module: dynamo, oncall: pt2",scan when time allows,2025-09-12T20:16:04Z
issue,160322,`torch.Tensor` methods return type annotation too broad for `torch.Tensor` subclasses,https://github.com/pytorch/pytorch/issues/160322,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: typing,route to design/research decision before patch planning,2025-09-02T03:09:22Z
issue,161129,autograd assertion error in cuda backend when privateuse1 is renamed,https://github.com/pytorch/pytorch/issues/161129,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: privateuse1",route to design/research decision before patch planning,2025-08-29T20:17:25Z
issue,160489,[CI] Mergebot tests are no longer updatable,https://github.com/pytorch/pytorch/issues/160489,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: ci, module: tests",route to design/research decision before patch planning,2025-08-26T20:06:48Z
issue,89116,Implement generic batch normalization layer.,https://github.com/pytorch/pytorch/issues/89116,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: nn, oncall: distributed",route to design/research decision before patch planning,2025-08-25T10:22:07Z
issue,109691,Extends the functionality of `nn.BatchNorm1d`.,https://github.com/pytorch/pytorch/issues/109691,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: nn, oncall: distributed",route to design/research decision before patch planning,2025-08-25T10:21:04Z
issue,158709,autograd precision: float32 gradients of bfloat16 weights,https://github.com/pytorch/pytorch/issues/158709,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: autograd,route to design/research decision before patch planning,2025-08-20T16:29:03Z
issue,125422,Tensor roll with different shifts,https://github.com/pytorch/pytorch/issues/125422,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: numpy,route to design/research decision before patch planning,2025-08-11T19:08:53Z
issue,159765,Add nn.GradBank for gradient scaling to prevent vanishing/exploding gradients,https://github.com/pytorch/pytorch/issues/159765,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: nn,route to design/research decision before patch planning,2025-08-08T18:54:19Z
issue,112303,"[dynamo]RFC/Feature request - cache, guard, and reuse InliningInstructionTranslator inline_call",https://github.com/pytorch/pytorch/issues/112303,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",route to design/research decision before patch planning,2025-08-01T22:16:05Z
issue,80168,GEGLU activation,https://github.com/pytorch/pytorch/issues/80168,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: nn,route to design/research decision before patch planning,2025-07-30T18:41:46Z
issue,130513,RuntimeError Occurred In getBar1SizeOfGpu when initializing torch RPC,https://github.com/pytorch/pytorch/issues/130513,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: tensorpipe, oncall: distributed",route to design/research decision before patch planning,2025-07-30T14:08:49Z
issue,155322,Segmentation fault after manual tensor assignment with autograd enabled on CUDA,https://github.com/pytorch/pytorch/issues/155322,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: edge cases",route to design/research decision before patch planning,2025-07-29T14:18:11Z
issue,104849,"[feature request] torch.mix function to generalize/symmetrize addcmul + can serve as a more versatlie, generalized ""optimization update rule"" function",https://github.com/pytorch/pytorch/issues/104849,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: python frontend,route to design/research decision before patch planning,2025-07-29T12:41:47Z
issue,104247,"[proposal] ""Name"" string attribute for modules, parameters, buffers, tensors for more pleasant debugging (including distributed FSDP2 logging)",https://github.com/pytorch/pytorch/issues/104247,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label,route to design/research decision before patch planning,2025-07-29T12:40:35Z
issue,158704,Enable DataLoaders to prefetch batches across epochs,https://github.com/pytorch/pytorch/issues/158704,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: dataloader,route to design/research decision before patch planning,2025-07-23T18:44:54Z
issue,69431,"[feature request] `torch.to(obj, device, dtype)` supporting recursive lists/dicts/tuples of tensors probably by uplifting/promoting `torch.distributed.utils._recursive_to`",https://github.com/pytorch/pytorch/issues/69431,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label,route to design/research decision before patch planning,2025-07-19T20:30:25Z
issue,157547,Several `torch.*` functions raise uninformative `NotImplementedError`s when called with integer `dtype`,https://github.com/pytorch/pytorch/issues/157547,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: error checking,route to design/research decision before patch planning,2025-07-14T17:18:48Z
issue,150742,Install pytorch from pypi using local CUDA build,https://github.com/pytorch/pytorch/issues/150742,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: ci, oncall: releng",route to design/research decision before patch planning,2025-07-09T17:01:38Z
issue,61654,Allow creation of pseudo devices for testing purposes,https://github.com/pytorch/pytorch/issues/61654,good volunteer slice,20,label says issue is suitable for a scoped external contribution; old update age or low-urgency/stale label; component/oncall labels: module: internals,scan when time allows,2025-07-09T13:55:56Z
issue,141627,Support reductions in FlexAttention's score_mod/mask_mod,https://github.com/pytorch/pytorch/issues/141627,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: flex attention, module: higher order operators, module: pt2-dispatcher, oncall: pt2",route to design/research decision before patch planning,2025-07-08T00:06:21Z
issue,154260,Pypi Support for Windows arm64,https://github.com/pytorch/pytorch/issues/154260,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: arm, module: windows, oncall: releng",route to design/research decision before patch planning,2025-06-30T12:12:45Z
issue,155983,as_tensor of list of tensors should keep grad history,https://github.com/pytorch/pytorch/issues/155983,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: autograd,route to design/research decision before patch planning,2025-06-28T17:22:17Z
issue,154767,Turn gradient_accumulate into a separate aten op,https://github.com/pytorch/pytorch/issues/154767,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: autograd,route to design/research decision before patch planning,2025-06-28T15:07:49Z
issue,152877,Motivate Pytorch's forward mode AD APIs with training examples,https://github.com/pytorch/pytorch/issues/152877,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: docs, module: forward ad",route to design/research decision before patch planning,2025-06-28T15:07:02Z
issue,155986,Vulkan interoperability,https://github.com/pytorch/pytorch/issues/155986,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: vulkan,route to design/research decision before patch planning,2025-06-17T17:10:23Z
issue,154906,load_state_dict with strict should be able to remove_duplicate?,https://github.com/pytorch/pytorch/issues/154906,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: nn,route to design/research decision before patch planning,2025-06-13T21:45:37Z
issue,155368,Change error to warnning when tensor size mismatch while loading params,https://github.com/pytorch/pytorch/issues/155368,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: nn,route to design/research decision before patch planning,2025-06-13T19:21:33Z
issue,155878,RFC: Add Adaptive Entropy-Gated Contrastive Fusion (AECF) for Robust Multimodal Attention Pooling,https://github.com/pytorch/pytorch/issues/155878,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: nn, module: sdpa",route to design/research decision before patch planning,2025-06-13T15:47:43Z
issue,122680,Please consider adding MIG (MI-rror with G-radient modification) to torch.nn,https://github.com/pytorch/pytorch/issues/122680,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: nn,route to design/research decision before patch planning,2025-06-08T10:32:43Z
issue,111441,torch.compile of simple loop takes 34 seconds,https://github.com/pytorch/pytorch/issues/111441,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: compile-time, module: higher order operators, module: pt2-dispatcher, oncall: pt2",route to design/research decision before patch planning,2025-06-04T02:23:26Z
issue,150698,[Feature Request] Memory optimization for backward propagation in GPU,https://github.com/pytorch/pytorch/issues/150698,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: memory usage",route to design/research decision before patch planning,2025-06-03T19:08:45Z
issue,149133,Implement einsum backprop rather than decomposing,https://github.com/pytorch/pytorch/issues/149133,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: python frontend",route to design/research decision before patch planning,2025-06-03T15:52:54Z
issue,115937,Building extensions with CMake,https://github.com/pytorch/pytorch/issues/115937,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: build, module: cpp-extensions",route to design/research decision before patch planning,2025-05-27T12:52:27Z
issue,153854,Better padding API,https://github.com/pytorch/pytorch/issues/153854,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: nn, module: padding",route to design/research decision before patch planning,2025-05-20T01:06:22Z
issue,153566,Improve PyTorch Logging to Distinguish ROCm from CUDA,https://github.com/pytorch/pytorch/issues/153566,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: rocm,route to design/research decision before patch planning,2025-05-15T20:23:45Z
issue,152875,Incorporate CUDA Memory Trimming Into DeviceCachingAllocator,https://github.com/pytorch/pytorch/issues/152875,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: windows",route to design/research decision before patch planning,2025-05-13T23:10:32Z
issue,153378,MPS backend silently ignores dimension mismatch,https://github.com/pytorch/pytorch/issues/153378,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: error checking, module: mps",route to design/research decision before patch planning,2025-05-13T02:03:01Z
issue,118225,Support for banded matrix operations,https://github.com/pytorch/pytorch/issues/118225,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: linear algebra,route to design/research decision before patch planning,2025-05-12T08:14:06Z
issue,153035,Add Split Softmax,https://github.com/pytorch/pytorch/issues/153035,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: nn,route to design/research decision before patch planning,2025-05-07T19:12:02Z
issue,150432,pytorch pip install instructions: always include the cuda index,https://github.com/pytorch/pytorch/issues/150432,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: docs, oncall: releng",route to design/research decision before patch planning,2025-05-04T14:50:30Z
issue,151722,Add maxcount Parameter to torch.unique and torch.unique_consecutive,https://github.com/pytorch/pytorch/issues/151722,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: performance, module: python frontend",route to design/research decision before patch planning,2025-04-28T18:40:11Z
issue,101699,[discussion] [feature request] Native tensor-backed array of strings (at least a read-only one) and basic string processing functions for addition into core + discussion of extremely basic data frames (also for reducing python object heap pressure),https://github.com/pytorch/pytorch/issues/101699,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label,route to design/research decision before patch planning,2025-04-16T08:16:38Z
issue,117806,Supporting custom attributes with `__torch_function__` tensor subclasses,https://github.com/pytorch/pytorch/issues/117806,good volunteer slice,20,"label says issue is suitable for a scoped external contribution; old update age or low-urgency/stale label; component/oncall labels: module: __torch_function__, module: dynamo",scan when time allows,2025-04-11T21:54:41Z
issue,25883,Python/C++ API Parity: torch.nn modules and functional,https://github.com/pytorch/pytorch/issues/25883,good volunteer slice,20,"label says issue is suitable for a scoped external contribution; old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: nn",scan when time allows,2025-04-11T20:00:45Z
issue,149118,Add dim parameter to torch.bucketize,https://github.com/pytorch/pytorch/issues/149118,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: python frontend,route to design/research decision before patch planning,2025-04-09T20:33:12Z
issue,4132,x.grad should be 0 but get NaN after x/0,https://github.com/pytorch/pytorch/issues/4132,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: docs, module: nans and infs",route to design/research decision before patch planning,2025-04-01T14:38:32Z
issue,132446,[Feature Request] Support Custom CPU Allocator to Enable torch.cpu.change_current_allocator,https://github.com/pytorch/pytorch/issues/132446,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: cpu,route to design/research decision before patch planning,2025-04-01T07:36:13Z
issue,148843,make `to_empty` a no-op if parameter/buffer already on `device`,https://github.com/pytorch/pytorch/issues/148843,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: nn,route to design/research decision before patch planning,2025-03-15T20:41:29Z
issue,144228,grad_fn function disobeys broadcast rules,https://github.com/pytorch/pytorch/issues/144228,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: autograd,route to design/research decision before patch planning,2025-03-04T18:27:51Z
issue,148345,Installation of `pytorch==2.6.0+cu124` doesn't install `triton` and `nvidia` libraries,https://github.com/pytorch/pytorch/issues/148345,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: binaries,route to design/research decision before patch planning,2025-03-03T18:10:28Z
issue,147437,"To avoid default eps ineffective in some precision(float16), dynamically set the default value of eps in optimizers such as Adam based on precision.",https://github.com/pytorch/pytorch/issues/147437,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: optimizer,route to design/research decision before patch planning,2025-02-26T16:24:01Z
issue,146980,Feature Request: Add dlfloat16 support as a new dtype,https://github.com/pytorch/pytorch/issues/146980,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label,route to design/research decision before patch planning,2025-02-16T15:45:20Z
issue,147048,Feature Request: rsample for Von Mises distribution,https://github.com/pytorch/pytorch/issues/147048,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: distributions,route to design/research decision before patch planning,2025-02-13T18:23:37Z
issue,144159,Allow generic python data structure input for torch.autograd.Function,https://github.com/pytorch/pytorch/issues/144159,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: autograd,route to design/research decision before patch planning,2025-02-04T01:51:28Z
issue,145739,No broadcasting by default.,https://github.com/pytorch/pytorch/issues/145739,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: python frontend,route to design/research decision before patch planning,2025-02-03T19:09:13Z
issue,144536,[bug report template format] Simplify version information with HTML tags,https://github.com/pytorch/pytorch/issues/144536,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: collect_env.py,route to design/research decision before patch planning,2025-01-14T02:16:08Z
issue,120589,Softmax plus one for Flash attention,https://github.com/pytorch/pytorch/issues/120589,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: sdpa,route to design/research decision before patch planning,2025-01-12T08:32:20Z
issue,84782,macOS Pyinstaller: libc++abi: terminating with uncaught exception of type c10::Error: Type c10::intrusive_ptr<ConvPackedParamsBase<2>> could not be converted to any of the known types,https://github.com/pytorch/pytorch/issues/84782,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: cpp-extensions, module: m1, module: third_party",route to design/research decision before patch planning,2025-01-07T19:23:23Z
issue,136433,SAC doesn't support nesting with different recompute plans,https://github.com/pytorch/pytorch/issues/136433,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: activation checkpointing, module: autograd",route to design/research decision before patch planning,2025-01-06T02:16:57Z
issue,138118,[autograd] do not clear all variables when retain_graph=False,https://github.com/pytorch/pytorch/issues/138118,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: autograd,route to design/research decision before patch planning,2025-01-06T02:16:07Z
issue,140378,Differentiable quadratic programing solver,https://github.com/pytorch/pytorch/issues/140378,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: autograd,route to design/research decision before patch planning,2025-01-06T02:06:23Z
issue,142320,Proposal for Official Cross-Platform Support for 3D Gaussian Splatting,https://github.com/pytorch/pytorch/issues/142320,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: nn,route to design/research decision before patch planning,2024-12-23T18:08:35Z
issue,120189,Making Mamba first-class citizen in PyTorch,https://github.com/pytorch/pytorch/issues/120189,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: higher order operators, module: pt2-dispatcher, oncall: pt2",route to design/research decision before patch planning,2024-12-14T22:33:49Z
issue,142381,Add a linter for Docker images which are not included in docker-builds.yml,https://github.com/pytorch/pytorch/issues/142381,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: ci,route to design/research decision before patch planning,2024-12-09T17:11:20Z
issue,141678,torch.cuda adds support for the changing `CUDA_VISIBLE_DEVICES`,https://github.com/pytorch/pytorch/issues/141678,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: third_party",route to design/research decision before patch planning,2024-11-30T20:36:52Z
issue,44945,Fuse softmax and masking in MultiheadAttention,https://github.com/pytorch/pytorch/issues/44945,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: nn,route to design/research decision before patch planning,2024-11-27T19:47:15Z
issue,35458,Shared-QK transformer for the transformer (nn.activation.MultiheadAttention) module in PyTorch?,https://github.com/pytorch/pytorch/issues/35458,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: nn,route to design/research decision before patch planning,2024-11-27T19:47:12Z
issue,74363,"Efficient very large (e.g., 31x31) depth-wise convolution",https://github.com/pytorch/pytorch/issues/74363,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: convolution,route to design/research decision before patch planning,2024-11-20T05:33:00Z
issue,50566,Enhance supported fill value type for constant pad in functional.pad,https://github.com/pytorch/pytorch/issues/50566,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: nn, module: padding",route to design/research decision before patch planning,2024-11-10T01:44:00Z
issue,138952,[Feature request] Flag to prevent certain parameters from being typecast by nn.module.to(),https://github.com/pytorch/pytorch/issues/138952,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: nn,route to design/research decision before patch planning,2024-11-08T00:39:28Z
issue,138502,Contribution Proposal: ScatterND Implementation for PyTorch,https://github.com/pytorch/pytorch/issues/138502,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: scatter & gather ops,route to design/research decision before patch planning,2024-11-04T18:14:38Z
issue,111538,Propose to add constant padding mode to the `torch.nn.functional.grid_sample` function,https://github.com/pytorch/pytorch/issues/111538,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: nn,route to design/research decision before patch planning,2024-11-03T03:22:28Z
issue,106360,Improving save_on_cpu's performance by overlapping memory transfers with compute,https://github.com/pytorch/pytorch/issues/106360,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: autograd, oncall: distributed",route to design/research decision before patch planning,2024-10-31T15:58:23Z
issue,71386,Direct Implementation of K-Nearest neighbor (KNN) in pytorch,https://github.com/pytorch/pytorch/issues/71386,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: nn,route to design/research decision before patch planning,2024-10-29T05:23:51Z
issue,138906,"RuntimeError: ""addmm_cuda"" not implemented for 'ComplexHalf'",https://github.com/pytorch/pytorch/issues/138906,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: complex, module: cuda",route to design/research decision before patch planning,2024-10-28T15:46:40Z
issue,137680,[FEA] Add nvsci support in pytorch multiprocessing module for nvidia jetson devices.,https://github.com/pytorch/pytorch/issues/137680,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda,route to design/research decision before patch planning,2024-10-17T06:27:09Z
issue,135816,Can we remove @torch._dynamo_disable from torch.utils.checkpoint.checkpoint?,https://github.com/pytorch/pytorch/issues/135816,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",route to design/research decision before patch planning,2024-09-24T06:12:00Z
issue,113530,"Make ""torch.load"" multi threaded process",https://github.com/pytorch/pytorch/issues/113530,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: serialization,route to design/research decision before patch planning,2024-09-20T16:59:41Z
issue,43101,New function for CTC decoding,https://github.com/pytorch/pytorch/issues/43101,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: nn,route to design/research decision before patch planning,2024-09-14T12:00:32Z
issue,107224,"[CPP API] Add Adadelta, Adamax, ASGD, NAdam, RAdam and Rprop",https://github.com/pytorch/pytorch/issues/107224,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: optimizer",route to design/research decision before patch planning,2024-09-01T18:44:21Z
issue,133288,Give PrivateUse2 and PrivateUse3 the same treatment as PrivateUse1,https://github.com/pytorch/pytorch/issues/133288,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: backend, module: privateuse1",route to design/research decision before patch planning,2024-08-29T14:17:01Z
issue,133566,Performance: Autodiff stores inputs of convolution layers with non-trainable weight,https://github.com/pytorch/pytorch/issues/133566,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: memory usage",route to design/research decision before patch planning,2024-08-29T13:44:17Z
issue,134325,Proposal: When serializing to state_dict/parameters use different separators for ParameterList and ParameterDict,https://github.com/pytorch/pytorch/issues/134325,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: nn, module: serialization",route to design/research decision before patch planning,2024-08-26T22:43:19Z
issue,133466,"In-place operations should be deprecated and in the future, they should be removed from PyTorch.",https://github.com/pytorch/pytorch/issues/133466,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: nn,route to design/research decision before patch planning,2024-08-16T20:35:14Z
issue,132378,per sample gradient with torch.autograd.grad,https://github.com/pytorch/pytorch/issues/132378,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: vmap",route to design/research decision before patch planning,2024-08-13T22:19:06Z
issue,129867,`saved_tensors_hooks`/`_pop_saved_tensors_default_hooks` API extension,https://github.com/pytorch/pytorch/issues/129867,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: activation checkpointing, module: autograd, module: multithreading",route to design/research decision before patch planning,2024-08-13T15:38:13Z
issue,105465,[proposal] Bit ops: e.g. setbit/getbit/togglebit/byteswap,https://github.com/pytorch/pytorch/issues/105465,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: python frontend,route to design/research decision before patch planning,2024-08-08T20:48:17Z
issue,131881,Decouple some generic APIs with CUDA building,https://github.com/pytorch/pytorch/issues/131881,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda,route to design/research decision before patch planning,2024-08-08T07:00:36Z
issue,128380,Conformal Prediction framework to enhance reliability in risk sensitive industrial applications,https://github.com/pytorch/pytorch/issues/128380,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label,route to design/research decision before patch planning,2024-08-07T20:46:50Z
issue,131927,exclude_dim=[...] as a keyword argument to reduce type operations,https://github.com/pytorch/pytorch/issues/131927,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: reductions,route to design/research decision before patch planning,2024-07-29T04:35:01Z
issue,128283,control output type in matmul,https://github.com/pytorch/pytorch/issues/128283,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label,route to design/research decision before patch planning,2024-07-25T13:20:01Z
issue,130428,Extend the device context manager to support existing tensors as well,https://github.com/pytorch/pytorch/issues/130428,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: python frontend,route to design/research decision before patch planning,2024-07-16T16:47:03Z
issue,130480,[Feature Request] Lazy upcasting for mmap'd state dicts,https://github.com/pytorch/pytorch/issues/130480,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: nn, module: serialization",route to design/research decision before patch planning,2024-07-11T20:07:11Z
issue,127996,Tensor subclass constructor via autograd function is error prone re mutation,https://github.com/pytorch/pytorch/issues/127996,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: autograd,route to design/research decision before patch planning,2024-06-28T16:12:42Z
issue,120107,[feature request] keep gradcheck numerical jacobian from being depeciated,https://github.com/pytorch/pytorch/issues/120107,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: autograd,route to design/research decision before patch planning,2024-06-28T13:39:47Z
issue,128793,"GradType: a subset of dtype that is differentiable, containing all float and complex dtypes",https://github.com/pytorch/pytorch/issues/128793,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: docs",route to design/research decision before patch planning,2024-06-27T20:11:38Z
issue,128934,version inquiry,https://github.com/pytorch/pytorch/issues/128934,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: serialization,route to design/research decision before patch planning,2024-06-27T09:30:36Z
issue,50122,"[proposal] Pseudo-functions to support common gradient patch use-cases like replacing inf / nan, gradient clipping / reversal",https://github.com/pytorch/pytorch/issues/50122,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: autograd,route to design/research decision before patch planning,2024-06-25T19:19:52Z
issue,124902,Add support for IPC features for PrivateUse devices,https://github.com/pytorch/pytorch/issues/124902,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: multiprocessing, module: privateuse1",route to design/research decision before patch planning,2024-06-06T11:32:19Z
issue,127932,Add SinusoidalPositionalEmbedding module for use in Transformers and Diffusion models,https://github.com/pytorch/pytorch/issues/127932,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: nn,route to design/research decision before patch planning,2024-06-05T02:20:50Z
issue,127699,torch.Tensor for optimizer parameters,https://github.com/pytorch/pytorch/issues/127699,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: optimizer,route to design/research decision before patch planning,2024-06-03T19:52:37Z
issue,127284,Fused AdamW maybe should accept lr_dict directly?,https://github.com/pytorch/pytorch/issues/127284,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: optimizer,route to design/research decision before patch planning,2024-05-28T14:58:02Z
issue,126762,Feature request to add the SReLU activation function,https://github.com/pytorch/pytorch/issues/126762,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: nn,route to design/research decision before patch planning,2024-05-22T14:23:28Z
issue,125767,RFC: Integrate Arm Compute Library (ACL) into PyTorch as a submodule and add build support in setup.py,https://github.com/pytorch/pytorch/issues/125767,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: arm, module: build",route to design/research decision before patch planning,2024-05-10T23:44:51Z
issue,121791,[RFC] Set Float32 Precision for CONV/RNN,https://github.com/pytorch/pytorch/issues/121791,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label,route to design/research decision before patch planning,2024-05-08T06:45:02Z
issue,125223,torch.uniform_() is single-threaded on CPU,https://github.com/pytorch/pytorch/issues/125223,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: distributions, module: nn, module: performance",route to design/research decision before patch planning,2024-05-06T01:52:56Z
issue,58122,Generate special clear error messages for known common misuses in TorchScript,https://github.com/pytorch/pytorch/issues/58122,good volunteer slice,20,label says issue is suitable for a scoped external contribution; old update age or low-urgency/stale label; component/oncall labels: module: bootcamp,scan when time allows,2024-04-21T17:55:38Z
issue,49227,Looking for a more convenient way to getter value of the upper/lower triangular matrix into 1D,https://github.com/pytorch/pytorch/issues/49227,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: bc-breaking, module: numpy",route to design/research decision before patch planning,2024-04-15T14:06:34Z
issue,63265,torch.scatter_: support index.size(d) > src.size(d),https://github.com/pytorch/pytorch/issues/63265,good volunteer slice,20,label says issue is suitable for a scoped external contribution; old update age or low-urgency/stale label; component/oncall labels: module: scatter & gather ops,scan when time allows,2024-04-03T02:24:43Z
issue,47055,"[types] torch.utils.data.{Dataset, Sampler} are not Sized",https://github.com/pytorch/pytorch/issues/47055,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: data,route to design/research decision before patch planning,2024-04-02T15:54:07Z
issue,122551,Global Response Normalization (GRN),https://github.com/pytorch/pytorch/issues/122551,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: nn,route to design/research decision before patch planning,2024-03-25T15:23:00Z
issue,121933,Please provide `total_repeat_length` field to PyTorch `repeat` / `repeat_interleave` API,https://github.com/pytorch/pytorch/issues/121933,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: tensor creation,route to design/research decision before patch planning,2024-03-20T16:46:51Z
issue,122099,No reduction of disk usage on slices of torch tensor,https://github.com/pytorch/pytorch/issues/122099,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: serialization,route to design/research decision before patch planning,2024-03-19T17:17:42Z
issue,119475,`autograd.Function` leads to `pylint` warnings in valid user code,https://github.com/pytorch/pytorch/issues/119475,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: typing",route to design/research decision before patch planning,2024-03-19T15:05:31Z
issue,109052,Allow reductions to write into pinned memory,https://github.com/pytorch/pytorch/issues/109052,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: memory usage, module: reductions",route to design/research decision before patch planning,2024-03-14T02:24:02Z
issue,58228,"""_amp_foreach_non_finite_check_and_unscale_cuda"" not implemented for 'ComplexFloat'",https://github.com/pytorch/pytorch/issues/58228,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision), module: complex, module: cuda",route to design/research decision before patch planning,2024-03-10T09:07:08Z
issue,120700,Invert the model architecture from the unknown neural network model (the python object of the model),https://github.com/pytorch/pytorch/issues/120700,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: nn,route to design/research decision before patch planning,2024-03-01T08:14:44Z
issue,120072,Please consider adding more optimization support for multi-layer recurrent neural networks to improve fitting speed.,https://github.com/pytorch/pytorch/issues/120072,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: nn, module: rnn",route to design/research decision before patch planning,2024-02-23T21:57:56Z
issue,100628,Feature request: fast way to approximate the diagonal of the hessian,https://github.com/pytorch/pytorch/issues/100628,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: functorch",route to design/research decision before patch planning,2024-02-21T11:13:19Z
issue,114462,Transformer with convolutional network,https://github.com/pytorch/pytorch/issues/114462,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: nn,route to design/research decision before patch planning,2024-02-13T23:54:42Z
issue,102447,"Add additional ""sigmoid"" approximation to GeLu activation?",https://github.com/pytorch/pytorch/issues/102447,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: nn,route to design/research decision before patch planning,2024-02-08T13:46:03Z
issue,110761,Using ChainedScheduler with ReduceLROnPlateau leads to unexpected keyword argument error,https://github.com/pytorch/pytorch/issues/110761,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: lrscheduler, module: optimizer",route to design/research decision before patch planning,2024-02-07T11:10:11Z
issue,20822,"[Proposal] Data reading framework for PyTorch (Hive, MySQL, S3 etc.)",https://github.com/pytorch/pytorch/issues/20822,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: dataloader,route to design/research decision before patch planning,2024-02-05T23:35:05Z
issue,68879,Stop gradient option for padding,https://github.com/pytorch/pytorch/issues/68879,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: nn",route to design/research decision before patch planning,2024-02-02T15:50:13Z
issue,118133,Better test disabling system,https://github.com/pytorch/pytorch/issues/118133,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: devx,route to design/research decision before patch planning,2024-02-01T14:05:00Z
issue,99278,Build error on libstc++ header stl_alogbase.h on riscv,https://github.com/pytorch/pytorch/issues/99278,good volunteer slice,20,label says issue is suitable for a scoped external contribution; old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2024-01-21T11:52:05Z
issue,117217,"BatchNorm1d, running_mean and running_var",https://github.com/pytorch/pytorch/issues/117217,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: nn,route to design/research decision before patch planning,2024-01-11T22:18:05Z
issue,72831,Toggling deterministic mode for individual autograd backward functions,https://github.com/pytorch/pytorch/issues/72831,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: determinism",route to design/research decision before patch planning,2024-01-10T18:32:09Z
issue,110295,Tests modify global state cause later tests to fail,https://github.com/pytorch/pytorch/issues/110295,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: ci, module: devx",route to design/research decision before patch planning,2024-01-09T21:53:06Z
issue,115993,[FR] Add a way to reserve memory that survives `torch.cuda.empty_cache()`,https://github.com/pytorch/pytorch/issues/115993,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: memory usage",route to design/research decision before patch planning,2024-01-08T23:34:36Z
issue,116555,The performance of `torch.index_select` and regular indexing differs dramatically based on the size of the tensor it is indexing,https://github.com/pytorch/pytorch/issues/116555,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: advanced indexing, module: autograd, module: cpu, module: performance",route to design/research decision before patch planning,2024-01-08T18:58:39Z
issue,116354,Abstract torch.device for GPU/NE/TPU computations in the cloud-based agent,https://github.com/pytorch/pytorch/issues/116354,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label,route to design/research decision before patch planning,2024-01-08T18:37:47Z
issue,115852,`groups` parameter for `nn.Linear`,https://github.com/pytorch/pytorch/issues/115852,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: nn,route to design/research decision before patch planning,2024-01-08T13:32:33Z
issue,116490,Legitimize/harden differentiable optimizers in torch.optim,https://github.com/pytorch/pytorch/issues/116490,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: optimizer",route to design/research decision before patch planning,2023-12-28T07:38:44Z
issue,105255,[discussion] Integrate widely used utilities from fvcore into the core repo,https://github.com/pytorch/pytorch/issues/105255,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: lrscheduler, module: nn, oncall: distributed",route to design/research decision before patch planning,2023-12-22T00:11:54Z
issue,115273,Consider improving project management (12 pages of HighPriority tickets > 1 year old),https://github.com/pytorch/pytorch/issues/115273,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: devx,route to design/research decision before patch planning,2023-12-18T18:08:24Z
issue,115588,`torch.nn.KLDivLoss` fails `test_grad` and `test_gradgrad`,https://github.com/pytorch/pytorch/issues/115588,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: nn,route to design/research decision before patch planning,2023-12-12T21:59:19Z
issue,115585,`torch.nn.CTCLoss` doesn't seem to properly support noncontiguous inputs for 1d `target` with `target_dtype=torch.long`,https://github.com/pytorch/pytorch/issues/115585,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: nn,route to design/research decision before patch planning,2023-12-11T20:52:25Z
issue,108491,Suppport Fused AdamW on CPU,https://github.com/pytorch/pytorch/issues/108491,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: optimizer, module: performance",route to design/research decision before patch planning,2023-12-01T21:02:56Z
issue,114650,Error when calculating the Jacobian of torch.conj using forward-mode differentiation,https://github.com/pytorch/pytorch/issues/114650,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: complex",route to design/research decision before patch planning,2023-11-29T01:15:40Z
issue,40316,Refactor the adaptive avg pool code,https://github.com/pytorch/pytorch/issues/40316,good volunteer slice,20,"label says issue is suitable for a scoped external contribution; old update age or low-urgency/stale label; component/oncall labels: module: pooling, oncall: quantization",scan when time allows,2023-11-17T22:22:33Z
issue,113730,Can the official provide cpu offload method for large model inference?,https://github.com/pytorch/pytorch/issues/113730,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label,route to design/research decision before patch planning,2023-11-17T17:33:19Z
issue,110885,Depthwise conv3d slower than normal conv3d,https://github.com/pytorch/pytorch/issues/110885,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: nn, module: performance",route to design/research decision before patch planning,2023-11-03T18:53:06Z
issue,102201,scipy.ndimage.find_objects,https://github.com/pytorch/pytorch/issues/102201,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: nn,route to design/research decision before patch planning,2023-11-01T20:01:51Z
issue,16663,new_zeros is not traced correctly,https://github.com/pytorch/pytorch/issues/16663,good volunteer slice,20,label says issue is suitable for a scoped external contribution; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2023-11-01T17:43:43Z
issue,111563,"Higher-order derivatives extremely slow, increasing exponentially",https://github.com/pytorch/pytorch/issues/111563,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: autograd,route to design/research decision before patch planning,2023-11-01T05:36:13Z
issue,112352,RFC: [pytree] node registration namespaces,https://github.com/pytorch/pytorch/issues/112352,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: pytree,route to design/research decision before patch planning,2023-10-30T18:31:46Z
issue,110780,backward and grad behave inconsistently w.r.t. set_ on leaf variable,https://github.com/pytorch/pytorch/issues/110780,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: autograd,route to design/research decision before patch planning,2023-10-09T17:01:06Z
issue,92330,Simplify module backward hooks to use multi-grad hooks instead,https://github.com/pytorch/pytorch/issues/92330,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: bc-breaking, module: nn",route to design/research decision before patch planning,2023-10-05T22:31:45Z
issue,108090,[Optimizer Perf] Improve speed of _init_group to c++,https://github.com/pytorch/pytorch/issues/108090,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: optimizer, module: performance",route to design/research decision before patch planning,2023-10-05T15:10:28Z
issue,87734,Checkpointing Support for Modularized Optimizers,https://github.com/pytorch/pytorch/issues/87734,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: optimizer,route to design/research decision before patch planning,2023-10-04T21:58:58Z
issue,97469,Improve collectives fingerprinting,https://github.com/pytorch/pytorch/issues/97469,good volunteer slice,20,label says issue is suitable for a scoped external contribution; old update age or low-urgency/stale label; component/oncall labels: module: c10d,scan when time allows,2023-09-26T10:12:30Z
issue,109948,Simple script segfaulting when grad is enabled,https://github.com/pytorch/pytorch/issues/109948,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: edge cases",route to design/research decision before patch planning,2023-09-25T16:16:11Z
issue,71479,[numpy compat] torch.stack and torch.tensor doesn't support nested list+tensors (NumPy does support) - at least document the difference in the error message,https://github.com/pytorch/pytorch/issues/71479,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: numpy, module: viewing and reshaping",route to design/research decision before patch planning,2023-09-23T18:53:56Z
issue,103449,Asynchronous CUDA AveragedModel,https://github.com/pytorch/pytorch/issues/103449,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: optimizer,route to design/research decision before patch planning,2023-09-22T18:18:52Z
issue,109706,Make standard container classes satisfy container Protocols.,https://github.com/pytorch/pytorch/issues/109706,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: nn,route to design/research decision before patch planning,2023-09-22T18:04:05Z
issue,109582,Make Dropout take a dim=... argument,https://github.com/pytorch/pytorch/issues/109582,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: nn,route to design/research decision before patch planning,2023-09-22T16:27:39Z
issue,108692,Adding Maximal Update Parametrization (µP) to torch.nn.init,https://github.com/pytorch/pytorch/issues/108692,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: nn,route to design/research decision before patch planning,2023-09-14T14:56:07Z
issue,108784,Collect more env variables in `collect_env.py`,https://github.com/pytorch/pytorch/issues/108784,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label,route to design/research decision before patch planning,2023-09-12T11:14:52Z
issue,66813,Use `__slots__` for the `nn.Module` class,https://github.com/pytorch/pytorch/issues/66813,good volunteer slice,20,"label says issue is suitable for a scoped external contribution; old update age or low-urgency/stale label; component/oncall labels: module: nn, module: performance",scan when time allows,2023-08-28T14:13:25Z
issue,67087,Feature Request: 1d grid sample,https://github.com/pytorch/pytorch/issues/67087,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: nn,route to design/research decision before patch planning,2023-08-26T04:19:28Z
issue,62902,distributed/elastic/multiprocessing/api_test is flaky on ROCm,https://github.com/pytorch/pytorch/issues/62902,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: flaky-tests, module: rocm, oncall: r2p",route to design/research decision before patch planning,2023-08-22T21:32:48Z
issue,106784,"[ux] Suppot torch.tensor(set([1,2,3]))",https://github.com/pytorch/pytorch/issues/106784,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: python frontend,route to design/research decision before patch planning,2023-08-21T18:27:56Z
issue,105365,Quadric Layer,https://github.com/pytorch/pytorch/issues/105365,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: nn,route to design/research decision before patch planning,2023-08-09T20:19:27Z
issue,96319,views created in __torch_dispatch__ share storage but not version_counter,https://github.com/pytorch/pytorch/issues/96319,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: __torch_dispatch__, module: autograd, module: molly-guard",route to design/research decision before patch planning,2023-07-24T15:20:16Z
issue,39765,Support mainstream pruning techniques,https://github.com/pytorch/pytorch/issues/39765,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: pruning,route to design/research decision before patch planning,2023-07-14T12:23:52Z
issue,102337,Tensors that share same underlying storage to also share gradient storage,https://github.com/pytorch/pytorch/issues/102337,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: python frontend",route to design/research decision before patch planning,2023-07-13T14:37:49Z
issue,103588,LSTM/RNN operation agnostic,https://github.com/pytorch/pytorch/issues/103588,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: nn,route to design/research decision before patch planning,2023-07-07T16:15:11Z
issue,102417,_view_func but without keeping original view tensor alive,https://github.com/pytorch/pytorch/issues/102417,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: viewing and reshaping",route to design/research decision before patch planning,2023-06-23T15:24:01Z
issue,96764,Improve checkpoint thread-safety,https://github.com/pytorch/pytorch/issues/96764,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: multithreading",route to design/research decision before patch planning,2023-06-23T15:22:12Z
issue,96579,Wrong return type from operation on custom tensor inside registered hook,https://github.com/pytorch/pytorch/issues/96579,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: __torch_function__, module: autograd",route to design/research decision before patch planning,2023-06-23T15:16:09Z
issue,103109,Increased / more verbose type aliases for improved readability of user defined content,https://github.com/pytorch/pytorch/issues/103109,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: typing,route to design/research decision before patch planning,2023-06-19T20:22:13Z
issue,102905,nn.ChannelShuffle1d,https://github.com/pytorch/pytorch/issues/102905,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: nn,route to design/research decision before patch planning,2023-06-09T19:20:49Z
issue,102730,Best practices clarification for initialization strategies,https://github.com/pytorch/pytorch/issues/102730,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: initialization, module: nn, module: python frontend",route to design/research decision before patch planning,2023-06-09T19:12:08Z
issue,10756,Unexpected Behavior when Pointwise Operations Write to Expanded Tensors,https://github.com/pytorch/pytorch/issues/10756,good volunteer slice,20,"label says issue is suitable for a scoped external contribution; old update age or low-urgency/stale label; component/oncall labels: module: internals, module: partial aliasing",scan when time allows,2023-06-07T22:36:05Z
issue,102938,Support for efficiently processing categorical distributions with varying dimensions,https://github.com/pytorch/pytorch/issues/102938,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: distributions,route to design/research decision before patch planning,2023-06-05T15:30:35Z
issue,95374,Add oscillating activation functions to PyTorch.,https://github.com/pytorch/pytorch/issues/95374,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: loss,route to design/research decision before patch planning,2023-05-31T00:04:45Z
issue,44954,Optimization with constraints for torch.optim,https://github.com/pytorch/pytorch/issues/44954,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: optimizer,route to design/research decision before patch planning,2023-05-13T12:07:01Z
issue,100990,Extending compatibility of LibTorch,https://github.com/pytorch/pytorch/issues/100990,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: cpp,route to design/research decision before patch planning,2023-05-10T01:51:22Z
issue,48830,"Please add ""dim"" feature for function ""torch.masked_select""",https://github.com/pytorch/pytorch/issues/48830,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: numpy, module: sorting and selection",route to design/research decision before patch planning,2023-04-21T16:58:13Z
issue,83800,torch.var_mean is slower than layer norm,https://github.com/pytorch/pytorch/issues/83800,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: nn, module: performance",route to design/research decision before patch planning,2023-04-14T18:17:41Z
issue,97788,Add SSIM as Loss Function,https://github.com/pytorch/pytorch/issues/97788,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: loss,route to design/research decision before patch planning,2023-03-31T21:10:15Z
issue,98089,GroupNorm cpu/gpu parity tests fail with pretty large differences,https://github.com/pytorch/pytorch/issues/98089,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: nn,route to design/research decision before patch planning,2023-03-31T21:04:04Z
issue,97248,"A more systematic API for resolving the ""vmap-incompatible in-place operation"" error",https://github.com/pytorch/pytorch/issues/97248,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: functorch,route to design/research decision before patch planning,2023-03-21T22:35:35Z
issue,96555,early stopping,https://github.com/pytorch/pytorch/issues/96555,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label,route to design/research decision before patch planning,2023-03-14T17:53:20Z
issue,95916,autograd.functional.jacobian : tensor instead of function as input for reverse mode?,https://github.com/pytorch/pytorch/issues/95916,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: autograd,route to design/research decision before patch planning,2023-03-10T22:50:34Z
issue,95337,Changing behavior of module.to() to better support mixed real- and complex-valued parameters,https://github.com/pytorch/pytorch/issues/95337,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: complex, module: nn",route to design/research decision before patch planning,2023-03-03T20:11:58Z
issue,88325,Group losses in a common namespace,https://github.com/pytorch/pytorch/issues/88325,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: loss, module: nn",route to design/research decision before patch planning,2023-03-03T20:09:25Z
issue,93684,Can we rewrite numpy operators to pytorch operators?,https://github.com/pytorch/pytorch/issues/93684,good volunteer slice,20,label says issue is suitable for a scoped external contribution; old update age or low-urgency/stale label; component/oncall labels: module: numpy,scan when time allows,2023-02-21T15:31:52Z
issue,94904,"Optimizer ""Lion"" in Symbolic Discovery of Optimization Algorithms",https://github.com/pytorch/pytorch/issues/94904,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: optimizer,route to design/research decision before patch planning,2023-02-17T15:33:50Z
issue,84751,Allow passing dict (as opposed to OrderedDict) to nn.Sequential,https://github.com/pytorch/pytorch/issues/84751,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: nn,route to design/research decision before patch planning,2023-02-10T20:31:34Z
issue,62148,[feature request] Function to broadcast right,https://github.com/pytorch/pytorch/issues/62148,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: viewing and reshaping,route to design/research decision before patch planning,2023-02-06T23:24:59Z
issue,93154,Support for VeLO optimizer.,https://github.com/pytorch/pytorch/issues/93154,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: optimizer,route to design/research decision before patch planning,2023-02-03T20:06:26Z
issue,84524,nn.Softmax should not allow default/implicit/unset dim constructor argument,https://github.com/pytorch/pytorch/issues/84524,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: deprecation, module: nn",route to design/research decision before patch planning,2023-01-27T21:03:08Z
issue,89724,Sample Weighted BatchNorm1d,https://github.com/pytorch/pytorch/issues/89724,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: nn, module: norms and normalization",route to design/research decision before patch planning,2023-01-27T20:36:35Z
issue,87366,"Implementation of CG, and BICGSTAB methods",https://github.com/pytorch/pytorch/issues/87366,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: optimizer,route to design/research decision before patch planning,2023-01-27T20:36:23Z
issue,86590,"Easy way to ""freeze"" BatchNorm running_mean/running_var",https://github.com/pytorch/pytorch/issues/86590,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: nn,route to design/research decision before patch planning,2023-01-27T20:36:17Z
issue,59787,Remove support for multiple ellipses in slicing,https://github.com/pytorch/pytorch/issues/59787,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: advanced indexing, module: deprecation, module: python array api",route to design/research decision before patch planning,2023-01-11T08:04:38Z
issue,91073,Implement L1 and L2 gradient as hooks with the option of changing the weight decay value.,https://github.com/pytorch/pytorch/issues/91073,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: nn, module: optimizer",route to design/research decision before patch planning,2022-12-20T08:40:00Z
issue,84593,Uneven and/or Dynamically sized collectives,https://github.com/pytorch/pytorch/issues/84593,good volunteer slice,20,label says issue is suitable for a scoped external contribution; old update age or low-urgency/stale label; component/oncall labels: module: c10d,scan when time allows,2022-12-14T23:50:15Z
issue,80553,Nonliner conjugate gradient optimizer + Hager-Zhang line search,https://github.com/pytorch/pytorch/issues/80553,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: optimizer,route to design/research decision before patch planning,2022-11-25T02:54:51Z
issue,86641,RFC(from users): nn.Module behavior with in-place changes,https://github.com/pytorch/pytorch/issues/86641,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: functorch,route to design/research decision before patch planning,2022-10-11T13:51:18Z
issue,64067,Execute C++ based GTest unit tests from a python wrapper,https://github.com/pytorch/pytorch/issues/64067,good volunteer slice,20,label says issue is suitable for a scoped external contribution; old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2022-09-28T18:58:34Z
issue,67321,alias to generate tensor with random uniform distribution.,https://github.com/pytorch/pytorch/issues/67321,good volunteer slice,20,label says issue is suitable for a scoped external contribution; old update age or low-urgency/stale label; component/oncall labels: module: random,scan when time allows,2022-09-09T15:29:27Z
issue,83492,`Frozen` module for transfer learning.,https://github.com/pytorch/pytorch/issues/83492,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: nn,route to design/research decision before patch planning,2022-08-26T19:13:33Z
issue,83854,Pytorch/Nova CI should monitor service outages for major dependencies,https://github.com/pytorch/pytorch/issues/83854,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: ci,route to design/research decision before patch planning,2022-08-23T13:32:52Z
issue,76532,Add ability to add custom suffixes to tensor repr,https://github.com/pytorch/pytorch/issues/76532,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: printing, module: python frontend",route to design/research decision before patch planning,2022-08-23T06:52:09Z
issue,61343,Feature Request: Remove Optimizer Lazy State Initialization,https://github.com/pytorch/pytorch/issues/61343,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: optimizer,route to design/research decision before patch planning,2022-08-18T14:41:58Z
issue,67587,Re-implement `Optimizer.__repr__`,https://github.com/pytorch/pytorch/issues/67587,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: optimizer, module: python frontend",route to design/research decision before patch planning,2022-08-18T14:34:50Z
issue,79205,out-of-place functional optimizers: functional optimizers may not be composite compliant,https://github.com/pytorch/pytorch/issues/79205,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: __torch_dispatch__, module: optimizer",route to design/research decision before patch planning,2022-08-18T14:16:19Z
issue,80308,CosineAnnealingWarmRestarts with initial warm up and weight decay applied on consecutive cycles without warm up,https://github.com/pytorch/pytorch/issues/80308,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: optimizer,route to design/research decision before patch planning,2022-08-18T14:10:40Z
issue,83159,Expand Learning rate scheduling to any optimization hyperparameter,https://github.com/pytorch/pytorch/issues/83159,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: lrscheduler, module: optimizer",route to design/research decision before patch planning,2022-08-18T14:05:34Z
issue,79452,Use c10d broadcast_object in Zero,https://github.com/pytorch/pytorch/issues/79452,good volunteer slice,20,"label says issue is suitable for a scoped external contribution; old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, oncall: distributed",scan when time allows,2022-07-28T22:35:47Z
issue,33181,[feature request] [dataloader] Introduce Dataset.__collate__,https://github.com/pytorch/pytorch/issues/33181,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: dataloader,route to design/research decision before patch planning,2022-07-26T18:24:03Z
issue,68699,Transpose of a sparse tensor is not a view operation,https://github.com/pytorch/pytorch/issues/68699,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: sparse, module: viewing and reshaping",route to design/research decision before patch planning,2022-07-26T15:30:27Z
issue,81333,Reductions on tensors larger than GPU memory,https://github.com/pytorch/pytorch/issues/81333,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda,route to design/research decision before patch planning,2022-07-12T16:57:38Z
issue,80380,Support for learnable p Values in LPPOOL like Pool,https://github.com/pytorch/pytorch/issues/80380,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: nn,route to design/research decision before patch planning,2022-06-28T15:22:40Z
issue,79877,test_meta_vstack_cuda_int16 (__main__.TestMetaCUDA) Fails with DEBUG=1,https://github.com/pytorch/pytorch/issues/79877,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: autograd,route to design/research decision before patch planning,2022-06-25T03:25:08Z
issue,79842,Automatically use CUDA,https://github.com/pytorch/pytorch/issues/79842,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: python frontend,route to design/research decision before patch planning,2022-06-21T20:32:54Z
issue,68516,"[ux] [feature request] Arguments to modules(), named_modules(), children(), named_children() to filter modules of specific types",https://github.com/pytorch/pytorch/issues/68516,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: nn,route to design/research decision before patch planning,2022-06-02T18:12:51Z
issue,14844,Negative indexing for nn.Embedding inputs,https://github.com/pytorch/pytorch/issues/14844,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: nn,route to design/research decision before patch planning,2022-05-24T21:51:52Z
issue,72712,Better support for pypip packages implementing torch cuda extentions,https://github.com/pytorch/pytorch/issues/72712,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: binaries,route to design/research decision before patch planning,2022-03-31T11:28:20Z
issue,74904,Adding novel 'AdaFamily' optimizer,https://github.com/pytorch/pytorch/issues/74904,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: optimizer,route to design/research decision before patch planning,2022-03-30T09:14:02Z
issue,73143,Add nan-safe einsum and bilinear,https://github.com/pytorch/pytorch/issues/73143,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: linear algebra,route to design/research decision before patch planning,2022-03-14T07:20:41Z
issue,73172,LazyLinear with equal in_features and out_features,https://github.com/pytorch/pytorch/issues/73172,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: nn,route to design/research decision before patch planning,2022-03-01T18:18:43Z
issue,73513,Wrapper tensor level confusion,https://github.com/pytorch/pytorch/issues/73513,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: __torch_dispatch__,route to design/research decision before patch planning,2022-03-01T14:44:13Z
issue,67634,Feature request: Grid sample on complex tensors with float grid,https://github.com/pytorch/pytorch/issues/67634,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: complex, module: interpolation, module: nn",route to design/research decision before patch planning,2022-02-06T21:40:47Z
issue,68798,Clarify variables of BatchNorm*d functions,https://github.com/pytorch/pytorch/issues/68798,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: docs, module: nn, module: norms and normalization",route to design/research decision before patch planning,2022-01-23T18:03:00Z
issue,69757,[Feature request] nn.Model API: Standard model interface,https://github.com/pytorch/pytorch/issues/69757,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: nn,route to design/research decision before patch planning,2021-12-17T18:15:39Z
issue,44827,[RFC] Pipeline Parallelism in PyTorch,https://github.com/pytorch/pytorch/issues/44827,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: rpc, oncall: distributed",route to design/research decision before patch planning,2021-12-09T17:30:40Z
issue,68648,[feature request] Sticky/force regime of train/eval modes,https://github.com/pytorch/pytorch/issues/68648,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: nn, module: ux",route to design/research decision before patch planning,2021-11-27T11:41:20Z
issue,68514,Add common regularizations to PyTorch.,https://github.com/pytorch/pytorch/issues/68514,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: nn, module: norms and normalization",route to design/research decision before patch planning,2021-11-20T16:16:30Z
issue,65913,Support differentiability through clone and update,https://github.com/pytorch/pytorch/issues/65913,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: nn",route to design/research decision before patch planning,2021-11-05T17:35:03Z
issue,64730,Enable BatchNorm to use running mean/variance during train,https://github.com/pytorch/pytorch/issues/64730,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: nn, module: norms and normalization",route to design/research decision before patch planning,2021-10-07T13:13:35Z
issue,57724,Incorporate Posit Support,https://github.com/pytorch/pytorch/issues/57724,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label,route to design/research decision before patch planning,2021-09-17T13:28:03Z
issue,46168,[discussion] In-place gradient (grad_input) computation for better memory utilisation,https://github.com/pytorch/pytorch/issues/46168,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: memory usage",route to design/research decision before patch planning,2021-09-07T03:34:19Z
issue,49890,GEMM with int8 datatype throws RuntimeError on GPU,https://github.com/pytorch/pytorch/issues/49890,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: linear algebra,route to design/research decision before patch planning,2021-09-01T05:27:12Z
issue,46544,PyTorch NaN behavior and API design,https://github.com/pytorch/pytorch/issues/46544,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: nans and infs, module: numpy, module: ux",route to design/research decision before patch planning,2021-08-30T21:08:20Z
issue,60223,[testing] SkipInfo should error if `cls_name` and `test_name` combination is not valid,https://github.com/pytorch/pytorch/issues/60223,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: tests,route to design/research decision before patch planning,2021-08-30T02:53:38Z
issue,20704,Remove unpack() in torch/csrc/nn/type_checks.h and its caller functions in the codebase,https://github.com/pytorch/pytorch/issues/20704,good volunteer slice,20,"label says issue is suitable for a scoped external contribution; old update age or low-urgency/stale label; component/oncall labels: module: internals, module: nn",scan when time allows,2021-08-25T16:58:36Z
issue,63038,Use pytorch-probot for PyTorch-specific stuff + MOTD in Dr. CI comment,https://github.com/pytorch/pytorch/issues/63038,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: infra,route to design/research decision before patch planning,2021-08-16T21:38:33Z
issue,61490,Deprecate torch.(min|max|median|mode) to only return values and not indices,https://github.com/pytorch/pytorch/issues/61490,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: deprecation, module: numpy, module: python array api, module: reductions",route to design/research decision before patch planning,2021-08-12T16:14:00Z
issue,62155,[docs] Change code snippet formatting to be directly copy-pasteable into REPL or into source files,https://github.com/pytorch/pytorch/issues/62155,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: docs,route to design/research decision before patch planning,2021-07-27T09:08:08Z
issue,61901,Reducing over empty dimensions for reductions without identity,https://github.com/pytorch/pytorch/issues/61901,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: numpy, module: reductions",route to design/research decision before patch planning,2021-07-22T13:16:16Z
issue,61664,Support adding new keyword-only parameters without breaking FC,https://github.com/pytorch/pytorch/issues/61664,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: codegen, module: internals",route to design/research decision before patch planning,2021-07-19T17:19:53Z
issue,61585,"Proposal: allow using axis, axes, dim and dims interchangeably",https://github.com/pytorch/pytorch/issues/61585,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: python array api, module: ux",route to design/research decision before patch planning,2021-07-14T23:50:47Z
issue,59868,Python scalars should be promoted to the same `dtype` as the respective tensor,https://github.com/pytorch/pytorch/issues/59868,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: python array api, module: type promotion",route to design/research decision before patch planning,2021-06-16T21:05:19Z
issue,51962,PyTorch for scientific computing,https://github.com/pytorch/pytorch/issues/51962,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label,route to design/research decision before patch planning,2021-05-05T18:59:06Z
issue,41803,out-variant for tensor.bitwise_and (exists for torch.bitwise_and) + bitwise_friends,https://github.com/pytorch/pytorch/issues/41803,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: numpy, module: ux",route to design/research decision before patch planning,2021-03-24T15:26:45Z
issue,35801,Add KFAC optimizer,https://github.com/pytorch/pytorch/issues/35801,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: optimizer,route to design/research decision before patch planning,2021-03-09T10:55:54Z
issue,51156,[Feature Request] Support tensor creation from objects that implement the __array__ interface,https://github.com/pytorch/pytorch/issues/51156,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: numpy,route to design/research decision before patch planning,2021-03-06T17:44:18Z
issue,52207,Wasserstein metric,https://github.com/pytorch/pytorch/issues/52207,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: nn,route to design/research decision before patch planning,2021-02-26T17:46:36Z
issue,52036,Automatic splitting of batch during model feed-forwarding to avoid memory errors caused by the too big batch_size,https://github.com/pytorch/pytorch/issues/52036,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: memory usage,route to design/research decision before patch planning,2021-02-11T09:27:35Z
issue,43450,Training Large Neural Networks with Constant Memory using a New Execution Algorithm,https://github.com/pytorch/pytorch/issues/43450,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: memory usage, module: nn",route to design/research decision before patch planning,2021-02-08T05:17:12Z
issue,50204,Find the joint eigenvalue of two matrices.,https://github.com/pytorch/pytorch/issues/50204,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: linear algebra,route to design/research decision before patch planning,2021-02-03T11:48:09Z
issue,46829,Mixing Numpy's arrays and PyTorch tensors,https://github.com/pytorch/pytorch/issues/46829,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: numpy, module: ux",route to design/research decision before patch planning,2021-01-27T14:23:19Z
issue,50712,ignore_index for nn_mse_loss,https://github.com/pytorch/pytorch/issues/50712,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: loss, module: nn",route to design/research decision before patch planning,2021-01-20T14:07:49Z
issue,45051,Custom Datatypes in Tensors,https://github.com/pytorch/pytorch/issues/45051,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: numpy,route to design/research decision before patch planning,2021-01-12T16:43:34Z
issue,8837,Inconsistency in implementation of _LRScheduler,https://github.com/pytorch/pytorch/issues/8837,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: nn, module: optimizer",route to design/research decision before patch planning,2021-01-05T17:15:17Z
issue,49338,a time series library,https://github.com/pytorch/pytorch/issues/49338,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label,route to design/research decision before patch planning,2020-12-21T21:45:00Z
issue,46516,Batchnorm support for tracking buffer statistics when using gradient accumulation,https://github.com/pytorch/pytorch/issues/46516,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: norms and normalization,route to design/research decision before patch planning,2020-12-19T02:10:02Z
issue,48747,[bug][libtorch] torch::load seeks to the start of the input stream before reading,https://github.com/pytorch/pytorch/issues/48747,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: serialization,route to design/research decision before patch planning,2020-12-03T01:31:46Z
issue,48494,Inconsistent type of property stride of Conv1d and MaxPool1d,https://github.com/pytorch/pytorch/issues/48494,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: nn, module: ux",route to design/research decision before patch planning,2020-11-27T11:43:47Z
issue,47915,sparse filter layers (more specifically convolutions),https://github.com/pytorch/pytorch/issues/47915,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: sparse,route to design/research decision before patch planning,2020-11-16T13:28:19Z
issue,47037,"Change ""_next_index()"" in DataLoader to a public and stable API",https://github.com/pytorch/pytorch/issues/47037,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: dataloader,route to design/research decision before patch planning,2020-10-29T14:39:41Z
issue,6657,[feature request] Stochastic Variance Reduced Gradient (SVRG) optimizer,https://github.com/pytorch/pytorch/issues/6657,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: optimizer,route to design/research decision before patch planning,2020-10-20T14:40:01Z
issue,44591,[Feature] Fused Matmul & Min/Max/Sum/Prod,https://github.com/pytorch/pytorch/issues/44591,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: linear algebra, module: performance",route to design/research decision before patch planning,2020-10-06T09:49:47Z
issue,45808,Support for Locality Sensitive Hashing Optimizations,https://github.com/pytorch/pytorch/issues/45808,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label,route to design/research decision before patch planning,2020-10-05T20:52:47Z
issue,45133,[Feature Request] Add support for Hidden Markov Models in torch.distributions,https://github.com/pytorch/pytorch/issues/45133,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: distributions,route to design/research decision before patch planning,2020-10-01T00:01:23Z
issue,45295,Adaptive usage of memory during training inference,https://github.com/pytorch/pytorch/issues/45295,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: nn",route to design/research decision before patch planning,2020-09-28T15:37:30Z
issue,44777,Hybrid Memory,https://github.com/pytorch/pytorch/issues/44777,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: memory usage",route to design/research decision before patch planning,2020-09-18T09:00:43Z
issue,42652,Add non_blocking=True copy of torch.Storage,https://github.com/pytorch/pytorch/issues/42652,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label,route to design/research decision before patch planning,2020-08-10T22:12:26Z
issue,4660,[Feature proposal] Add MC-derived optimizers,https://github.com/pytorch/pytorch/issues/4660,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: optimizer,route to design/research decision before patch planning,2020-08-04T18:07:49Z
issue,41559,momentum in BatchNorm,https://github.com/pytorch/pytorch/issues/41559,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: docs, module: nn",route to design/research decision before patch planning,2020-07-21T19:24:25Z
issue,41556,Implement LSH Optimizations for Enhanced CPU-Only Performance,https://github.com/pytorch/pytorch/issues/41556,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: performance",route to design/research decision before patch planning,2020-07-17T17:20:11Z
issue,39780,LRP-based explainability,https://github.com/pytorch/pytorch/issues/39780,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label,route to design/research decision before patch planning,2020-06-10T17:47:44Z
issue,3904,Implement DE in pytorch.optim,https://github.com/pytorch/pytorch/issues/3904,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: optimizer,route to design/research decision before patch planning,2020-05-29T14:18:10Z
issue,37155,Per-cluster biases in AdaptiveLogSoftmaxWithLoss,https://github.com/pytorch/pytorch/issues/37155,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: nn,route to design/research decision before patch planning,2020-04-28T19:05:20Z
pr,184042,Tma scatter add remove software pipelining,https://github.com/pytorch/pytorch/pull/184042,stale/low urgency,17,PR has not updated for at least 45 days,review when ready,2026-05-22T22:08:36Z
pr,184753,Fix Dynamo frozenset copy identity semantics,https://github.com/pytorch/pytorch/pull/184753,stale/low urgency,17,"PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan",review when ready,2026-05-21T21:46:27Z
pr,184605,Add CPython Dynamo agentic loop plan,https://github.com/pytorch/pytorch/pull/184605,stale/low urgency,17,PR has not updated for at least 45 days,review when ready,2026-05-21T19:51:43Z
pr,182269,[MPS] Fix channels_last batch_norm backward,https://github.com/pytorch/pytorch/pull/182269,stale/low urgency,17,PR has not updated for at least 45 days; CI lane labels: ciflow/mps,review when ready,2026-05-21T06:25:07Z
pr,183386,Add tests for OneCycleLR zero divisor initialization behavior,https://github.com/pytorch/pytorch/pull/183386,stale/low urgency,17,PR has not updated for at least 45 days,review when ready,2026-05-19T22:50:52Z
pr,184116,[distributed] Add NVFP4 scaled matmul + reduce-scatter fast path #182915,https://github.com/pytorch/pytorch/pull/184116,stale/low urgency,17,"PR has not updated for at least 45 days; CI lane labels: ciflow/h100-symm-mem, ciflow/torchtitan",review when ready,2026-05-19T17:20:45Z
pr,184067,[submodule] Drop unused third_party/gemmlowp submodule,https://github.com/pytorch/pytorch/pull/184067,stale/low urgency,17,PR has not updated for at least 45 days; CI lane labels: ciflow/trunk,review when ready,2026-05-19T15:21:11Z
pr,183976,[CUDA] Correct inner scope Math.cuh int i -> int j,https://github.com/pytorch/pytorch/pull/183976,stale/low urgency,17,PR has not updated for at least 45 days,review when ready,2026-05-19T14:45:02Z
pr,183568,Fix timeout in `test_share_memory` on error,https://github.com/pytorch/pytorch/pull/183568,stale/low urgency,17,PR has not updated for at least 45 days; CI lane labels: ciflow/h100-symm-mem,review when ready,2026-05-18T21:32:41Z
pr,182424,Warning oversized storage copy,https://github.com/pytorch/pytorch/pull/182424,stale/low urgency,17,PR has not updated for at least 45 days,review when ready,2026-05-18T09:53:37Z
pr,183203,perf(dataloader): zero-copy contiguous-slice fast path for TensorDataset,https://github.com/pytorch/pytorch/pull/183203,stale/low urgency,17,PR has not updated for at least 45 days,review when ready,2026-05-16T01:06:58Z
pr,183726,[poc] yet another nonstrict api,https://github.com/pytorch/pytorch/pull/183726,stale/low urgency,17,PR has not updated for at least 45 days,review when ready,2026-05-15T21:54:44Z
pr,183787,[PyTorch] Support selective TorchScript tensor data elision (#183787),https://github.com/pytorch/pytorch/pull/183787,stale/low urgency,17,PR has not updated for at least 45 days,review when ready,2026-05-14T22:48:56Z
pr,177450,DataPipes: prevent failing properties from masking invalid-iterator errors,https://github.com/pytorch/pytorch/pull/177450,stale/low urgency,17,PR has not updated for at least 45 days,review when ready,2026-05-14T21:34:59Z
pr,175320,[MPS] Ensure pool2d inputs are contiguous in target memory format,https://github.com/pytorch/pytorch/pull/175320,stale/low urgency,17,PR has not updated for at least 45 days; CI lane labels: ciflow/mps,review when ready,2026-05-14T19:40:35Z
pr,182114,Add torch.utils.nan_guard for forward-pass NaN/Inf debugging,https://github.com/pytorch/pytorch/pull/182114,stale/low urgency,17,PR has not updated for at least 45 days,review when ready,2026-05-14T17:51:21Z
pr,182583,Skip RST<->MD source conversions in doc-redirects-check,https://github.com/pytorch/pytorch/pull/182583,stale/low urgency,17,PR has not updated for at least 45 days,review when ready,2026-05-14T16:37:32Z
pr,181262,"Show node.meta[""custom""][""stream""] in Graph.__str__",https://github.com/pytorch/pytorch/pull/181262,stale/low urgency,17,PR has not updated for at least 45 days; CI lane labels: ciflow/trunk,review when ready,2026-05-14T16:23:10Z
pr,183405,Fix unused groups in test_convs,https://github.com/pytorch/pytorch/pull/183405,stale/low urgency,17,PR has not updated for at least 45 days,review when ready,2026-05-14T16:06:31Z
pr,183502,Remove outdated code for macOS < 12,https://github.com/pytorch/pytorch/pull/183502,stale/low urgency,17,PR has not updated for at least 45 days,review when ready,2026-05-14T16:06:15Z
pr,182788,Fix decorators of `ProcessGroupGlooTest` & `ProcessGroupNCCLTest`,https://github.com/pytorch/pytorch/pull/182788,stale/low urgency,17,PR has not updated for at least 45 days,review when ready,2026-05-14T16:04:27Z
pr,182836,[ROCm][CI] Update mi200 workflows to use arc label,https://github.com/pytorch/pytorch/pull/182836,stale/low urgency,17,PR has not updated for at least 45 days,review when ready,2026-05-14T16:03:52Z
pr,182746,[MPS] Memory-efficient SDPA backward (opt-in),https://github.com/pytorch/pytorch/pull/182746,stale/low urgency,17,"PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/mps",review when ready,2026-05-14T16:03:47Z
pr,182873,test: skip some test_non_standard_bool_values test cases,https://github.com/pytorch/pytorch/pull/182873,stale/low urgency,17,PR has not updated for at least 45 days,review when ready,2026-05-14T16:03:28Z
pr,183037,[Distributed] Add `ranks_only` to `StateDictOptions` for controlling which ranks receive the full state dict,https://github.com/pytorch/pytorch/pull/183037,stale/low urgency,17,PR has not updated for at least 45 days; CI lane labels: ciflow/torchtitan,review when ready,2026-05-14T16:02:20Z
pr,171686,Add failing test documenting implicit step during `LRScheduler` initialization,https://github.com/pytorch/pytorch/pull/171686,stale/low urgency,17,PR has not updated for at least 45 days,review when ready,2026-05-12T23:39:52Z
pr,182857,[inductor][mtia] Skip matvec bmm/addmm decomposition on MTIA; add MTIA matvec bmm lowering (#182857),https://github.com/pytorch/pytorch/pull/182857,stale/low urgency,17,"PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan",review when ready,2026-05-12T15:37:03Z
pr,166625,Fix skip condition in test_overrides,https://github.com/pytorch/pytorch/pull/166625,stale/low urgency,17,PR has not updated for at least 45 days,review when ready,2026-05-12T10:35:17Z
pr,182934,[RFC][distributed] Add parallel_map with collective operation interception…,https://github.com/pytorch/pytorch/pull/182934,stale/low urgency,17,PR has not updated for at least 45 days; CI lane labels: ciflow/torchtitan,review when ready,2026-05-11T22:15:00Z
pr,182975,Document 5 public APIs in docs/source/utils.md,https://github.com/pytorch/pytorch/pull/182975,stale/low urgency,17,PR has not updated for at least 45 days,review when ready,2026-05-11T21:26:01Z
pr,176430,Don't use sharding for tests with a single shard,https://github.com/pytorch/pytorch/pull/176430,stale/low urgency,17,PR has not updated for at least 45 days,review when ready,2026-05-11T11:21:08Z
pr,174768,Normalize object names in TreeProfiler test,https://github.com/pytorch/pytorch/pull/174768,stale/low urgency,17,PR has not updated for at least 45 days,review when ready,2026-05-11T11:20:52Z
pr,165986,Skip tests requiring NCCL if no GPU is found,https://github.com/pytorch/pytorch/pull/165986,stale/low urgency,17,PR has not updated for at least 45 days,review when ready,2026-05-11T11:20:51Z
pr,174959,Make server creation in distributed/elastic/test_control_plane more reliable,https://github.com/pytorch/pytorch/pull/174959,stale/low urgency,17,PR has not updated for at least 45 days,review when ready,2026-05-11T10:01:53Z
pr,177126,Fix search for OpenMP using e.g. GCC,https://github.com/pytorch/pytorch/pull/177126,stale/low urgency,17,PR has not updated for at least 45 days,review when ready,2026-05-11T08:45:15Z
pr,169936,Fix `TestSaveLoad.test_version_error`,https://github.com/pytorch/pytorch/pull/169936,stale/low urgency,17,PR has not updated for at least 45 days,review when ready,2026-05-11T08:45:03Z
pr,176709,Remove Eigen d_cnjg and r_cnjg source reference,https://github.com/pytorch/pytorch/pull/176709,stale/low urgency,17,PR has not updated for at least 45 days,review when ready,2026-05-11T07:56:57Z
pr,141842,Refactor NJT to hold metadata on nested int,https://github.com/pytorch/pytorch/pull/141842,stale/low urgency,17,"PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/trunk",review when ready,2026-05-06T20:42:48Z
pr,119977,[NJT] Factory function support,https://github.com/pytorch/pytorch/pull/119977,stale/low urgency,17,PR has not updated for at least 45 days; CI lane labels: ciflow/inductor,review when ready,2026-05-04T22:34:38Z
pr,132414,Ensure FakeTensor post-creation has associated nested int cached,https://github.com/pytorch/pytorch/pull/132414,stale/low urgency,17,PR has not updated for at least 45 days; CI lane labels: ciflow/inductor,review when ready,2026-05-04T22:34:33Z
pr,168223,Add failing test for negative epoch crash in LR schedulers (ZeroDivisionError),https://github.com/pytorch/pytorch/pull/168223,stale/low urgency,17,PR has not updated for at least 45 days,review when ready,2026-04-04T12:42:00Z
pr,130752,support zb1p and zb2p algorithms,https://github.com/pytorch/pytorch/pull/130752,stale/low urgency,17,"PR has not updated for at least 45 days; CI lane labels: ciflow/torchtitan, ciflow/trunk",review when ready,2026-03-26T21:24:16Z
pr,168895,Add `pad_to_multiple_of` to `pad_sequence` (C++ only),https://github.com/pytorch/pytorch/pull/168895,stale/low urgency,17,PR has not updated for at least 45 days,review when ready,2026-02-03T06:32:28Z
pr,167541,[MPS] Fix crashes due to a race condition in the MPS backend,https://github.com/pytorch/pytorch/pull/167541,stale/low urgency,17,PR has not updated for at least 45 days,review when ready,2026-01-26T10:21:33Z
pr,162079,Tune linalg_eigh_cusolver: better heuristic for syevj_batched selection on cuda,https://github.com/pytorch/pytorch/pull/162079,stale/low urgency,17,PR has not updated for at least 45 days,review when ready,2025-11-15T20:20:24Z
pr,152158,Add AC_TRACER Infra TorchDispatchMode key,https://github.com/pytorch/pytorch/pull/152158,stale/low urgency,17,PR has not updated for at least 45 days,review when ready,2025-07-24T18:26:34Z
pr,102613,Support building pytorch using MKL ILP64 model.,https://github.com/pytorch/pytorch/pull/102613,stale/low urgency,17,PR has not updated for at least 45 days,review when ready,2025-06-19T08:43:08Z
pr,147927,Custom ops support arbitrary input types by migrating to python dispatcher,https://github.com/pytorch/pytorch/pull/147927,stale/low urgency,17,PR has not updated for at least 45 days; CI lane labels: ciflow/inductor,review when ready,2025-05-30T15:02:48Z
pr,136835,Add back DistributedDataParallel types that were lost when pyi was removed,https://github.com/pytorch/pytorch/pull/136835,stale/low urgency,17,PR has not updated for at least 45 days,review when ready,2025-05-06T09:21:08Z
pr,142114,[DO NOT MERGE][WIP] CI: Dispatch PR events to the out-of-tree test infra,https://github.com/pytorch/pytorch/pull/142114,stale/low urgency,17,PR has not updated for at least 45 days,review when ready,2025-04-16T07:00:50Z
pr,146101,(WIP) Update NJT ops to check data for raggedness check,https://github.com/pytorch/pytorch/pull/146101,stale/low urgency,17,PR has not updated for at least 45 days; CI lane labels: ciflow/inductor,review when ready,2025-04-15T16:01:07Z
pr,145922,Update NestedInt equality to take into account all metadata,https://github.com/pytorch/pytorch/pull/145922,stale/low urgency,17,PR has not updated for at least 45 days; CI lane labels: ciflow/inductor,review when ready,2025-04-08T21:20:16Z
pr,146318,Hack AC to not clear recomputed activations,https://github.com/pytorch/pytorch/pull/146318,stale/low urgency,17,PR has not updated for at least 45 days,review when ready,2025-04-07T21:46:21Z
pr,138519,[Inductor] introduce comm buffer planning,https://github.com/pytorch/pytorch/pull/138519,stale/low urgency,17,PR has not updated for at least 45 days; CI lane labels: ciflow/inductor,review when ready,2025-03-06T02:03:37Z
pr,140321,[scan] Refactored testcases,https://github.com/pytorch/pytorch/pull/140321,stale/low urgency,17,PR has not updated for at least 45 days,review when ready,2025-01-18T18:23:50Z
pr,119496,Switch batch norm stack to consolidated ops,https://github.com/pytorch/pytorch/pull/119496,stale/low urgency,17,"PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/mps",review when ready,2024-09-24T23:02:33Z
pr,130462,[testing] Allow creating offsets within AC region for NJT without SAC,https://github.com/pytorch/pytorch/pull/130462,stale/low urgency,17,PR has not updated for at least 45 days,review when ready,2024-09-23T19:55:20Z
pr,117905,Factory function and basic .sizes() support for C++ NestedTensor,https://github.com/pytorch/pytorch/pull/117905,stale/low urgency,17,PR has not updated for at least 45 days,review when ready,2024-08-26T17:00:50Z
pr,110155,[TESTING] Capture scalar/dynamic by default,https://github.com/pytorch/pytorch/pull/110155,stale/low urgency,17,PR has not updated for at least 45 days; CI lane labels: ciflow/inductor,review when ready,2024-04-07T06:26:26Z
pr,108303,[optim] Make casting to match params a hook (2nd try),https://github.com/pytorch/pytorch/pull/108303,stale/low urgency,17,PR has not updated for at least 45 days; CI lane labels: ciflow/trunk,review when ready,2024-01-12T19:19:46Z
pr,82997,Implement refs.var as a real reference,https://github.com/pytorch/pytorch/pull/82997,stale/low urgency,17,PR has not updated for at least 45 days,review when ready,2023-04-17T00:35:59Z
pr,184358,Add XPU to externalId,https://github.com/pytorch/pytorch/pull/184358,draft/noise,15,PR is draft; PR closes or references open issue numbers; PR has not updated for at least 45 days,wait for author to mark ready,2026-05-25T00:03:21Z
pr,183507,einsum: fast path for small-K batched-dot to avoid cuBLAS launch overhead,https://github.com/pytorch/pytorch/pull/183507,draft/noise,15,PR is draft; PR closes or references open issue numbers; PR has not updated for at least 45 days,wait for author to mark ready,2026-05-18T02:07:30Z
pr,183009,[Inductor] Fall back native_layer_norm; route var_mean through inductor lowering,https://github.com/pytorch/pytorch/pull/183009,draft/noise,15,"PR is draft; PR closes or references open issue numbers; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for author to mark ready,2026-05-11T03:11:51Z
pr,153893,[PP] Fix double backward error in stage_backward,https://github.com/pytorch/pytorch/pull/153893,draft/noise,15,PR is draft; PR closes or references open issue numbers; PR has not updated for at least 45 days; CI lane labels: ciflow/torchtitan,wait for author to mark ready,2026-03-20T22:40:30Z
pr,124624,[NJT] Allow construction of NJT within graph using offsets from inputs,https://github.com/pytorch/pytorch/pull/124624,draft/noise,15,PR is draft; PR closes or references open issue numbers; PR has not updated for at least 45 days,wait for author to mark ready,2024-08-12T20:07:16Z
issue,158986,`nvlink error: Entry function '<redacted for brevity>' uses too much data for compiler-generated constants`,https://github.com/pytorch/pytorch/issues/158986,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: build, module: cuda",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-01-07T16:46:46Z
issue,156010,`torch.prod` or `torch.special.entr` triggers `CUDA driver error: invalid argument` on GPU unless kernel cache is cleared,https://github.com/pytorch/pytorch/issues/156010,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda,confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-01-04T08:16:18Z
issue,165798,matmul/mm/bmm crashes on CUDA 13.0 + Libtorch 2.9 on Ubuntu 24.04,https://github.com/pytorch/pytorch/issues/165798,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: cpp",confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-12-27T06:21:53Z
issue,171092,Load nVidia libraries in reverse order so that the newest version can take control,https://github.com/pytorch/pytorch/issues/171092,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: cuda, module: third_party",confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-12-22T18:56:07Z
issue,163510,building pytorch c++ extension with cuda 13 causes unexpected behavior when pytorch is built with cuda 12.8,https://github.com/pytorch/pytorch/issues/163510,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: cpp-extensions",confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-12-20T20:24:14Z
issue,170278,macOS Parallel Builds Fail with jobs > 4,https://github.com/pytorch/pytorch/issues/170278,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: ci, module: macos",confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-12-18T21:19:28Z
issue,170524,RuntimeError occurred when I train “Resnet”,https://github.com/pytorch/pytorch/issues/170524,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: cudnn, module: third_party",confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-12-17T01:15:40Z
issue,170038,Conv2D numerical mismatch vs TensorFlow,https://github.com/pytorch/pytorch/issues/170038,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: convolution,confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-12-15T09:52:08Z
issue,170016,Conv2d produces numerically inconsistent results compared to other DL frameworks (TensorFlow / Paddle / MindSpore),https://github.com/pytorch/pytorch/issues/170016,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: convolution,confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-12-11T16:49:44Z
issue,169342,[MPS] Batch inference produces incorrect results when chunk() views are passed through conv operations,https://github.com/pytorch/pytorch/issues/169342,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: correctness (silent), module: mps",confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-12-09T11:42:07Z
issue,169082,Reduce Overhead and Max Autotune NaN,https://github.com/pytorch/pytorch/issues/169082,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: inductor, oncall: pt2",confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-12-04T01:48:55Z
issue,168362,Triton 3.4.0 wheel hash mismatch between PyPI metadata and download server,https://github.com/pytorch/pytorch/issues/168362,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: binaries,confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-11-21T17:41:59Z
issue,167820,Why torch==2.9 compile qwen3 model with block ptr will crash?,https://github.com/pytorch/pytorch/issues/167820,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: vllm, oncall: pt2",confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-11-18T06:07:13Z
issue,137260,`import torch` takes forever,https://github.com/pytorch/pytorch/issues/137260,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label,confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-11-16T01:37:22Z
issue,167833,Incorrect SHA256 on index for torchvision==0.24.1+cu130,https://github.com/pytorch/pytorch/issues/167833,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: binaries,confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-11-15T20:12:01Z
issue,114206,[PT2] Compile Cold Start (Persistent Cacheing) - AOTAutograd may be bottleneck when `TORCHINDUCTOR_FX_GRAPH_CACHE=1`,https://github.com/pytorch/pytorch/issues/114206,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: aotdispatch, module: compile-time, module: dynamo, module: performance",confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-11-06T01:38:21Z
issue,105700,Don't use weak ref finalization for freeing resources when code objects die,https://github.com/pytorch/pytorch/issues/105700,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-11-05T19:46:28Z
issue,82218,OOM during backward() leads to memory leaks,https://github.com/pytorch/pytorch/issues/82218,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: memory usage",confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-11-04T16:24:09Z
issue,60477,ROCm miopenStatusInternalError /MIOpen/src/sqlite_db.cpp:109: open memvfs: unable to open database file,https://github.com/pytorch/pytorch/issues/60477,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: rocm,confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-11-04T05:15:26Z
issue,104259,ImportError: libcudnn.so.8: cannot open shared object file: No such file or directory,https://github.com/pytorch/pytorch/issues/104259,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: cuda",confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-11-01T14:44:05Z
issue,163311,[build] `CMake (or ninja)` configuration can't index `uv` environment for incremental compilation,https://github.com/pytorch/pytorch/issues/163311,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: build,confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-10-29T02:08:13Z
issue,166303,Pytorch Operators on older pytorch version,https://github.com/pytorch/pytorch/issues/166303,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: docs,confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-10-27T16:55:38Z
issue,165590,RuntimeError: non-positive groups is not supported,https://github.com/pytorch/pytorch/issues/165590,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: nn,confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-10-17T18:59:35Z
issue,163923,Enabling activation checkpoint debug makes throughput gradually slow down to crawl.,https://github.com/pytorch/pytorch/issues/163923,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: activation checkpointing,confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-10-15T11:44:30Z
issue,159224,"Import Error: Undefined Symbol, _PyCode_SetExtra: Torch Extensions doesn't seem to support Python 3.12",https://github.com/pytorch/pytorch/issues/159224,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label,confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-10-12T17:07:15Z
issue,155344,PyTorch CPP Extensions fail when same kernel is compiled more than once on ROCm servers,https://github.com/pytorch/pytorch/issues/155344,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: cpp-extensions, module: rocm",confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-10-09T17:09:00Z
issue,99573,"Jacfwd become slower after update pytorch (""We’ve integrated functorch into PyTorch---Documentation"")",https://github.com/pytorch/pytorch/issues/99573,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: functorch,confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-09-29T23:25:58Z
issue,163347,Incorrect bf16 results in Conv,https://github.com/pytorch/pytorch/issues/163347,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: bfloat16, module: correctness (silent), module: mkldnn",confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-09-24T19:17:48Z
issue,137524,FSDP-2 doesnt do overlapping when composed with TP,https://github.com/pytorch/pytorch/issues/137524,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-09-19T18:06:09Z
issue,161104,Eager CUDAGraph + stream performance,https://github.com/pytorch/pytorch/issues/161104,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: cuda graphs, module: performance",confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-09-15T17:24:12Z
issue,129992,Got this error when just running my flask script,https://github.com/pytorch/pytorch/issues/129992,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: internals,confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-09-06T11:00:36Z
issue,45273,"Training fast with small dataset, slow with large dataset",https://github.com/pytorch/pytorch/issues/45273,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision), module: data parallel, module: performance",confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-09-03T15:48:42Z
issue,161494,OOM,https://github.com/pytorch/pytorch/issues/161494,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: pt2-dispatcher,confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-08-26T13:50:47Z
issue,160846,Can't run a model trained in MPS-system on CPU-only system.,https://github.com/pytorch/pytorch/issues/160846,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: serialization,confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-08-19T19:24:18Z
issue,146769,MPS Error on sequoia 15.3: NDArray dimension length > INT_MAX',https://github.com/pytorch/pytorch/issues/146769,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: mps,confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-08-17T16:02:17Z
issue,159447,dataloader hits CUDA error: invalid argument in data.pin_memory(device) on RTX 3090,https://github.com/pytorch/pytorch/issues/159447,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: dataloader",confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-08-11T17:41:04Z
issue,160261,CPU: Pytorch >= 2.7.0 is broken,https://github.com/pytorch/pytorch/issues/160261,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: binaries,confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-08-11T14:50:53Z
issue,159669,Torch Compile. CudaGraphs Memory leak,https://github.com/pytorch/pytorch/issues/159669,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: oncall: pt2,confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-08-10T17:47:35Z
issue,134331,"CMake Error: When installing PyTorch from source, CUDA not being detected.",https://github.com/pytorch/pytorch/issues/134331,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: build,confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-08-06T03:30:24Z
issue,129658,[TensorDict - compile] dynamo generator compatibility,https://github.com/pytorch/pytorch/issues/129658,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-08-01T23:36:27Z
issue,159309,torch.nn.InstanceNorm2d CPU/GPU Inconsistency,https://github.com/pytorch/pytorch/issues/159309,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: half, module: numerical-reproducibility",confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-07-30T01:37:56Z
issue,150998,Unsupported operand 118,https://github.com/pytorch/pytorch/issues/150998,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: serialization,confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-07-29T14:46:09Z
issue,159196,torch.compile: device of tensor switches after the next statement is run. no repro,https://github.com/pytorch/pytorch/issues/159196,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: oncall: pt2,confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-07-28T19:12:35Z
issue,158906,The make lint command got stuck.,https://github.com/pytorch/pytorch/issues/158906,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: lint,confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-07-24T08:34:24Z
issue,111693,"[export] 14k models: AssertionError: graph-captured input # 2, of type <class 'torch.nn.parameter.Parameter'>, is not among original inputs of types",https://github.com/pytorch/pytorch/issues/111693,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: oncall: export, oncall: pt2",confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-07-11T18:16:07Z
issue,157237,"Unexpected, batch size and device dependent NaN propagation in Conv1d",https://github.com/pytorch/pytorch/issues/157237,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: arm, module: correctness (silent), module: macos, module: nans and infs",confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-07-09T16:15:13Z
issue,151740,mps and cpu backends produce different training results with FFT and Adam,https://github.com/pytorch/pytorch/issues/151740,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: fft, module: mps",confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-07-08T23:06:34Z
issue,157709,/lib/x86_64-linux-gnu/libstdc++.so.6: version `GLIBCXX_3.4.30' not found when IMPORT TORCH,https://github.com/pytorch/pytorch/issues/157709,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label,confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-07-07T22:28:00Z
issue,157398,Cannot copy data from one gpu to another using torch,https://github.com/pytorch/pytorch/issues/157398,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: cuda",confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-07-07T22:27:32Z
issue,157360,torch.Tensor.addmm_ The calculation result is inconsistent with the formula calculation result,https://github.com/pytorch/pytorch/issues/157360,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: linear algebra,confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-07-02T23:58:35Z
issue,151667,[MPS] MultiheadAttention with masks and dropout produces NaNs,https://github.com/pytorch/pytorch/issues/151667,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: macos, module: mps, module: nans and infs",confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-06-28T21:26:45Z
issue,151310,cpp_extension.load 'sources' does not support Chinese paths,https://github.com/pytorch/pytorch/issues/151310,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: cpp-extensions, module: windows",confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-06-27T00:43:46Z
issue,155060,MPS Memory Leak,https://github.com/pytorch/pytorch/issues/155060,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: memory usage, module: mps",confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-06-13T16:57:04Z
issue,152783,"undefined symbol: __nvJitLinkCreate_12_8, version libnvJitLink.so.12",https://github.com/pytorch/pytorch/issues/152783,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: binaries,confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-05-29T13:04:38Z
issue,154302,Inconsistent output of `torch.func.jvp` calculation,https://github.com/pytorch/pytorch/issues/154302,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: functorch",confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-05-27T17:18:30Z
issue,154266,torch dynamo fails on GH200 with world size 5,https://github.com/pytorch/pytorch/issues/154266,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: dtensor, oncall: distributed, oncall: pt2",confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-05-27T15:02:59Z
issue,153564,Sum difference for equal channels of tensor,https://github.com/pytorch/pytorch/issues/153564,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label,confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-05-23T17:37:40Z
issue,143646,"Getting ""Could not initialize NNPACK! Reason: Unsupported hardware."" warning even though NNPACK is enabled",https://github.com/pytorch/pytorch/issues/143646,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: nnpack,confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-05-22T12:59:24Z
issue,152790,[CXX11ABI] torch 2.6.0-cu126 and cu124 have different exported symbols,https://github.com/pytorch/pytorch/issues/152790,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: cuda",confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-05-16T13:12:13Z
issue,153574,Documentation Preview and Build Not Reflecting Latest Changes,https://github.com/pytorch/pytorch/issues/153574,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: build, module: doc infra, module: docs",confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-05-15T16:00:10Z
issue,153033,missalignment with differenet shape in F.linear with bf16 dtype,https://github.com/pytorch/pytorch/issues/153033,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: bfloat16, module: linear algebra, module: padding",confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-05-09T14:42:04Z
issue,90752,Adam (fused=True) issues,https://github.com/pytorch/pytorch/issues/90752,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: optimizer, module: performance",confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-05-02T21:36:34Z
issue,152592,[ratter-build] Cannot detect CUDA when build from source,https://github.com/pytorch/pytorch/issues/152592,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: build,confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-05-01T14:32:14Z
issue,152135,Unaccaptable OOMs all the time.,https://github.com/pytorch/pytorch/issues/152135,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: memory usage",confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-04-28T17:18:23Z
issue,149508,Adam optimizer ValueError: beta1 as a Tensor,https://github.com/pytorch/pytorch/issues/149508,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: optimizer,confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-04-26T04:45:40Z
issue,150912,using torch.compile with torchao at the same time cause stack overflow error,https://github.com/pytorch/pytorch/issues/150912,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-04-21T21:33:10Z
issue,128290,[apple M2]libc++abi: terminating due to uncaught exception of type c10::Error: Could not find schema for aten::randn,https://github.com/pytorch/pytorch/issues/128290,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: build, module: cpp, module: macos",confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-04-16T00:45:05Z
issue,151201,I can't compile Pytorch 2.0.0 because ninja: error: build.ninja:10911: multiple rules generate caffe2/CMakeFiles/torch_cpu.dir/__/aten/src/ATen/UfuncCPUKernel_add.cpp.DEFAULT.cpp.o,https://github.com/pytorch/pytorch/issues/151201,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: build,confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-04-14T14:59:25Z
issue,119546,Too many spawn process,https://github.com/pytorch/pytorch/issues/119546,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: multiprocessing,confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-04-14T03:42:51Z
issue,150921,`M(*[torch.from_numpy(v).to('cpu') for v in inp])` hang when start with `multiprocessing.Process`,https://github.com/pytorch/pytorch/issues/150921,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: multiprocessing,confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-04-10T02:03:07Z
issue,149279,CUDA Assertion Error in Scatter Operation During Training (RTX5090 cu128),https://github.com/pytorch/pytorch/issues/149279,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: scatter & gather ops, module: windows",confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-04-09T22:05:39Z
issue,71545,pip installation SSLError [SSL: CERTIFICATE_VERIFY_FAILED],https://github.com/pytorch/pytorch/issues/71545,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: binaries,confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-04-02T00:46:41Z
issue,149937,torch.inverse can't export onnx opset16,https://github.com/pytorch/pytorch/issues/149937,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: onnx,confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-03-26T01:27:59Z
issue,130486,"RuntimeError: NVML_SUCCESS == DriverAPI::get()->nvmlInit_v2_() INTERNAL ASSERT FAILED at ""../c10/cuda/CUDACachingAllocator.cpp"":813, please report a bug to PyTorch.",https://github.com/pytorch/pytorch/issues/130486,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda,confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-03-09T08:17:43Z
issue,148169,"PyTorch's nightly version no longer includes the CU118, CU124, and CU121 versions",https://github.com/pytorch/pytorch/issues/148169,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: cuda",confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-03-03T18:28:27Z
issue,146579,Pytorch model (.pth format) not loading in KServe environment,https://github.com/pytorch/pytorch/issues/146579,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: serialization,confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-02-07T12:07:52Z
issue,145827,What changed in torch 2.5.0 that cause photomaker to fail?,https://github.com/pytorch/pytorch/issues/145827,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label,confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-01-29T16:31:42Z
issue,47128,AMP doesn't gracefully handle optimizers for disabled regions,https://github.com/pytorch/pytorch/issues/47128,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision),confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-01-17T19:09:54Z
issue,130728,torch.cuda.empty_cache() doesn't work correctly for the first instance of nn.Module set to cuda (and a potential method to deal with that).,https://github.com/pytorch/pytorch/issues/130728,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: memory usage",confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-01-16T14:31:20Z
issue,143494,sympy.C.ConstantInteger has no method name,https://github.com/pytorch/pytorch/issues/143494,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: dynamic shapes, module: fx, oncall: pt2",confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-01-13T17:17:15Z
issue,144071,torch-nightly doesn't support tesla v100,https://github.com/pytorch/pytorch/issues/144071,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: cuda",confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-01-10T04:06:17Z
issue,111547,Bug with as_strided_tensorimpl for MPS devices,https://github.com/pytorch/pytorch/issues/111547,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: mps,confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-01-09T19:15:13Z
issue,126862,"worse results by using the MPS backend, compared to the CPU",https://github.com/pytorch/pytorch/issues/126862,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: mps",confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-01-09T18:26:34Z
issue,143906,How to correctly asynchronously copy a GPU tensor to a CPU tensor in another process without introducing blocking?,https://github.com/pytorch/pytorch/issues/143906,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-01-03T18:13:46Z
issue,116766,[Dynamo][DeepSpeed] torch._dynamo.exc.InternalTorchDynamoError: NestedUserFunctionVariable() has no type,https://github.com/pytorch/pytorch/issues/116766,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: distributed, oncall: pt2",confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-01-02T18:21:36Z
issue,143852,[CPU][Operator] one channel_shuffle test of the operator benchmark has a Performance fluctuation issue,https://github.com/pytorch/pytorch/issues/143852,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: nn, module: performance",confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-01-02T06:39:46Z
issue,143905,The special size tensor containing batches has a difference of a few tens of thousands in calculation results between CPU and GPU,https://github.com/pytorch/pytorch/issues/143905,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda,confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-12-30T00:50:14Z
issue,143800,The tensor-based computation of exponentiation and logarithmic operations is much slower than using NumPy,https://github.com/pytorch/pytorch/issues/143800,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: performance",confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-12-26T14:53:30Z
issue,143651,`MemoryDenyWriteExecute` in systemd service causes `RuntimeError: could not create a primitive`,https://github.com/pytorch/pytorch/issues/143651,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: error checking, module: mkldnn",confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-12-23T23:37:38Z
issue,110946,RuntimeError: out_ptr == out_accessor[thread_count_nonzero[tid + 1]].data() INTERNAL ASSERT FAILED,https://github.com/pytorch/pytorch/issues/110946,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: advanced indexing,confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-12-22T04:23:10Z
issue,143338,c10::SmallVector unusable with gcc/g++ 12 and 13 with `-O3`,https://github.com/pytorch/pytorch/issues/143338,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: build, module: cpp",confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-12-18T03:16:39Z
issue,143288,EXCEPTION : /python3.11/distutils/core.py,https://github.com/pytorch/pytorch/issues/143288,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: third_party, oncall: pt2",confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-12-17T08:04:09Z
issue,143259,s.isIntegral(false) INTERNAL ASSERT FAILED,https://github.com/pytorch/pytorch/issues/143259,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: autograd,confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-12-16T16:20:57Z
issue,137393,Cannot allocate memory for thread-local data: ABORT,https://github.com/pytorch/pytorch/issues/137393,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: memory usage,confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-12-14T15:06:49Z
issue,142302,torch.nn.functional.normalize get nan,https://github.com/pytorch/pytorch/issues/142302,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: nn,confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-12-13T19:47:58Z
issue,139709,less GPU memory used in Centos than ubuntu,https://github.com/pytorch/pytorch/issues/139709,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: memory usage,confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-12-06T22:52:26Z
issue,126141,Inconsistent gradients in reverse mode and forward mode AD when using `MultiheadAttention`,https://github.com/pytorch/pytorch/issues/126141,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: nn",confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-11-27T20:02:08Z
issue,126388,RAM Not Freed on CPU After Moving Model with Multiple Transformers to CUDA,https://github.com/pytorch/pytorch/issues/126388,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: memory usage, module: nn",confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-11-27T20:02:07Z
issue,141465,Inconsistent Bias Broadcasting Results in PyTorch GPU `F.linear(...)` and `F.conv2d(...)` Operations,https://github.com/pytorch/pytorch/issues/141465,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: cudnn",confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-11-27T16:34:05Z
issue,132226,关于yolov5在mac设备上使用pytorch的mps加速出现的各种问题,https://github.com/pytorch/pytorch/issues/132226,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: mps,confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-11-27T07:56:28Z
issue,141482,Can't use dataparallel,https://github.com/pytorch/pytorch/issues/141482,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: data parallel,confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-11-26T01:45:33Z
issue,140096,upsample_bilinear backward is super slow in bf16 as compared to fp32,https://github.com/pytorch/pytorch/issues/140096,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: nn, module: performance",confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-11-15T07:41:36Z
issue,139977,Assigning the weight of the torch.nn.Embedding object to other variables and then performing subsequent operations on it will cause unstable training,https://github.com/pytorch/pytorch/issues/139977,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: nn,confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-11-13T02:37:14Z
issue,140282,torch.norm produces wrong results,https://github.com/pytorch/pytorch/issues/140282,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: linear algebra,confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-11-12T00:19:31Z
issue,139884,Persistent memory leak from failed pinned memory allocation,https://github.com/pytorch/pytorch/issues/139884,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: memory format, module: memory usage",confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-11-11T18:11:11Z
issue,111359,I have a trouble with to_symmetric,https://github.com/pytorch/pytorch/issues/111359,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: sparse,confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-11-06T06:29:21Z
issue,137078,"Missing librt.so.1, libdl.so.2, libz.so.1 and libpthread.so.0 in PyTorch Source Build (aarch64)",https://github.com/pytorch/pytorch/issues/137078,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: arm, module: build",confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-11-05T05:24:01Z
issue,34948,TracerWarning: Output nr 1. of the traced function does not match the corresponding output of the Python function,https://github.com/pytorch/pytorch/issues/34948,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-11-01T05:52:54Z
issue,124197,Libtorch crashes docker when included in header file,https://github.com/pytorch/pytorch/issues/124197,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: abi,confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-10-31T17:18:50Z
issue,138821,"Failed to load image Python extension: libc10_cuda.so: cannot open shared object file: No such file or directory warn(f""Failed to load image Python extension: {e}"")",https://github.com/pytorch/pytorch/issues/138821,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: binaries,confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-10-25T23:00:05Z
issue,138455,Tried to instantiate dummy base class,https://github.com/pytorch/pytorch/issues/138455,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: testing,confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-10-23T15:05:36Z
issue,130530,Fail to offload FSDP model weights and optimizer states without using CPUOffload(offload_params=True),https://github.com/pytorch/pytorch/issues/130530,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-10-23T14:53:30Z
issue,121067,RuntimeError,https://github.com/pytorch/pytorch/issues/121067,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda,confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-10-22T18:39:21Z
issue,138208,NaN value produced by F.conv2d (cuda) with large values in input,https://github.com/pytorch/pytorch/issues/138208,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: nans and infs",confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-10-22T07:40:01Z
issue,137964,Network training silently fails with MPS backend,https://github.com/pytorch/pytorch/issues/137964,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: mps",confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-10-18T19:52:17Z
issue,75949,Potential memory leak in Adam optimizer in AMD chips (CPU),https://github.com/pytorch/pytorch/issues/75949,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: memory usage, module: optimizer",confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-10-11T18:34:16Z
issue,87550,[onnx] export repeat_interleave TypeError: z_(): incompatible function arguments,https://github.com/pytorch/pytorch/issues/87550,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: onnx,confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-09-19T03:33:05Z
issue,135693,Python's REPL trying to do invoke `.T` / `.mH` / `.mT` on 0-dim tensor when doing tab-completion for some reason (default install of Python on new Ubuntu 24.04),https://github.com/pytorch/pytorch/issues/135693,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label,confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-09-13T22:10:37Z
issue,94855,torch.compile breaks reproducibility,https://github.com/pytorch/pytorch/issues/94855,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: determinism, module: inductor, module: random, oncall: pt2",confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-09-12T18:41:14Z
issue,134993,[bug] import torch failed after install the cpu version of pytorch,https://github.com/pytorch/pytorch/issues/134993,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: binaries,confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-09-04T00:40:33Z
issue,133860,torch::tensor function does not work,https://github.com/pytorch/pytorch/issues/133860,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: cpp,confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-08-24T05:08:19Z
issue,104598,ERROR: THESE PACKAGES DO NOT MATCH THE HASHES FROM THE REQUIREMENTS FILE.,https://github.com/pytorch/pytorch/issues/104598,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: binaries,confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-08-21T12:32:49Z
issue,133808,pytorch/aten/src/ATen/cuda/CUDABlas.cpp:1422:28: error,https://github.com/pytorch/pytorch/issues/133808,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: build, module: cuda",confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-08-20T01:25:47Z
issue,133093,Inconsistent results of Conv2d in pytorch between CPU and CUDA,https://github.com/pytorch/pytorch/issues/133093,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: cuda",confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-08-19T17:28:18Z
issue,114455,Torch Cpu Memory Leak with FastApi uvicorn.,https://github.com/pytorch/pytorch/issues/114455,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: memory usage,confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-08-13T21:37:36Z
issue,131185,"RuntimeError: Expected grad_output.numel() <= std::numeric_limits<int32_t>::max() to be true, but got false.",https://github.com/pytorch/pytorch/issues/131185,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: 64-bit, module: autograd, module: error checking, module: nn",confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-08-13T15:49:47Z
issue,30507,"LibTorch, Error in 'xxx': free(): invalid pointer",https://github.com/pytorch/pytorch/issues/30507,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: cpp,confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-08-01T03:26:50Z
issue,120606,“libc10.dylib” can’t be opened because Apple cannot check it for malicious software.,https://github.com/pytorch/pytorch/issues/120606,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: docs, module: macos",confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-07-28T09:58:02Z
issue,58392,torch.cuda.amp.autocast nan values when model.eval(),https://github.com/pytorch/pytorch/issues/58392,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision),confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-07-10T15:23:19Z
issue,129643,"There is a bug, only when self",https://github.com/pytorch/pytorch/issues/129643,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: loss, module: nn",confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-07-05T03:23:23Z
issue,128939,Unable to install PyTorch on M1 Macos with Python 3.10.14,https://github.com/pytorch/pytorch/issues/128939,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: macos",confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-07-01T08:05:33Z
issue,129303,Incorrect index from torch.mode,https://github.com/pytorch/pytorch/issues/129303,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: reductions,confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-06-26T16:40:33Z
issue,128564,internal asset failure,https://github.com/pytorch/pytorch/issues/128564,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: linear algebra,confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-06-13T15:47:41Z
issue,127560,[libtorch] duplicate symbol of vtable,https://github.com/pytorch/pytorch/issues/127560,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: cpp",confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-06-10T17:28:02Z
issue,126775,The significant difference between the outputs of torch.reciprocal running on cpu and gpu,https://github.com/pytorch/pytorch/issues/126775,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: numerical-stability",confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-06-08T09:07:01Z
issue,127978,torch.addr outputs NaN when computing on inf vectors,https://github.com/pytorch/pytorch/issues/127978,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: complex, module: nans and infs",confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-06-06T16:56:07Z
issue,128131,Unable to record Memory consumption with `torch.cuda.memory._record_memory_history()`,https://github.com/pytorch/pytorch/issues/128131,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda,confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-06-06T15:38:15Z
issue,121222,Segmentation Fault in `torch.nn.Conv1d` starting from torch 2.2.0,https://github.com/pytorch/pytorch/issues/121222,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: cuda",confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-05-27T17:36:30Z
issue,50779,pytorch is built without _GLIBCXX_USE_CXX11_ABI and can cause std::regex crashes (probably),https://github.com/pytorch/pytorch/issues/50779,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: build,confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-05-26T01:10:33Z
issue,126760,"""torch.orgqr"" performs differently on cpu and gpu",https://github.com/pytorch/pytorch/issues/126760,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: nans and infs,confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-05-21T23:07:24Z
issue,125754,Excution difference between Mac and Linux,https://github.com/pytorch/pytorch/issues/125754,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: arm, module: macos",confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-05-11T00:05:08Z
issue,125911,"set_device只支持int,但如果碰到k8s分卡vcuda时报错segmentation fault",https://github.com/pytorch/pytorch/issues/125911,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda,confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-05-10T11:06:46Z
issue,47990,Is nn.Conv2d equivalent with Unfold + Matrix Multiplication + Fold ?,https://github.com/pytorch/pytorch/issues/47990,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: correctness (silent)",confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-05-02T17:56:18Z
issue,122707,unresovled symbol when compiling a torch c++ app that using libtorch,https://github.com/pytorch/pytorch/issues/122707,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: cpp,confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-04-12T21:40:38Z
issue,117617,"torch.topk, when the input contains +nan and -nan, the result is not sure; sometimes -nan is treated greater than normal number, sometimes -nan is the least;",https://github.com/pytorch/pytorch/issues/117617,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: python frontend,confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-04-10T07:19:06Z
issue,5953,"LBFGS always give nan results, why",https://github.com/pytorch/pytorch/issues/5953,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: numerical-stability, module: optimizer",confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-04-02T07:54:21Z
issue,122548,Unexpected device casting from CPU-pinned tensor to GPU tensor.,https://github.com/pytorch/pytorch/issues/122548,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: python frontend,confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-03-29T02:20:22Z
issue,71222,torch.linalg.lstsq is nondeterministic,https://github.com/pytorch/pytorch/issues/71222,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: determinism, module: linear algebra",confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-03-21T03:09:31Z
issue,121646,Installation fails on python 3.7-3.12,https://github.com/pytorch/pytorch/issues/121646,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: binaries,confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-03-16T18:21:12Z
issue,120905,F.conv2d double precision error,https://github.com/pytorch/pytorch/issues/120905,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: nn",confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-03-11T17:25:18Z
issue,121287,Libtorch - clang: error: unsupported option '-fopenmp',https://github.com/pytorch/pytorch/issues/121287,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: cpp, module: macos",confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-03-07T15:10:28Z
issue,121190,The result of BF16 to FP32 is different in class LlamaRMSNorm training twice.,https://github.com/pytorch/pytorch/issues/121190,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: bfloat16, module: numerical-stability",confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-03-07T05:50:46Z
issue,120791,about tensor.new(storage) issue,https://github.com/pytorch/pytorch/issues/120791,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: python frontend,confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-03-06T17:43:12Z
issue,70048,Tensor transfer between gpus doesnt work,https://github.com/pytorch/pytorch/issues/70048,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda,confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-02-27T23:06:31Z
issue,120167,"When loading a text file with torch.load, it exhausts the system's physical memory.",https://github.com/pytorch/pytorch/issues/120167,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: memory usage, module: serialization",confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-02-27T17:15:05Z
issue,119147,Incorrect Results from cdist_backward Operator with Input Dimensions Exceeding int32_max,https://github.com/pytorch/pytorch/issues/119147,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: 64-bit, module: edge cases",confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-02-21T07:47:34Z
issue,120081,undefined reference to `__atomic_exchange_1',https://github.com/pytorch/pytorch/issues/120081,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: build,confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-02-17T02:36:30Z
issue,118263,Getting error while loading LLAMA_2_7B,https://github.com/pytorch/pytorch/issues/118263,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label,confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-01-25T18:18:17Z
issue,118172,"""PyTorch installation failed.""",https://github.com/pytorch/pytorch/issues/118172,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: cuda",confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-01-24T15:41:02Z
issue,117962,runtime error: unknown parameter,https://github.com/pytorch/pytorch/issues/117962,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label,confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-01-24T03:27:34Z
issue,117210,"When running the code model.to('mps'), the program gets stuck, but it runs fine on the CPU",https://github.com/pytorch/pytorch/issues/117210,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: mps,confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-01-19T07:15:50Z
issue,117469,"Cannot import torch on Ubuntu 20.04 from clean pip install (shared library issues, libcufft and libcudart)",https://github.com/pytorch/pytorch/issues/117469,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: cuda",confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-01-16T18:43:51Z
issue,73568,Bug when using `nn.Linear`,https://github.com/pytorch/pytorch/issues/73568,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: linear algebra",confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-01-15T12:23:40Z
issue,116931,Previous Torch versions,https://github.com/pytorch/pytorch/issues/116931,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: docs",confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-01-12T18:12:15Z
issue,116589,Cannot pickle torch model in version 2.1 when using `autograd.Function` as a model's attribute,https://github.com/pytorch/pytorch/issues/116589,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: serialization, module: third_party",confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-01-10T11:59:45Z
issue,116882,RuntimeError: CUDA out of memory.,https://github.com/pytorch/pytorch/issues/116882,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label,confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-01-09T20:58:03Z
issue,117032,losing precision in torch.sum,https://github.com/pytorch/pytorch/issues/117032,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label,confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-01-09T20:32:31Z
issue,114465,Installation error: 'CMake Error at third_party/benchmark/CMakeLists.txt:304 (message): Failed to determine the source files for the regular expression backend',https://github.com/pytorch/pytorch/issues/114465,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: build,confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-01-02T10:05:50Z
issue,50803,pytorch DDP hangs at .backward() call,https://github.com/pytorch/pytorch/issues/50803,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: ddp, module: deadlock",confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-01-01T18:09:04Z
issue,48580,torch.save() fails when attempting to save to mounted drives,https://github.com/pytorch/pytorch/issues/48580,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: serialization,confirm whether existing evidence is enough; otherwise wait for a reproducer,2023-12-23T00:48:38Z
issue,114534,`PYTORCH_NO_CUDA_MEMORY_CACHING=1` with `torch.multiprocessing` shared tensors seems to perform use-after-free,https://github.com/pytorch/pytorch/issues/114534,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: multiprocessing",confirm whether existing evidence is enough; otherwise wait for a reproducer,2023-12-05T16:24:37Z
issue,114572,"In version 2.1, libtorch needs to be woken up every time it is called after the model is initialized, which means that every time the model is called, it is very slow to predict the first picture.",https://github.com/pytorch/pytorch/issues/114572,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: cpp,confirm whether existing evidence is enough; otherwise wait for a reproducer,2023-11-27T04:56:17Z
issue,113840,torch.einsum is stuck in mp.Process,https://github.com/pytorch/pytorch/issues/113840,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: multiprocessing,confirm whether existing evidence is enough; otherwise wait for a reproducer,2023-11-21T03:19:19Z
issue,113150,"Aborted (core dumped) after Run Pytorch2.0.0, which I compiled by myself",https://github.com/pytorch/pytorch/issues/113150,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: build,confirm whether existing evidence is enough; otherwise wait for a reproducer,2023-11-13T17:21:05Z
issue,111158,RuntimeError in run_streaming_llama.py When Using Accelerate with Streaming LLMa Model on A4500 GPU,https://github.com/pytorch/pytorch/issues/111158,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: error checking,confirm whether existing evidence is enough; otherwise wait for a reproducer,2023-11-07T02:02:40Z
issue,113035,Training a network SUPER slow with Pytorch 2.1,https://github.com/pytorch/pytorch/issues/113035,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: performance",confirm whether existing evidence is enough; otherwise wait for a reproducer,2023-11-06T22:35:20Z
issue,111508,LBFGS accuracy difference between CPU and GPU,https://github.com/pytorch/pytorch/issues/111508,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: optimizer,confirm whether existing evidence is enough; otherwise wait for a reproducer,2023-11-06T21:59:26Z
issue,85211,OSError libstdc++.so.6 at import,https://github.com/pytorch/pytorch/issues/85211,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: binaries,confirm whether existing evidence is enough; otherwise wait for a reproducer,2023-11-02T13:58:23Z
issue,112569,"RuntimeError: NVML_SUCCESS == r INTERNAL ASSERT FAILED at ""/opt/conda/conda-bld/pytorch_1695392035891/work/c10/cuda/CUDACachingAllocator.cpp"":1154, please report a bug to PyTorch",https://github.com/pytorch/pytorch/issues/112569,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: error checking",confirm whether existing evidence is enough; otherwise wait for a reproducer,2023-11-01T22:02:13Z
issue,111419,nonnull error,https://github.com/pytorch/pytorch/issues/111419,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: build,confirm whether existing evidence is enough; otherwise wait for a reproducer,2023-10-31T18:17:17Z
issue,111896,ImportError: cannot import name 'external_utils' from partially initialized module 'torch._dynamo',https://github.com/pytorch/pytorch/issues/111896,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: docker",confirm whether existing evidence is enough; otherwise wait for a reproducer,2023-10-30T17:13:17Z
issue,110944,gdb core dump when enable DEBUG mode to compile cpu torch in centos!!!,https://github.com/pytorch/pytorch/issues/110944,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label,confirm whether existing evidence is enough; otherwise wait for a reproducer,2023-10-12T13:14:17Z
issue,110701,Error with monai SwinUNETR and checkpointing,https://github.com/pytorch/pytorch/issues/110701,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda,confirm whether existing evidence is enough; otherwise wait for a reproducer,2023-10-12T08:51:37Z
issue,110883,MAX_JOBS ignored when compiling pytorch from source,https://github.com/pytorch/pytorch/issues/110883,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: build,confirm whether existing evidence is enough; otherwise wait for a reproducer,2023-10-10T17:49:36Z
issue,110594,Multiprocessing takes forever after on .get() with mp.Queue() (Possible Deadlock),https://github.com/pytorch/pytorch/issues/110594,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: multiprocessing,confirm whether existing evidence is enough; otherwise wait for a reproducer,2023-10-06T10:57:56Z
issue,110588,libtorch.so: error adding symbols: file in wrong format,https://github.com/pytorch/pytorch/issues/110588,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: build,confirm whether existing evidence is enough; otherwise wait for a reproducer,2023-10-05T14:40:29Z
issue,108493,RWKV + Adam exp_avg_sq will change from positive to negative after loss.backward(),https://github.com/pytorch/pytorch/issues/108493,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: optimizer,confirm whether existing evidence is enough; otherwise wait for a reproducer,2023-09-27T18:59:36Z
issue,108676,RuntimeError when calling conv_transpose2d with groups,https://github.com/pytorch/pytorch/issues/108676,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: nn,confirm whether existing evidence is enough; otherwise wait for a reproducer,2023-09-27T13:27:19Z
issue,104857,Torch's `LayerNorm` and Adam optimizer vs those in tensorflow,https://github.com/pytorch/pytorch/issues/104857,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: nn, module: numerical-stability, module: optimizer",confirm whether existing evidence is enough; otherwise wait for a reproducer,2023-09-22T18:26:41Z
issue,109453,Inconsistent Behavior of `ConvTranspose2d` on CPU and CUDA,https://github.com/pytorch/pytorch/issues/109453,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: nn,confirm whether existing evidence is enough; otherwise wait for a reproducer,2023-09-22T15:31:53Z
issue,109193,"F.conv2d(input, weight, bias, self.stride, RuntimeError: cuDNN error: CUDNN_STATUS_MAPPING_ERROR",https://github.com/pytorch/pytorch/issues/109193,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: cudnn, module: nn",confirm whether existing evidence is enough; otherwise wait for a reproducer,2023-09-14T14:45:17Z
issue,86547,Segmentation fault (core dumped) in RTX3090,https://github.com/pytorch/pytorch/issues/86547,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda,confirm whether existing evidence is enough; otherwise wait for a reproducer,2023-09-04T06:24:01Z
issue,108013,Installation with rocm5.6 results in error: assert len(weights) == expected_node_count AssertionError,https://github.com/pytorch/pytorch/issues/108013,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: build, module: rocm",confirm whether existing evidence is enough; otherwise wait for a reproducer,2023-08-28T16:01:56Z
issue,107276,CUBLAS_STATUS_NOT_SUPPORTED,https://github.com/pytorch/pytorch/issues/107276,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: cublas, module: cuda",confirm whether existing evidence is enough; otherwise wait for a reproducer,2023-08-17T03:06:54Z
issue,107115,H100 works differently than rtx4090 on same model,https://github.com/pytorch/pytorch/issues/107115,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: numerical-reproducibility,confirm whether existing evidence is enough; otherwise wait for a reproducer,2023-08-15T12:26:15Z
issue,107133,torch.inverse throws error when DP but not in DDP or single GPU,https://github.com/pytorch/pytorch/issues/107133,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: data parallel, module: edge cases, module: linear algebra",confirm whether existing evidence is enough; otherwise wait for a reproducer,2023-08-14T18:46:45Z
issue,21018,free(): invalid pointer Aborted (core dumped),https://github.com/pytorch/pytorch/issues/21018,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label,confirm whether existing evidence is enough; otherwise wait for a reproducer,2023-08-04T10:01:58Z
issue,105941,Differences in the results of conv2d calculations in PyTorch 1.8,https://github.com/pytorch/pytorch/issues/105941,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: cudnn,confirm whether existing evidence is enough; otherwise wait for a reproducer,2023-07-29T15:38:11Z
issue,106126,Libtorch report C10 error when compiling on my own project,https://github.com/pytorch/pytorch/issues/106126,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: cpp",confirm whether existing evidence is enough; otherwise wait for a reproducer,2023-07-28T21:53:43Z
issue,2539,"DataLoader converts cuda FloatTensor into cpu DoubleTensor when shape is (n,)",https://github.com/pytorch/pytorch/issues/2539,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: dataloader,confirm whether existing evidence is enough; otherwise wait for a reproducer,2023-07-25T20:31:08Z
issue,104284,`F.conv1d` and `F.conv2d` propagate `nan`'s incorrectly when minibatch > 15,https://github.com/pytorch/pytorch/issues/104284,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: arm, module: macos, module: nans and infs, module: nn",confirm whether existing evidence is enough; otherwise wait for a reproducer,2023-07-20T14:26:06Z
issue,55663,torch.nn.functional.log_softmax on Arm CPU gives partly NaN results,https://github.com/pytorch/pytorch/issues/55663,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: arm,confirm whether existing evidence is enough; otherwise wait for a reproducer,2023-07-19T19:10:40Z
issue,104587,Conv1d step-by-step numerical error,https://github.com/pytorch/pytorch/issues/104587,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: arm, module: determinism, module: macos",confirm whether existing evidence is enough; otherwise wait for a reproducer,2023-07-17T17:35:33Z
issue,103752,Pytorch not calling to C code from a docker container,https://github.com/pytorch/pytorch/issues/103752,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: __torch_function__, module: docker",confirm whether existing evidence is enough; otherwise wait for a reproducer,2023-07-03T02:41:55Z
issue,91942,jacrev over huber function,https://github.com/pytorch/pytorch/issues/91942,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: functorch,confirm whether existing evidence is enough; otherwise wait for a reproducer,2023-06-23T09:27:21Z
issue,103423,The document does not emphasize hidden range in nn.MaxPool2d,https://github.com/pytorch/pytorch/issues/103423,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: docs,confirm whether existing evidence is enough; otherwise wait for a reproducer,2023-06-18T09:19:33Z
issue,103424,The document does not emphasize hidden range in nn.Embedding,https://github.com/pytorch/pytorch/issues/103424,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: docs,confirm whether existing evidence is enough; otherwise wait for a reproducer,2023-06-18T08:36:41Z
issue,102078,AddressSanitizer: heap-buffer-overflow in test_comprehensive_nn_functional_embedding_bag_cpu_bfloat16,https://github.com/pytorch/pytorch/issues/102078,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: decompositions, module: embedding",confirm whether existing evidence is enough; otherwise wait for a reproducer,2023-05-31T04:10:02Z
issue,101404,problem of compilation for torch2.0,https://github.com/pytorch/pytorch/issues/101404,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: build, module: third_party",confirm whether existing evidence is enough; otherwise wait for a reproducer,2023-05-24T04:11:38Z
issue,63847,RuntimError:CUDA error: an illegal memory access was encountered when copying data to CUDA,https://github.com/pytorch/pytorch/issues/63847,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda,confirm whether existing evidence is enough; otherwise wait for a reproducer,2023-05-22T03:05:08Z
issue,101880,Crash on Python // PyArrow //,https://github.com/pytorch/pytorch/issues/101880,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label,confirm whether existing evidence is enough; otherwise wait for a reproducer,2023-05-19T20:35:07Z
issue,101331,Pytorch 2.1.0.dev20230512 cuda not available,https://github.com/pytorch/pytorch/issues/101331,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: cuda",confirm whether existing evidence is enough; otherwise wait for a reproducer,2023-05-16T01:13:03Z
issue,101321,Speed when installing from source is very low with CUDA 11,https://github.com/pytorch/pytorch/issues/101321,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: build, module: cuda, module: performance",confirm whether existing evidence is enough; otherwise wait for a reproducer,2023-05-15T14:14:16Z
issue,99268,Embedding layer tensor shape,https://github.com/pytorch/pytorch/issues/99268,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: embedding,confirm whether existing evidence is enough; otherwise wait for a reproducer,2023-04-17T15:52:07Z
issue,25190,Error in python3: double free or corruption (fasttop),https://github.com/pytorch/pytorch/issues/25190,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: cudnn, module: dataloader, module: multiprocessing",confirm whether existing evidence is enough; otherwise wait for a reproducer,2023-04-10T14:05:31Z
issue,98142,double free or corruption (fasttop),https://github.com/pytorch/pytorch/issues/98142,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label,confirm whether existing evidence is enough; otherwise wait for a reproducer,2023-04-10T13:45:17Z
issue,98204,NaN appears when initializing tensor,https://github.com/pytorch/pytorch/issues/98204,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: nans and infs,confirm whether existing evidence is enough; otherwise wait for a reproducer,2023-04-04T13:56:02Z
issue,97872,nn.linear not support bfloat16,https://github.com/pytorch/pytorch/issues/97872,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,confirm whether existing evidence is enough; otherwise wait for a reproducer,2023-03-31T15:19:48Z
issue,97865,Unable to install torch on python 3.8.16,https://github.com/pytorch/pytorch/issues/97865,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: binaries,confirm whether existing evidence is enough; otherwise wait for a reproducer,2023-03-30T06:55:34Z
issue,97847,CUDA 10.2 cudnn 8.2.4 run Conv2d error,https://github.com/pytorch/pytorch/issues/97847,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: cudnn",confirm whether existing evidence is enough; otherwise wait for a reproducer,2023-03-29T14:20:31Z
issue,97135,Incorrect gradient calculation for upsample nearest on CUDA,https://github.com/pytorch/pytorch/issues/97135,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: interpolation, module: nn",confirm whether existing evidence is enough; otherwise wait for a reproducer,2023-03-24T19:28:09Z
issue,80588,Semi-reproducible random torch.baddbmm NaNs,https://github.com/pytorch/pytorch/issues/80588,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: nans and infs,confirm whether existing evidence is enough; otherwise wait for a reproducer,2023-03-06T14:17:03Z
issue,95677,Unable to import ``torch.linalg``,https://github.com/pytorch/pytorch/issues/95677,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label,confirm whether existing evidence is enough; otherwise wait for a reproducer,2023-03-02T16:35:35Z
issue,90115,Couldn't install pytorch 2.0,https://github.com/pytorch/pytorch/issues/90115,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: binaries,confirm whether existing evidence is enough; otherwise wait for a reproducer,2023-02-16T17:03:28Z
issue,93900,"Why does the torch model have no memory leaks under gpu, but there is a memory leak under cpu, torch version 1.10.1",https://github.com/pytorch/pytorch/issues/93900,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label,confirm whether existing evidence is enough; otherwise wait for a reproducer,2023-02-16T15:07:32Z
issue,94542,torch.lgamma CUDA driver error,https://github.com/pytorch/pytorch/issues/94542,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: special,confirm whether existing evidence is enough; otherwise wait for a reproducer,2023-02-13T22:33:41Z
issue,92794,Segmentation fault when running torch.ge,https://github.com/pytorch/pytorch/issues/92794,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label,confirm whether existing evidence is enough; otherwise wait for a reproducer,2023-01-27T02:08:38Z
issue,91245,Segmentation faults in DataLoader (in latest torch version).,https://github.com/pytorch/pytorch/issues/91245,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: dataloader,confirm whether existing evidence is enough; otherwise wait for a reproducer,2023-01-23T07:40:12Z
issue,91970,Unknown CUDA graph CaptureStatus21852,https://github.com/pytorch/pytorch/issues/91970,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label,confirm whether existing evidence is enough; otherwise wait for a reproducer,2023-01-10T22:41:57Z
issue,91072,Unexpected behavior when running torch.max in cuda,https://github.com/pytorch/pytorch/issues/91072,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda,confirm whether existing evidence is enough; otherwise wait for a reproducer,2022-12-20T18:03:42Z
issue,90521,[A ERROR in Docker] RuntimeError: CUDA error: no kernel image is available for execution on the device,https://github.com/pytorch/pytorch/issues/90521,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: docker",confirm whether existing evidence is enough; otherwise wait for a reproducer,2022-12-09T17:31:10Z
issue,90138,p.block != nullptr && p.block->ptr != nullptr INTERNAL ASSERT FAILED,https://github.com/pytorch/pytorch/issues/90138,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label,confirm whether existing evidence is enough; otherwise wait for a reproducer,2022-12-06T18:16:47Z
issue,88948,torch.linalg.matrix_rank memory leak,https://github.com/pytorch/pytorch/issues/88948,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: linear algebra, module: memory usage",confirm whether existing evidence is enough; otherwise wait for a reproducer,2022-11-16T14:04:53Z
issue,89041,"cross compile pytoch using cmake , get an error : protobuf::protoc: command not found",https://github.com/pytorch/pytorch/issues/89041,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: build,confirm whether existing evidence is enough; otherwise wait for a reproducer,2022-11-16T00:18:47Z
issue,87597,ninja: build stopped: subcommand failed,https://github.com/pytorch/pytorch/issues/87597,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: build,confirm whether existing evidence is enough; otherwise wait for a reproducer,2022-10-24T16:04:50Z
issue,86684,[ONNX] AssertionError: A mismatch between the number of arguments (5) and their descriptors (4) was found at symbolic function 'scatter',https://github.com/pytorch/pytorch/issues/86684,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: onnx,confirm whether existing evidence is enough; otherwise wait for a reproducer,2022-10-15T22:21:35Z
issue,86877,compile torch from source,https://github.com/pytorch/pytorch/issues/86877,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: build,confirm whether existing evidence is enough; otherwise wait for a reproducer,2022-10-14T02:34:57Z
issue,86919,Importing torch 1.12.0 breaks subprocess module,https://github.com/pytorch/pytorch/issues/86919,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: macos",confirm whether existing evidence is enough; otherwise wait for a reproducer,2022-10-13T18:45:45Z
issue,75458,Placing model on bfloat16 on CPU make it freeze/hang,https://github.com/pytorch/pytorch/issues/75458,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: bfloat16, module: deadlock, module: intel",confirm whether existing evidence is enough; otherwise wait for a reproducer,2022-10-07T21:50:10Z
issue,84705,Set dtype if tensor converted to numpy,https://github.com/pytorch/pytorch/issues/84705,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: numpy,confirm whether existing evidence is enough; otherwise wait for a reproducer,2022-09-13T23:43:48Z
issue,84370,Installation prefix is not passed to CMake appropriately,https://github.com/pytorch/pytorch/issues/84370,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: build,confirm whether existing evidence is enough; otherwise wait for a reproducer,2022-09-01T21:38:51Z
issue,83901,pytorch 1.12.1 Adam Optimizer Malfunction!!!,https://github.com/pytorch/pytorch/issues/83901,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: optimizer,confirm whether existing evidence is enough; otherwise wait for a reproducer,2022-08-23T14:12:51Z
issue,82546,Libtorch C++ torch::stack error,https://github.com/pytorch/pytorch/issues/82546,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: cpp,confirm whether existing evidence is enough; otherwise wait for a reproducer,2022-08-03T14:16:07Z
issue,82217,backward not available for index and mask,https://github.com/pytorch/pytorch/issues/82217,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: sorting and selection",confirm whether existing evidence is enough; otherwise wait for a reproducer,2022-07-27T08:07:01Z
issue,70702,conv3d padding=`same` gradgradcheck fails on CUDA,https://github.com/pytorch/pytorch/issues/70702,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: convolution, module: determinism",confirm whether existing evidence is enough; otherwise wait for a reproducer,2022-07-25T21:16:21Z
issue,81257,High GPU context memory on Torch 1.11.0 but none on Torch 1.10.0,https://github.com/pytorch/pytorch/issues/81257,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda,confirm whether existing evidence is enough; otherwise wait for a reproducer,2022-07-19T23:03:24Z
issue,80302,AttributeError: 'LinearPackedParams' object has no attribute '_modules',https://github.com/pytorch/pytorch/issues/80302,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: nn, oncall: quantization",confirm whether existing evidence is enough; otherwise wait for a reproducer,2022-07-05T15:53:20Z
issue,28950,ConnectionResetError when using dataLoader with pin_memory=True,https://github.com/pytorch/pytorch/issues/28950,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: dataloader,confirm whether existing evidence is enough; otherwise wait for a reproducer,2022-06-23T09:12:28Z
issue,78831,Mac M1 Build Failure on DEBUG=1,https://github.com/pytorch/pytorch/issues/78831,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: arm, module: build, module: macos",confirm whether existing evidence is enough; otherwise wait for a reproducer,2022-06-03T23:52:35Z
issue,76024,The prediction results of different equipment are inconsistent,https://github.com/pytorch/pytorch/issues/76024,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: nn,confirm whether existing evidence is enough; otherwise wait for a reproducer,2022-05-31T16:41:29Z
issue,77893,Segmentation fault in _remove_batch_dim,https://github.com/pytorch/pytorch/issues/77893,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: vmap,confirm whether existing evidence is enough; otherwise wait for a reproducer,2022-05-23T17:11:00Z
issue,76960,"When using Rsqrt, the output of the 1/x process is very likely to have nan/inf",https://github.com/pytorch/pytorch/issues/76960,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label,confirm whether existing evidence is enough; otherwise wait for a reproducer,2022-05-09T04:27:11Z
issue,76959,"When using Lambda, the output of the 1/x process is very likely to have nan/inf",https://github.com/pytorch/pytorch/issues/76959,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label,confirm whether existing evidence is enough; otherwise wait for a reproducer,2022-05-09T04:24:45Z
issue,76956,GlobalAvgPool2d causes the inconsistency of output between frameworks,https://github.com/pytorch/pytorch/issues/76956,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label,confirm whether existing evidence is enough; otherwise wait for a reproducer,2022-05-09T04:23:40Z
issue,76955,GaussianNoise causes the inconsistency of output between frameworks,https://github.com/pytorch/pytorch/issues/76955,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label,confirm whether existing evidence is enough; otherwise wait for a reproducer,2022-05-09T04:22:57Z
issue,76954,ReduceSum causes the inconsistency of output between frameworks,https://github.com/pytorch/pytorch/issues/76954,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label,confirm whether existing evidence is enough; otherwise wait for a reproducer,2022-05-09T04:21:51Z
issue,76643,Pytorch can't process special unicode,https://github.com/pytorch/pytorch/issues/76643,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: serialization,confirm whether existing evidence is enough; otherwise wait for a reproducer,2022-05-08T00:42:14Z
issue,76798,`gradcheck` for `torch.solve` may trigger INTERNAL ASSERT FAILED,https://github.com/pytorch/pytorch/issues/76798,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: autograd,confirm whether existing evidence is enough; otherwise wait for a reproducer,2022-05-05T14:28:45Z
issue,75963,Add build support for GCC 11.2,https://github.com/pytorch/pytorch/issues/75963,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: build,confirm whether existing evidence is enough; otherwise wait for a reproducer,2022-04-25T17:08:41Z
issue,56771,Problems with initial communication between GPUs,https://github.com/pytorch/pytorch/issues/56771,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda,confirm whether existing evidence is enough; otherwise wait for a reproducer,2022-04-18T18:20:17Z
issue,75265,Kernel Dies while using conv2d layer / function,https://github.com/pytorch/pytorch/issues/75265,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label,confirm whether existing evidence is enough; otherwise wait for a reproducer,2022-04-09T15:34:24Z
issue,2482,"""Shared memory manager connection has timed out""",https://github.com/pytorch/pytorch/issues/2482,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: multiprocessing,confirm whether existing evidence is enough; otherwise wait for a reproducer,2022-04-09T07:32:36Z
issue,74670,Can't install Pytorch,https://github.com/pytorch/pytorch/issues/74670,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: build,confirm whether existing evidence is enough; otherwise wait for a reproducer,2022-04-01T17:49:06Z
issue,65027,CUDA out of memory,https://github.com/pytorch/pytorch/issues/65027,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: memory usage",confirm whether existing evidence is enough; otherwise wait for a reproducer,2022-03-30T10:08:13Z
issue,69203,Error in SVD cusolver on Linux,https://github.com/pytorch/pytorch/issues/69203,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: linear algebra,confirm whether existing evidence is enough; otherwise wait for a reproducer,2022-03-24T23:32:33Z
issue,35754,pytorch 1.4.0 hangs when using with CUDA >= 10.1,https://github.com/pytorch/pytorch/issues/35754,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda,confirm whether existing evidence is enough; otherwise wait for a reproducer,2022-02-22T11:56:24Z
issue,72105,ImportError: cannot import name '_VF' from partially initialized module 'torch',https://github.com/pytorch/pytorch/issues/72105,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label,confirm whether existing evidence is enough; otherwise wait for a reproducer,2022-02-21T12:53:57Z
issue,72591,"Can't forward pass conv2d with kernel_size=1, and padding=1",https://github.com/pytorch/pytorch/issues/72591,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: arm, module: build, module: convolution, module: macos",confirm whether existing evidence is enough; otherwise wait for a reproducer,2022-02-14T17:48:23Z
issue,69966,torch.nn.DataParallel caused inference failure with cpu set as device on NV machine,https://github.com/pytorch/pytorch/issues/69966,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: data parallel,confirm whether existing evidence is enough; otherwise wait for a reproducer,2022-02-13T22:02:39Z
issue,71704,M1 Pro Apple Silicon chip support.,https://github.com/pytorch/pytorch/issues/71704,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: arm, module: macos, module: nnpack",confirm whether existing evidence is enough; otherwise wait for a reproducer,2022-01-25T03:11:08Z
issue,71272,"UserWarning: Seems like `optimizer.step()` has been overridden after learning rate scheduler initialization. Please, make sure to call `optimizer.step()` before `lr_scheduler.step()`. See more details at https://pytorch.org/docs/stable/optim.html#how-to-adjust-learning-rate warnings.warn(""Seems like `optimizer.step()` has been overridden after learning rate scheduler",https://github.com/pytorch/pytorch/issues/71272,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: lrscheduler, module: optimizer",confirm whether existing evidence is enough; otherwise wait for a reproducer,2022-01-20T16:33:17Z
issue,70038,comile error,https://github.com/pytorch/pytorch/issues/70038,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: build, module: mkldnn, module: third_party",confirm whether existing evidence is enough; otherwise wait for a reproducer,2022-01-08T00:34:33Z
issue,70519,RuntimeError: cublas runtime error,https://github.com/pytorch/pytorch/issues/70519,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: cublas,confirm whether existing evidence is enough; otherwise wait for a reproducer,2022-01-06T13:57:42Z
issue,60341,C++ version 1.9.0 libtorch dynamic load fails -- GCC only,https://github.com/pytorch/pytorch/issues/60341,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: abi, module: binaries, module: cpp",confirm whether existing evidence is enough; otherwise wait for a reproducer,2021-12-21T14:50:49Z
issue,24398,Significantly slower in latest version than in 0.4.0,https://github.com/pytorch/pytorch/issues/24398,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: performance",confirm whether existing evidence is enough; otherwise wait for a reproducer,2021-12-20T00:54:32Z
issue,70008,Torch function runtime seemingly dependent on scipy call,https://github.com/pytorch/pytorch/issues/70008,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: linear algebra, module: performance",confirm whether existing evidence is enough; otherwise wait for a reproducer,2021-12-17T15:04:44Z
issue,66781,lr_scheduler.py / list index out of range,https://github.com/pytorch/pytorch/issues/66781,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: lrscheduler, module: optimizer",confirm whether existing evidence is enough; otherwise wait for a reproducer,2021-12-13T15:25:36Z
issue,69448,RuntimeError: PyTorch convert function for op 'inverse' not implemented.,https://github.com/pytorch/pytorch/issues/69448,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label,confirm whether existing evidence is enough; otherwise wait for a reproducer,2021-12-08T21:33:10Z
issue,68864,Incorrect accuracy pytorch(1.0.1) and libtorch because of nn.LSTM,https://github.com/pytorch/pytorch/issues/68864,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label,confirm whether existing evidence is enough; otherwise wait for a reproducer,2021-11-30T01:37:22Z
issue,68144,Unstable buggy calculation.,https://github.com/pytorch/pytorch/issues/68144,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: optimizer,confirm whether existing evidence is enough; otherwise wait for a reproducer,2021-11-25T00:53:03Z
issue,67740,`torch.utils.data.random_split` example broken in 1.10,https://github.com/pytorch/pytorch/issues/67740,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: data,confirm whether existing evidence is enough; otherwise wait for a reproducer,2021-11-10T19:01:37Z
issue,67835,It seems to be a bug！！！,https://github.com/pytorch/pytorch/issues/67835,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label,confirm whether existing evidence is enough; otherwise wait for a reproducer,2021-11-05T20:16:28Z
issue,63726,a problem happened in torch.randperm,https://github.com/pytorch/pytorch/issues/63726,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: random,confirm whether existing evidence is enough; otherwise wait for a reproducer,2021-10-09T08:49:35Z
issue,64889,"""RuntimeError: CUDA error: out of memory"" for no reason",https://github.com/pytorch/pytorch/issues/64889,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: memory usage",confirm whether existing evidence is enough; otherwise wait for a reproducer,2021-09-13T16:12:07Z
issue,64560,Private bytes consumption exception,https://github.com/pytorch/pytorch/issues/64560,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: memory usage",confirm whether existing evidence is enough; otherwise wait for a reproducer,2021-09-10T22:00:33Z
issue,64804,RuntimeError: one of the variables needed for gradient computation has been modified by an inplace operation,https://github.com/pytorch/pytorch/issues/64804,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: nn,confirm whether existing evidence is enough; otherwise wait for a reproducer,2021-09-10T12:38:42Z
issue,64143,AttributeError: module 'torchvision.models' has no attribute 'load_model',https://github.com/pytorch/pytorch/issues/64143,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: serialization,confirm whether existing evidence is enough; otherwise wait for a reproducer,2021-09-05T03:23:38Z
issue,64156,Get zero when using torch.matmul and torch.dot with 1-D tensor in torch-1.9.0-cp39-none-macosx_11_0_arm64.whl,https://github.com/pytorch/pytorch/issues/64156,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: arm, module: correctness (silent)",confirm whether existing evidence is enough; otherwise wait for a reproducer,2021-08-31T17:32:56Z
issue,61341,inerror: reference to ‘DeviceType’ is ambiguous,https://github.com/pytorch/pytorch/issues/61341,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: build, module: cpp",confirm whether existing evidence is enough; otherwise wait for a reproducer,2021-08-22T15:48:23Z
issue,63322,lstm's input h0 and c0 bug,https://github.com/pytorch/pytorch/issues/63322,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: nans and infs, module: numpy, module: rnn",confirm whether existing evidence is enough; otherwise wait for a reproducer,2021-08-17T18:05:54Z
issue,62644,waste too much time in first two pictures,https://github.com/pytorch/pytorch/issues/62644,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label,confirm whether existing evidence is enough; otherwise wait for a reproducer,2021-08-11T03:13:14Z
issue,59628,what(): result type Float can't be cast to the desired output type Long,https://github.com/pytorch/pytorch/issues/59628,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: cpp,confirm whether existing evidence is enough; otherwise wait for a reproducer,2021-08-07T03:17:55Z
issue,61660,Pytorch model load failure in Gunicorn with Gevent workers,https://github.com/pytorch/pytorch/issues/61660,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: dataloader, module: nn, module: numpy",confirm whether existing evidence is enough; otherwise wait for a reproducer,2021-08-03T22:21:05Z
issue,62152,Runtime Error: Expected a 'cuda' device type for generator but found 'cpu' in google colab,https://github.com/pytorch/pytorch/issues/62152,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label,confirm whether existing evidence is enough; otherwise wait for a reproducer,2021-07-26T10:02:42Z
issue,61243,Pytorch 1.5+ is slower than pytorch 1.3,https://github.com/pytorch/pytorch/issues/61243,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: performance,confirm whether existing evidence is enough; otherwise wait for a reproducer,2021-07-26T08:35:24Z
issue,59626,multi scripted model on the same gpu on multi gpu machine,https://github.com/pytorch/pytorch/issues/59626,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: multi-gpu,confirm whether existing evidence is enough; otherwise wait for a reproducer,2021-07-09T09:31:39Z
issue,60936,The usage of get_stoi() in 0.10.0 and stoi[] in previous versions,https://github.com/pytorch/pytorch/issues/60936,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label,confirm whether existing evidence is enough; otherwise wait for a reproducer,2021-06-30T00:03:22Z
issue,60923,RuntimeError: derivative for im2col_backward is not implemented,https://github.com/pytorch/pytorch/issues/60923,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: nn",confirm whether existing evidence is enough; otherwise wait for a reproducer,2021-06-29T06:13:34Z
issue,60094,bundled libiomp5 causing segfaults in other libraries that use libomp,https://github.com/pytorch/pytorch/issues/60094,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: macos, module: openmp",confirm whether existing evidence is enough; otherwise wait for a reproducer,2021-06-23T16:20:31Z
issue,56928,torch-1.8.1 wheel verification fails with distlib,https://github.com/pytorch/pytorch/issues/56928,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: binaries,confirm whether existing evidence is enough; otherwise wait for a reproducer,2021-06-22T13:40:05Z
issue,59883,"matmul causing cuDNN error: CUDNN_STATUS_INTERNAL_ERROR at future, unrelated Conv2d",https://github.com/pytorch/pytorch/issues/59883,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: cudnn",confirm whether existing evidence is enough; otherwise wait for a reproducer,2021-06-14T20:56:27Z
issue,36649,"RuntimeError: Expected cuda::check_device({sparse_, r_, t, dense}) to be true, but got false.",https://github.com/pytorch/pytorch/issues/36649,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: data parallel,confirm whether existing evidence is enough; otherwise wait for a reproducer,2021-06-07T22:36:29Z
issue,54975,CUBLAS_STATUS_EXECUTION_FAILED error on torch >= 1.8.0 and CUDA 11.1,https://github.com/pytorch/pytorch/issues/54975,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda,confirm whether existing evidence is enough; otherwise wait for a reproducer,2021-05-19T04:38:03Z
issue,58135,Segmentation fault need help,https://github.com/pytorch/pytorch/issues/58135,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda,confirm whether existing evidence is enough; otherwise wait for a reproducer,2021-05-13T23:31:32Z
issue,57794,CUDA error: an illegal memory access was encountered,https://github.com/pytorch/pytorch/issues/57794,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda,confirm whether existing evidence is enough; otherwise wait for a reproducer,2021-05-08T02:24:48Z
issue,57796,BatchNorm grad calculation is imprecise,https://github.com/pytorch/pytorch/issues/57796,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: numerical-stability",confirm whether existing evidence is enough; otherwise wait for a reproducer,2021-05-07T22:13:28Z
issue,57491,setting all seed still get different result.,https://github.com/pytorch/pytorch/issues/57491,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: determinism, module: random",confirm whether existing evidence is enough; otherwise wait for a reproducer,2021-05-05T23:40:05Z
issue,54799,"Vertices=torch.matmul(vertices.unsqueeze(0), rotations_init), RuntimeError: CUDA error: CUBLAS_STATUS_EXECUTION_FAILED when calling `cublasSgemmStridedBatched in CentOS",https://github.com/pytorch/pytorch/issues/54799,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: numpy",confirm whether existing evidence is enough; otherwise wait for a reproducer,2021-04-22T18:22:05Z
issue,49832,RuntimeError: CUDA error: unspecified launch failure,https://github.com/pytorch/pytorch/issues/49832,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda,confirm whether existing evidence is enough; otherwise wait for a reproducer,2021-04-21T15:50:14Z
issue,56019,when building apk with libtorch,https://github.com/pytorch/pytorch/issues/56019,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: build,confirm whether existing evidence is enough; otherwise wait for a reproducer,2021-04-15T10:35:42Z
issue,12641,Install Jetson TX2 Max Regcount Error,https://github.com/pytorch/pytorch/issues/12641,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: build, module: jetson",confirm whether existing evidence is enough; otherwise wait for a reproducer,2021-04-15T10:32:43Z
issue,55035,DataLoader performance drop on 4-channel images?,https://github.com/pytorch/pytorch/issues/55035,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: dataloader,confirm whether existing evidence is enough; otherwise wait for a reproducer,2021-04-07T14:31:57Z
issue,54716,Segmentation fault in PyTorch dataloader,https://github.com/pytorch/pytorch/issues/54716,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: dataloader,confirm whether existing evidence is enough; otherwise wait for a reproducer,2021-04-01T18:14:05Z
issue,54827,Import error,https://github.com/pytorch/pytorch/issues/54827,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label,confirm whether existing evidence is enough; otherwise wait for a reproducer,2021-03-30T16:58:01Z
issue,54487,"RuntimeError: iter.device(arg).is_cuda() INTERNAL ASSERT FAILED at ""/opt/conda/conda-bld/pytorch_1607370116979/work/aten/src/ATen/native/cuda/Loops.cuh"":94, please report a bug to PyTorch.",https://github.com/pytorch/pytorch/issues/54487,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: tensoriterator",confirm whether existing evidence is enough; otherwise wait for a reproducer,2021-03-26T02:01:58Z
issue,54577,Question when I use Pytorch Mobile,https://github.com/pytorch/pytorch/issues/54577,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: oncall: mobile,confirm whether existing evidence is enough; otherwise wait for a reproducer,2021-03-24T20:33:38Z
issue,24397,SyncBatchNorm error when using model.eval() with DistributedDataParallel,https://github.com/pytorch/pytorch/issues/24397,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: autograd, oncall: distributed",confirm whether existing evidence is enough; otherwise wait for a reproducer,2021-03-20T11:38:07Z
issue,53464,libtorch reports bad_alloc when used together with darknet,https://github.com/pytorch/pytorch/issues/53464,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label,confirm whether existing evidence is enough; otherwise wait for a reproducer,2021-03-15T17:17:58Z
issue,52989,lr_scheduler _triangular2_scale_fn calculation is overflowed,https://github.com/pytorch/pytorch/issues/52989,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: optimizer,confirm whether existing evidence is enough; otherwise wait for a reproducer,2021-03-02T14:23:05Z
issue,10950,Cannot run torch in different sub-interpreters,https://github.com/pytorch/pytorch/issues/10950,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: cpp,confirm whether existing evidence is enough; otherwise wait for a reproducer,2021-03-01T21:41:32Z
issue,52515,torch.bmm incorrect with pytorch 1.7.1 and cuda 11,https://github.com/pytorch/pytorch/issues/52515,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: cublas, module: cuda",confirm whether existing evidence is enough; otherwise wait for a reproducer,2021-02-22T23:15:36Z
issue,52307,Model produces same outputs for different inputs when using GPU,https://github.com/pytorch/pytorch/issues/52307,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda,confirm whether existing evidence is enough; otherwise wait for a reproducer,2021-02-20T20:54:05Z
issue,51275,Error of running,https://github.com/pytorch/pytorch/issues/51275,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label,confirm whether existing evidence is enough; otherwise wait for a reproducer,2021-01-29T07:56:16Z
issue,40133,Why does Libtorch use more memory than Pytorch does in Python？,https://github.com/pytorch/pytorch/issues/40133,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: memory usage,confirm whether existing evidence is enough; otherwise wait for a reproducer,2021-01-21T12:38:13Z
issue,46806,Compile Pytorch with MAGMA Issue,https://github.com/pytorch/pytorch/issues/46806,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: build,confirm whether existing evidence is enough; otherwise wait for a reproducer,2021-01-20T15:14:54Z
issue,50261,ModuleNotFoundError: No module named '__torch__',https://github.com/pytorch/pytorch/issues/50261,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: multiprocessing, module: serialization",confirm whether existing evidence is enough; otherwise wait for a reproducer,2021-01-08T18:13:08Z
issue,49669,"While training the model, GPU util reaches 100% but no progress happends.",https://github.com/pytorch/pytorch/issues/49669,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: deadlock",confirm whether existing evidence is enough; otherwise wait for a reproducer,2020-12-30T03:11:35Z
issue,49870,Tensorboard Error,https://github.com/pytorch/pytorch/issues/49870,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label,confirm whether existing evidence is enough; otherwise wait for a reproducer,2020-12-29T20:31:57Z
issue,48579,test_fs_pool fails,https://github.com/pytorch/pytorch/issues/48579,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: multiprocessing, module: tests",confirm whether existing evidence is enough; otherwise wait for a reproducer,2020-12-21T09:50:23Z
issue,47832,Error during distributed training,https://github.com/pytorch/pytorch/issues/47832,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,confirm whether existing evidence is enough; otherwise wait for a reproducer,2020-12-04T08:37:37Z
issue,32584,Segmentation Fault (core dumped) with 1.4.0,https://github.com/pytorch/pytorch/issues/32584,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label,confirm whether existing evidence is enough; otherwise wait for a reproducer,2020-10-27T20:19:39Z
issue,46153,Getting this error while installing torch and torchvision!,https://github.com/pytorch/pytorch/issues/46153,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label,confirm whether existing evidence is enough; otherwise wait for a reproducer,2020-10-12T21:24:32Z
issue,28943,The `@` operation uses too much memory on GPU,https://github.com/pytorch/pytorch/issues/28943,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: memory usage,confirm whether existing evidence is enough; otherwise wait for a reproducer,2020-10-09T21:32:55Z
issue,44855,RuntimeError: Caught RuntimeError in replica 1 on device 1.,https://github.com/pytorch/pytorch/issues/44855,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: data parallel,confirm whether existing evidence is enough; otherwise wait for a reproducer,2020-09-17T03:49:10Z
issue,42581,Build script complains that gcc is < 6 even though gcc version is 8,https://github.com/pytorch/pytorch/issues/42581,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label,confirm whether existing evidence is enough; otherwise wait for a reproducer,2020-09-14T00:41:06Z
issue,44024,"When the pytorch training data reaches a certain epoch, the memory remains unchanged?",https://github.com/pytorch/pytorch/issues/44024,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label,confirm whether existing evidence is enough; otherwise wait for a reproducer,2020-09-05T03:01:25Z
issue,43859,Multi-machine multi-gpu training won't start on CNGrid,https://github.com/pytorch/pytorch/issues/43859,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,confirm whether existing evidence is enough; otherwise wait for a reproducer,2020-09-03T04:57:55Z
issue,35870,CUDNN_STATUS_INTERNAL_ERROR with GPU RTX 8000,https://github.com/pytorch/pytorch/issues/35870,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: cudnn",confirm whether existing evidence is enough; otherwise wait for a reproducer,2020-06-25T23:50:02Z
issue,4334,torch.cuda.device_count() returns 1 using 4 TitanX setup.,https://github.com/pytorch/pytorch/issues/4334,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda,confirm whether existing evidence is enough; otherwise wait for a reproducer,2020-06-25T15:31:14Z
issue,39832,Torch Distributed Asynch docs with invalid ENV args,https://github.com/pytorch/pytorch/issues/39832,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: docs, module: rpc",confirm whether existing evidence is enough; otherwise wait for a reproducer,2020-06-15T18:09:13Z
issue,3863,Considerable slowdown in Adam.step after a number of epochs with multiple losses,https://github.com/pytorch/pytorch/issues/3863,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: optimizer, module: performance",confirm whether existing evidence is enough; otherwise wait for a reproducer,2020-05-26T16:19:25Z
issue,26400,RuntimeError: tensor.ndimension() == static_cast<int64_t>(expected_size.size()) INTERNAL ASSERT FAILED,https://github.com/pytorch/pytorch/issues/26400,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: multi-gpu",confirm whether existing evidence is enough; otherwise wait for a reproducer,2020-05-15T05:47:59Z
issue,38155,"Python script name with ""profile.py"" will run twice",https://github.com/pytorch/pytorch/issues/38155,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label,confirm whether existing evidence is enough; otherwise wait for a reproducer,2020-05-13T00:39:15Z
issue,36721,[1.4.1] Cuda build fails,https://github.com/pytorch/pytorch/issues/36721,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: build,confirm whether existing evidence is enough; otherwise wait for a reproducer,2020-04-24T10:19:18Z
issue,37057,CUDA error: device-side assert triggered @ model.cuda(),https://github.com/pytorch/pytorch/issues/37057,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda,confirm whether existing evidence is enough; otherwise wait for a reproducer,2020-04-22T18:23:23Z
issue,37066,Saved model behaves differently on same data,https://github.com/pytorch/pytorch/issues/37066,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: serialization,confirm whether existing evidence is enough; otherwise wait for a reproducer,2020-04-22T18:07:42Z
issue,37067,can't wrap two models in the same class,https://github.com/pytorch/pytorch/issues/37067,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label,confirm whether existing evidence is enough; otherwise wait for a reproducer,2020-04-22T18:04:17Z
issue,35990,"[JIT] Tensor method API behavior discrepancy, Tensor.detach(..)",https://github.com/pytorch/pytorch/issues/35990,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,confirm whether existing evidence is enough; otherwise wait for a reproducer,2020-04-07T18:09:53Z
issue,34937,[JIT]torch.jit.export invaild for pytorch1.4,https://github.com/pytorch/pytorch/issues/34937,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,confirm whether existing evidence is enough; otherwise wait for a reproducer,2020-03-20T18:31:57Z
issue,25904,Python hang after using torch.exp(),https://github.com/pytorch/pytorch/issues/25904,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: deadlock,confirm whether existing evidence is enough; otherwise wait for a reproducer,2020-03-10T14:53:52Z
issue,30900,CUDA error: initialization error (multiprocessing) with Python 3.7,https://github.com/pytorch/pytorch/issues/30900,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: multiprocessing,confirm whether existing evidence is enough; otherwise wait for a reproducer,2020-02-12T19:43:37Z
issue,22597,Autograd profiler memory leak when use_cuda=True,https://github.com/pytorch/pytorch/issues/22597,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: autograd,confirm whether existing evidence is enough; otherwise wait for a reproducer,2020-02-07T22:52:41Z
issue,30830,Torch getting stuck transfering model to GPU in multiple GPU setting,https://github.com/pytorch/pytorch/issues/30830,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: multiprocessing",confirm whether existing evidence is enough; otherwise wait for a reproducer,2020-02-04T07:48:07Z
issue,32662,Second-order gradient cause segfault over time,https://github.com/pytorch/pytorch/issues/32662,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label,confirm whether existing evidence is enough; otherwise wait for a reproducer,2020-01-30T18:07:59Z
issue,32443,Segmentation fault in lazyInitCUDA -> CUDAHooks::initCUDA -> THCMagma_init -> magma_init,https://github.com/pytorch/pytorch/issues/32443,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: third_party",confirm whether existing evidence is enough; otherwise wait for a reproducer,2020-01-21T22:52:20Z
issue,31660,Cuda error 59 : device-side assert triggered,https://github.com/pytorch/pytorch/issues/31660,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda,confirm whether existing evidence is enough; otherwise wait for a reproducer,2020-01-02T15:00:07Z
issue,31004,The inference speed of the torch compiled manually is slower than the torch build from official binaries?,https://github.com/pytorch/pytorch/issues/31004,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: performance",confirm whether existing evidence is enough; otherwise wait for a reproducer,2019-12-10T01:24:56Z
issue,30955,error C3203: “templated_iterator”: 未专用化的 类 模板 不能用作 模板 变量，该变量属于 模板 参数“_Ty1”，应为 real 类型,https://github.com/pytorch/pytorch/issues/30955,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: internals,confirm whether existing evidence is enough; otherwise wait for a reproducer,2019-12-09T18:28:21Z
issue,29891,Get wrong precision when multi nodes run in docker,https://github.com/pytorch/pytorch/issues/29891,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,confirm whether existing evidence is enough; otherwise wait for a reproducer,2019-11-21T20:31:25Z
issue,28956,RuntimeError: CUDA error: unknown error,https://github.com/pytorch/pytorch/issues/28956,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda,confirm whether existing evidence is enough; otherwise wait for a reproducer,2019-10-31T13:31:21Z
issue,14352,[libtorch] Catkin_make compilation error,https://github.com/pytorch/pytorch/issues/14352,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: build,confirm whether existing evidence is enough; otherwise wait for a reproducer,2019-10-27T12:33:01Z
issue,27694,Deployment training model at C + + end,https://github.com/pytorch/pytorch/issues/27694,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: docs, oncall: jit",confirm whether existing evidence is enough; otherwise wait for a reproducer,2019-10-22T02:14:58Z
issue,22924,[data loader] Graceful data loader threads exit on KeyboardInterrupt,https://github.com/pytorch/pytorch/issues/22924,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: dataloader,confirm whether existing evidence is enough; otherwise wait for a reproducer,2019-10-01T17:27:35Z
issue,16654,pytorch was blocked at loss.backward,https://github.com/pytorch/pytorch/issues/16654,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,confirm whether existing evidence is enough; otherwise wait for a reproducer,2019-09-17T13:54:47Z
issue,20134,"ERROR: Command ""python setup.py egg_info"" when dockerfile build",https://github.com/pytorch/pytorch/issues/20134,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: build,confirm whether existing evidence is enough; otherwise wait for a reproducer,2019-06-26T13:52:53Z
issue,20805,LibTorch :About torch.jit.trace generate model.pt,https://github.com/pytorch/pytorch/issues/20805,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,confirm whether existing evidence is enough; otherwise wait for a reproducer,2019-06-21T03:35:46Z
issue,21179,Failed to install pytorch from source on ubuntu.,https://github.com/pytorch/pytorch/issues/21179,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: build,confirm whether existing evidence is enough; otherwise wait for a reproducer,2019-06-19T00:08:08Z
issue,19128,pytorch.version.cuda is None when compiling with CUDA support,https://github.com/pytorch/pytorch/issues/19128,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: build, module: cuda",confirm whether existing evidence is enough; otherwise wait for a reproducer,2019-05-28T09:47:17Z
issue,19029,C++ custom module not thread safe,https://github.com/pytorch/pytorch/issues/19029,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: cpp,confirm whether existing evidence is enough; otherwise wait for a reproducer,2019-04-11T08:06:40Z
pr,189297,[CI] Pin quack to torch's vendored SHA to fix B200 smoke cutlass import crash,https://github.com/pytorch/pytorch/pull/189297,draft/noise,13,"PR is draft; CI lane labels: ciflow/b200, ciflow/trunk",wait for author to mark ready,2026-07-09T00:20:52Z
pr,188739,flex_gemm: add quant lowerings,https://github.com/pytorch/pytorch/pull/188739,draft/noise,13,"PR is draft; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for author to mark ready,2026-07-08T23:12:51Z
pr,188112,flex_gemm: feed local reductions into epilogues,https://github.com/pytorch/pytorch/pull/188112,draft/noise,13,"PR is draft; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for author to mark ready,2026-07-08T23:12:51Z
pr,188470,flex_gemm: extend local reduce aux groups,https://github.com/pytorch/pytorch/pull/188470,draft/noise,13,"PR is draft; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for author to mark ready,2026-07-08T23:12:50Z
pr,189290,Harden Inductor SubprocPool against compile-worker death and teardown hangs,https://github.com/pytorch/pytorch/pull/189290,draft/noise,13,"PR is draft; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for author to mark ready,2026-07-08T21:49:50Z
pr,189230,Support multiple pools in cudagraph_trees,https://github.com/pytorch/pytorch/pull/189230,draft/noise,13,"PR is draft; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for author to mark ready,2026-07-08T21:35:35Z
pr,189304,Use freeze_rng_state in distribution rsample gradient tests,https://github.com/pytorch/pytorch/pull/189304,draft/noise,13,PR is draft,wait for author to mark ready,2026-07-08T21:31:26Z
pr,185424,Add CPython Dynamo agentic loop plan,https://github.com/pytorch/pytorch/pull/185424,draft/noise,13,PR is draft,wait for author to mark ready,2026-07-08T20:53:26Z
pr,189258,[inductor] Replace ExternKernel/ComputedBuffer plumbing tuples with dataclasses + Protocol,https://github.com/pytorch/pytorch/pull/189258,draft/noise,13,"PR is draft; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk",wait for author to mark ready,2026-07-08T19:12:16Z
pr,186302,Install Rust toolchain in PyTorch CI/CD build images (#186302),https://github.com/pytorch/pytorch/pull/186302,draft/noise,13,"PR is draft; CI lane labels: ciflow/docker, ciflow/inductor",wait for author to mark ready,2026-07-08T17:49:26Z
pr,188682,Limit ROCm build targets to architectures covered by testing,https://github.com/pytorch/pytorch/pull/188682,draft/noise,13,"PR is draft; CI lane labels: ciflow/rocm-mi300, ciflow/rocm-mi350, ciflow/trunk",wait for author to mark ready,2026-07-08T17:25:27Z
pr,187859,[functorch] Add torch._functorch.aot_autograd.compile_to_python,https://github.com/pytorch/pytorch/pull/187859,draft/noise,13,PR is draft; CI lane labels: ciflow/inductor,wait for author to mark ready,2026-07-08T16:57:39Z
pr,189253,[TEST] Refactor test_c10d_spawn_ucc.py with hw_classification,https://github.com/pytorch/pytorch/pull/189253,draft/noise,13,PR is draft,wait for author to mark ready,2026-07-08T12:01:34Z
pr,188653,Add test decorator for device-type-specific RNG seed setting.,https://github.com/pytorch/pytorch/pull/188653,draft/noise,13,PR is draft,wait for author to mark ready,2026-07-08T10:51:18Z
pr,188388,[WIP][XPU][Test]Port 12 UT test suites to Intel GPU,https://github.com/pytorch/pytorch/pull/188388,draft/noise,13,PR is draft; CI lane labels: ciflow/xpu,wait for author to mark ready,2026-07-08T08:28:23Z
pr,183041,[FSDP1] Test forward overlap with use_orig_params,https://github.com/pytorch/pytorch/pull/183041,draft/noise,13,PR is draft,wait for author to mark ready,2026-07-08T08:18:44Z
pr,189238,[XPU] Register linalg._powsum and _foreach_powsum.Scalar for XPU,https://github.com/pytorch/pytorch/pull/189238,draft/noise,13,PR is draft,wait for author to mark ready,2026-07-08T06:31:23Z
pr,185816,[WIP] [PoC] Enable CPU FlexAttention on AArch64,https://github.com/pytorch/pytorch/pull/185816,draft/noise,13,"PR is draft; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk",wait for author to mark ready,2026-07-07T17:14:11Z
pr,189097,[RFC] Expose Kineto custom logger registration,https://github.com/pytorch/pytorch/pull/189097,draft/noise,13,PR is draft,wait for author to mark ready,2026-07-07T13:56:30Z
pr,188195,Adding OpenReg allocator stats tests,https://github.com/pytorch/pytorch/pull/188195,draft/noise,13,PR is draft,wait for author to mark ready,2026-07-07T11:52:44Z
pr,187318,[xpu]Implement scaled_mm for XPU - combined,https://github.com/pytorch/pytorch/pull/187318,draft/noise,13,"PR is draft; CI lane labels: ciflow/b200, ciflow/h100, ciflow/inductor, ciflow/rocm-mi300",wait for author to mark ready,2026-07-07T08:21:54Z
pr,189020,Reorder condition to handle empty tensor the same for file_system & file_descriptor,https://github.com/pytorch/pytorch/pull/189020,draft/noise,13,PR is draft; CI lane labels: ciflow/trunk,wait for author to mark ready,2026-07-07T07:04:43Z
pr,182379,"[DTensor] Custom algorithm dispatch mechanism for DTensor, one-sided GEMM algorithm",https://github.com/pytorch/pytorch/pull/182379,draft/noise,13,"PR is draft; CI lane labels: ciflow/dtensor, ciflow/h100-symm-mem, ciflow/inductor, ciflow/torchtitan",wait for author to mark ready,2026-07-07T01:08:45Z
pr,189099,[Profiler][NOT FOR LAND] CI test debugging,https://github.com/pytorch/pytorch/pull/189099,draft/noise,13,PR is draft,wait for author to mark ready,2026-07-06T23:37:02Z
pr,182841,[CI] No-op docker change to test current build state,https://github.com/pytorch/pytorch/pull/182841,draft/noise,13,PR is draft,wait for author to mark ready,2026-07-06T22:51:36Z
pr,182838,"[CI] Fix docker builds: pin miniforge, use --extra-index-url",https://github.com/pytorch/pytorch/pull/182838,draft/noise,13,PR is draft,wait for author to mark ready,2026-07-06T22:51:34Z
pr,189081,Add tp_name property to Variable Tracker,https://github.com/pytorch/pytorch/pull/189081,draft/noise,13,"PR is draft; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for author to mark ready,2026-07-06T20:16:23Z
pr,181659,[FSDP2] Add recurse= to set_symm_mem_for_comm and set_force_sum_reduction_for_comms,https://github.com/pytorch/pytorch/pull/181659,draft/noise,13,"PR is draft; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for author to mark ready,2026-07-06T08:02:22Z
pr,188918,[dynamo] Model list subclass __new__/__init__ arg handling to match CPython,https://github.com/pytorch/pytorch/pull/188918,draft/noise,13,"PR is draft; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for author to mark ready,2026-07-03T21:13:49Z
pr,187540,"[DRAFT] Fix forward AD inconsistency for pow(abs(complex), complex)",https://github.com/pytorch/pytorch/pull/187540,draft/noise,13,PR is draft,wait for author to mark ready,2026-07-03T16:56:53Z
pr,188888,Add Reusable Event support to XPUEvent,https://github.com/pytorch/pytorch/pull/188888,draft/noise,13,PR is draft,wait for author to mark ready,2026-07-03T11:07:34Z
pr,188782,[CI] Benchmark TorchInductor perf against 2.13.0 RC,https://github.com/pytorch/pytorch/pull/188782,draft/noise,13,PR is draft,wait for author to mark ready,2026-07-02T22:50:16Z
pr,188781,[CI] Benchmark TorchInductor perf against 2.12.1 release (baseline),https://github.com/pytorch/pytorch/pull/188781,draft/noise,13,PR is draft,wait for author to mark ready,2026-07-02T10:36:31Z
pr,186282,Fix DTensor view-unflatten when one output dim is sharded by 2+ mesh axes,https://github.com/pytorch/pytorch/pull/186282,draft/noise,13,"PR is draft; CI lane labels: ciflow/dtensor, ciflow/inductor, ciflow/torchtitan",wait for author to mark ready,2026-07-02T04:49:37Z
pr,188593,"[ROCm] Add skipIfRocmVersionAtLeast([7, 14]) skips for ROCm 7.14 known failures",https://github.com/pytorch/pytorch/pull/188593,draft/noise,13,"PR is draft; CI lane labels: ciflow/dtensor, ciflow/inductor",wait for author to mark ready,2026-07-01T01:23:01Z
pr,188565,[ROCm] Add initial GFX1250 architecture support,https://github.com/pytorch/pytorch/pull/188565,draft/noise,13,"PR is draft; CI lane labels: ciflow/b200, ciflow/docker, ciflow/h100, ciflow/h100-symm-mem",wait for author to mark ready,2026-06-30T20:07:06Z
pr,186980,Move header-only c10/core headers to torch/headeronly,https://github.com/pytorch/pytorch/pull/186980,draft/noise,13,PR is draft; CI lane labels: ciflow/inductor,wait for author to mark ready,2026-06-30T16:04:35Z
pr,178305,[XPU][oneDNN] Enforce const/mutable pointer semantics for mkldnn xpu detail,https://github.com/pytorch/pytorch/pull/178305,draft/noise,13,PR is draft,wait for author to mark ready,2026-06-30T12:13:40Z
pr,188447,Add DTensor shard propagation skill,https://github.com/pytorch/pytorch/pull/188447,draft/noise,13,PR is draft,wait for author to mark ready,2026-06-29T19:31:11Z
pr,181145,[CuteDSL][Native Ops] group norm kernel,https://github.com/pytorch/pytorch/pull/181145,draft/noise,13,PR is draft,wait for author to mark ready,2026-06-28T18:52:57Z
pr,183688,[MPS] Add torch.mps.MPSGraph capture/replay via MTLIndirectCommandBuffer,https://github.com/pytorch/pytorch/pull/183688,draft/noise,13,PR is draft; CI lane labels: ciflow/mps,wait for author to mark ready,2026-06-27T22:52:51Z
pr,182749,[MPS] Zero-copy CPU->MPS via memcpy on Apple Silicon UMA,https://github.com/pytorch/pytorch/pull/182749,draft/noise,13,PR is draft; CI lane labels: ciflow/mps,wait for author to mark ready,2026-06-27T22:52:51Z
pr,182736,[MPS] Zero-copy MPS->CPU transfer via unified memory in copy_from_mps_,https://github.com/pytorch/pytorch/pull/182736,draft/noise,13,PR is draft; CI lane labels: ciflow/mps,wait for author to mark ready,2026-06-27T22:52:51Z
pr,182731,[MPS] Zero-copy scalar extraction via unified memory in _local_scalar_dense,https://github.com/pytorch/pytorch/pull/182731,draft/noise,13,PR is draft; CI lane labels: ciflow/mps,wait for author to mark ready,2026-06-27T22:52:51Z
pr,182791,[MPS] UMA fast path for fill/zero via CPU memset/std::fill,https://github.com/pytorch/pytorch/pull/182791,draft/noise,13,PR is draft; CI lane labels: ciflow/mps,wait for author to mark ready,2026-06-27T22:52:50Z
pr,180573,Avoid unnecessary refcount bump in inplace op Python bindings,https://github.com/pytorch/pytorch/pull/180573,draft/noise,13,PR is draft,wait for author to mark ready,2026-06-26T22:01:48Z
pr,188257,[RFC] S3-over-RDMA checkpoint storage backends via NVIDIA cuObject,https://github.com/pytorch/pytorch/pull/188257,draft/noise,13,PR is draft; CI lane labels: ciflow/torchtitan,wait for author to mark ready,2026-06-26T18:17:07Z
pr,181528,sdp: warn when enable_gqa=True silently falls back to MATH,https://github.com/pytorch/pytorch/pull/181528,draft/noise,13,PR is draft,wait for author to mark ready,2026-06-25T22:59:58Z
pr,188120,[FSDP2] Per-FQN offload policy map and cpu_offload_by_budget helper,https://github.com/pytorch/pytorch/pull/188120,draft/noise,13,"PR is draft; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for author to mark ready,2026-06-25T00:31:33Z
pr,188065,[AOTI] measure performance with AOTI_RUNTIME_CHECK_INPUTS=1,https://github.com/pytorch/pytorch/pull/188065,draft/noise,13,PR is draft,wait for author to mark ready,2026-06-24T19:03:06Z
pr,181557,[XPU] Enable grouped_mm and scaled_grouped_mm with OneDNN,https://github.com/pytorch/pytorch/pull/181557,draft/noise,13,"PR is draft; CI lane labels: ciflow/b200, ciflow/h100, ciflow/inductor, ciflow/rocm-mi300",wait for author to mark ready,2026-06-22T15:09:20Z
pr,187835,[profiler] Enable OVERHEAD activity type for the XPU backend,https://github.com/pytorch/pytorch/pull/187835,draft/noise,13,PR is draft,wait for author to mark ready,2026-06-22T14:21:55Z
pr,181187,[spmd types] fully_shard but in DTensor land,https://github.com/pytorch/pytorch/pull/181187,draft/noise,13,"PR is draft; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for author to mark ready,2026-06-22T03:48:43Z
pr,180317,[xpu][inductor] Add Dual GEMM SiLU-Mul fusion pass for XPU,https://github.com/pytorch/pytorch/pull/180317,draft/noise,13,PR is draft,wait for author to mark ready,2026-06-19T06:18:01Z
pr,187685,Name AOTAutograd backward gradient outputs grad_<primal> (#110700),https://github.com/pytorch/pytorch/pull/187685,draft/noise,13,"PR is draft; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for author to mark ready,2026-06-18T18:36:46Z
pr,187647,Remove XPU skips on test_partitioner_unbacked_symints and test_sdpa_s…,https://github.com/pytorch/pytorch/pull/187647,draft/noise,13,PR is draft; CI lane labels: ciflow/inductor,wait for author to mark ready,2026-06-18T11:25:08Z
pr,187612,docker-builds: move to default runner group (regular OSDC pool),https://github.com/pytorch/pytorch/pull/187612,draft/noise,13,PR is draft; CI lane labels: ciflow/docker,wait for author to mark ready,2026-06-17T23:59:57Z
pr,187620,[FSDP2] Add PartialOffloadPolicy for fractional CPU parameter offload,https://github.com/pytorch/pytorch/pull/187620,draft/noise,13,"PR is draft; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for author to mark ready,2026-06-17T22:49:51Z
pr,187564,"Register `linalg_qr`, `geqrf`, and `linalg_householder_product` for XPU",https://github.com/pytorch/pytorch/pull/187564,draft/noise,13,PR is draft; CI lane labels: ciflow/inductor,wait for author to mark ready,2026-06-17T09:16:14Z
pr,177649,BLAS sparse ops: dense arguments -- extend support beyond just dense and non-overlapping inputs.,https://github.com/pytorch/pytorch/pull/177649,draft/noise,13,PR is draft; CI lane labels: ciflow/trunk,wait for author to mark ready,2026-06-16T12:24:10Z
pr,186982,Move header-only c10/util concurrency/lifetime utils to torch/headeronly,https://github.com/pytorch/pytorch/pull/186982,draft/noise,13,PR is draft; CI lane labels: ciflow/inductor,wait for author to mark ready,2026-06-16T06:06:41Z
pr,186983,Move header-only c10/util math utils to torch/headeronly,https://github.com/pytorch/pytorch/pull/186983,draft/noise,13,PR is draft; CI lane labels: ciflow/inductor,wait for author to mark ready,2026-06-16T06:06:39Z
pr,186981,Move header-only c10/util metaprogramming/type utils to torch/headeronly,https://github.com/pytorch/pytorch/pull/186981,draft/noise,13,PR is draft; CI lane labels: ciflow/inductor,wait for author to mark ready,2026-06-16T06:06:39Z
pr,186985,Move header-only c10/util low-level utils to torch/headeronly,https://github.com/pytorch/pytorch/pull/186985,draft/noise,13,PR is draft; CI lane labels: ciflow/inductor,wait for author to mark ready,2026-06-16T06:06:31Z
pr,176393,Make avgpool3d support fp16/bf16,https://github.com/pytorch/pytorch/pull/176393,draft/noise,13,PR is draft; CI lane labels: ciflow/trunk,wait for author to mark ready,2026-06-15T03:33:18Z
pr,180460,Decompose all_gather+matmul into P2P+tiled mm for fine-grained overlap,https://github.com/pytorch/pytorch/pull/180460,draft/noise,13,"PR is draft; CI lane labels: ciflow/dtensor, ciflow/inductor, ciflow/torchtitan",wait for author to mark ready,2026-06-14T16:59:39Z
pr,187213,Issue-106571: Lint Fixes - Rule B007 - gb_registry_linter - Replace …,https://github.com/pytorch/pytorch/pull/187213,draft/noise,13,PR is draft,wait for author to mark ready,2026-06-12T21:17:36Z
pr,180087,[DTensor] Error on random ops with device mismatch,https://github.com/pytorch/pytorch/pull/180087,draft/noise,13,"PR is draft; CI lane labels: ciflow/dtensor, ciflow/inductor, ciflow/torchtitan",wait for author to mark ready,2026-06-10T07:55:25Z
pr,186852,[profiler][cupti] ProfilerObserver + record_function annotation routing,https://github.com/pytorch/pytorch/pull/186852,draft/noise,13,PR is draft,wait for author to mark ready,2026-06-10T03:28:56Z
pr,186851,[profiler][cupti] ProfilerObserver + record_function annotation routing,https://github.com/pytorch/pytorch/pull/186851,draft/noise,13,PR is draft,wait for author to mark ready,2026-06-10T02:16:15Z
pr,185217,Fix Debug/RelWithDebInfo PyTorch+XPU on Windows,https://github.com/pytorch/pytorch/pull/185217,draft/noise,13,PR is draft,wait for author to mark ready,2026-06-01T12:10:10Z
pr,185178,[FSDP][AC] Bypass SAC tracking for all-gather internals,https://github.com/pytorch/pytorch/pull/185178,draft/noise,13,"PR is draft; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for author to mark ready,2026-05-29T07:01:24Z
pr,179223,[Test] Refactor test_higher_order_ops.py to be device-generic,https://github.com/pytorch/pytorch/pull/179223,draft/noise,13,"PR is draft; CI lane labels: ciflow/dynamo, ciflow/inductor",wait for author to mark ready,2026-05-25T06:33:10Z
pr,185036,[inductor] Prototype channel-resident reduction suffix,https://github.com/pytorch/pytorch/pull/185036,draft/noise,5,"PR is draft; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for author to mark ready,2026-05-24T06:14:35Z
pr,185037,[inductor] Use persistent path for small fused online softmax,https://github.com/pytorch/pytorch/pull/185037,draft/noise,5,"PR is draft; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for author to mark ready,2026-05-24T06:14:34Z
pr,185027,[inductor] Fold CE backward one-hot row sums,https://github.com/pytorch/pytorch/pull/185027,draft/noise,5,"PR is draft; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for author to mark ready,2026-05-24T06:14:30Z
pr,185028,[inductor] Reassociate cat under surviving reductions,https://github.com/pytorch/pytorch/pull/185028,draft/noise,5,"PR is draft; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for author to mark ready,2026-05-24T06:14:26Z
pr,184941,[inductor] Enable reduction epilogue fusion by deferring split decision,https://github.com/pytorch/pytorch/pull/184941,draft/noise,5,"PR is draft; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for author to mark ready,2026-05-24T04:53:20Z
pr,185030,[inductor] Fold masked invariant expand reductions,https://github.com/pytorch/pytorch/pull/185030,draft/noise,5,"PR is draft; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for author to mark ready,2026-05-24T04:53:15Z
pr,185032,[inductor] Reuse expensive broadcast factors in pointwise tiling,https://github.com/pytorch/pytorch/pull/185032,draft/noise,5,"PR is draft; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for author to mark ready,2026-05-24T04:53:14Z
pr,184932,[inductor] Use non-online softmax for huge inner reductions,https://github.com/pytorch/pytorch/pull/184932,draft/noise,5,"PR is draft; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for author to mark ready,2026-05-24T03:27:57Z
pr,184936,[inductor] Avoid masks for zero-stride cat source loads,https://github.com/pytorch/pytorch/pull/184936,draft/noise,5,"PR is draft; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for author to mark ready,2026-05-24T03:27:52Z
pr,184940,[inductor] Enable reduction+transpose fusion via loop reordering,https://github.com/pytorch/pytorch/pull/184940,draft/noise,5,"PR is draft; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for author to mark ready,2026-05-24T02:15:28Z
pr,184929,[inductor] Sink single-use dropout randoms into fused epilogues,https://github.com/pytorch/pytorch/pull/184929,draft/noise,5,"PR is draft; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for author to mark ready,2026-05-24T01:27:26Z
pr,184998,[inductor] Use one warp for tiny inner persistent reductions,https://github.com/pytorch/pytorch/pull/184998,draft/noise,5,"PR is draft; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for author to mark ready,2026-05-24T01:25:53Z
pr,185003,[inductor] Tune small multi-output mix-order reductions,https://github.com/pytorch/pytorch/pull/185003,draft/noise,5,"PR is draft; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for author to mark ready,2026-05-24T01:25:46Z
pr,184931,[inductor] Use persistent reduction for R=2048 online softmax,https://github.com/pytorch/pytorch/pull/184931,draft/noise,5,"PR is draft; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for author to mark ready,2026-05-23T23:33:20Z
pr,184935,[inductor] Canonicalize duplicate mean and var reductions,https://github.com/pytorch/pytorch/pull/184935,draft/noise,5,"PR is draft; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for author to mark ready,2026-05-23T23:02:53Z
pr,184777,Refactor test_dlpack for out of tree backend,https://github.com/pytorch/pytorch/pull/184777,draft/noise,5,PR is draft; PR has not updated for at least 45 days,wait for author to mark ready,2026-05-23T14:55:29Z
pr,184724,[XPU][SDPA] Add XPU flash_attention_forward dispatch with dropout fallback,https://github.com/pytorch/pytorch/pull/184724,draft/noise,5,PR is draft; PR has not updated for at least 45 days,wait for author to mark ready,2026-05-22T05:18:57Z
pr,182811,Reenable test_addcmul_alpha_one_fma_parity dtypes F32 and F64 on XPU.,https://github.com/pytorch/pytorch/pull/182811,draft/noise,5,PR is draft; PR has not updated for at least 45 days,wait for author to mark ready,2026-05-21T14:14:04Z
pr,177636,Refactor platform support utilities,https://github.com/pytorch/pytorch/pull/177636,draft/noise,5,"PR is draft; PR has not updated for at least 45 days; CI lane labels: ciflow/b200, ciflow/dtensor, ciflow/h100, ciflow/inductor",wait for author to mark ready,2026-05-21T09:06:13Z
pr,184318,[opaque obj] Check opaque object schema types at dispatch,https://github.com/pytorch/pytorch/pull/184318,draft/noise,5,PR is draft; PR has not updated for at least 45 days,wait for author to mark ready,2026-05-20T20:10:14Z
pr,184498,[profiler] Propagate record_funtion_id from TorchOp ancestor to Kinet…,https://github.com/pytorch/pytorch/pull/184498,draft/noise,5,PR is draft; PR has not updated for at least 45 days,wait for author to mark ready,2026-05-20T14:21:28Z
pr,180888,Update requirements-docs.txt,https://github.com/pytorch/pytorch/pull/180888,draft/noise,5,PR is draft; PR has not updated for at least 45 days,wait for author to mark ready,2026-05-20T12:46:35Z
pr,182850,[inductor][symm_mem] Add low-contention allgather variants,https://github.com/pytorch/pytorch/pull/182850,draft/noise,5,"PR is draft; PR has not updated for at least 45 days; CI lane labels: ciflow/h100-symm-mem, ciflow/inductor, ciflow/torchtitan",wait for author to mark ready,2026-05-18T14:13:51Z
pr,183903,[XPU] Relax Muon float32 tolerance in test_state_dict_with_cuda_params,https://github.com/pytorch/pytorch/pull/183903,draft/noise,5,PR is draft; PR has not updated for at least 45 days,wait for author to mark ready,2026-05-18T02:33:45Z
pr,183594,TMA scatter_reduce min/max for sm_90+ (CUDA 12.8+),https://github.com/pytorch/pytorch/pull/183594,draft/noise,5,PR is draft; PR has not updated for at least 45 days,wait for author to mark ready,2026-05-13T20:07:40Z
pr,183360,Prototype lazy activation checkpoint compile,https://github.com/pytorch/pytorch/pull/183360,draft/noise,5,"PR is draft; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for author to mark ready,2026-05-13T03:28:06Z
pr,183359,Prototype lazy autograd.Function backward compile,https://github.com/pytorch/pytorch/pull/183359,draft/noise,5,"PR is draft; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for author to mark ready,2026-05-13T02:15:17Z
pr,183354,[CI] Match MI355 ROCm shards with trunk,https://github.com/pytorch/pytorch/pull/183354,draft/noise,5,PR is draft; PR has not updated for at least 45 days,wait for author to mark ready,2026-05-11T19:30:26Z
pr,183028,[Inductor] Generalize parallel Triton precompile to all multi-config kernels,https://github.com/pytorch/pytorch/pull/183028,draft/noise,5,"PR is draft; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for author to mark ready,2026-05-11T18:22:26Z
pr,183253,[BE]: Improve torch utils checkpoint typing,https://github.com/pytorch/pytorch/pull/183253,draft/noise,5,PR is draft; PR has not updated for at least 45 days; CI lane labels: ciflow/torchtitan,wait for author to mark ready,2026-05-11T10:16:32Z
pr,183211,[FSDP] Break cross-language ref cycle keeping FlatParameter alive,https://github.com/pytorch/pytorch/pull/183211,draft/noise,5,"PR is draft; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for author to mark ready,2026-05-11T03:11:52Z
pr,178850,[xpu][feature] Add fork-safe is_available() via PYTORCH_ZES_BASED_XPU_CHECK env var,https://github.com/pytorch/pytorch/pull/178850,draft/noise,5,PR is draft; PR has not updated for at least 45 days; CI lane labels: ciflow/trunk,wait for author to mark ready,2026-03-31T13:27:45Z
pr,153557,"[PP] wip, allow grad to be None",https://github.com/pytorch/pytorch/pull/153557,draft/noise,5,PR is draft; PR has not updated for at least 45 days; CI lane labels: ciflow/torchtitan,wait for author to mark ready,2026-03-20T22:40:44Z
pr,153317,[associative_scan] Autograd for additional inputs,https://github.com/pytorch/pytorch/pull/153317,draft/noise,5,PR is draft; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor,wait for author to mark ready,2025-07-09T18:22:39Z
pr,147653,Bf16 fused adam(W),https://github.com/pytorch/pytorch/pull/147653,draft/noise,5,PR is draft; PR has not updated for at least 45 days,wait for author to mark ready,2025-05-14T17:02:21Z
pr,142295,[inductor] Decide cooperative RSPLIT with same algorithm as split reductions,https://github.com/pytorch/pytorch/pull/142295,draft/noise,5,PR is draft; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor,wait for author to mark ready,2025-04-10T16:01:15Z
pr,143812,[inductor] Used fixed configs for contiguous reductions,https://github.com/pytorch/pytorch/pull/143812,draft/noise,5,PR is draft; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor,wait for author to mark ready,2025-04-10T15:54:03Z
pr,146593,[NOT FOR LANDING] experimental NVSHMEM integration,https://github.com/pytorch/pytorch/pull/146593,draft/noise,5,PR is draft; PR has not updated for at least 45 days,wait for author to mark ready,2025-04-09T18:59:30Z
pr,139524,Make anomaly mode asynchronous,https://github.com/pytorch/pytorch/pull/139524,draft/noise,5,PR is draft; PR has not updated for at least 45 days,wait for author to mark ready,2025-01-06T02:46:38Z
pr,125270,[do not review] Add tensor equality subsystem and get it working with compile,https://github.com/pytorch/pytorch/pull/125270,draft/noise,5,PR is draft; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor,wait for author to mark ready,2024-12-17T22:23:20Z
pr,125428,[do not review] support max_seqlen,https://github.com/pytorch/pytorch/pull/125428,draft/noise,5,PR is draft; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor,wait for author to mark ready,2024-10-07T03:35:08Z
pr,125326,[do not review] Use union find and new Python NestedInt with NJT,https://github.com/pytorch/pytorch/pull/125326,draft/noise,5,PR is draft; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor,wait for author to mark ready,2024-10-07T03:35:07Z
pr,126199,[POC] use a better default for num_threads,https://github.com/pytorch/pytorch/pull/126199,draft/noise,5,PR is draft; PR has not updated for at least 45 days,wait for author to mark ready,2024-07-15T05:46:38Z
pr,88106,WIP: feat: LARS optimizer,https://github.com/pytorch/pytorch/pull/88106,draft/noise,5,PR is draft; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor,wait for author to mark ready,2024-01-30T19:47:12Z
pr,84843,Support different NSE in batches of CSR and CSC tensors,https://github.com/pytorch/pytorch/pull/84843,draft/noise,5,PR is draft; PR has not updated for at least 45 days,wait for author to mark ready,2023-12-12T17:41:33Z
pr,85699,Add vector-Jacobian-products for a subset of nvFuser-supported prims; add backward support for nvprims,https://github.com/pytorch/pytorch/pull/85699,draft/noise,5,PR is draft; PR has not updated for at least 45 days,wait for author to mark ready,2023-04-17T00:36:00Z
pr,84545,Add nvfuser support for prims.copy_to,https://github.com/pytorch/pytorch/pull/84545,draft/noise,5,PR is draft; PR has not updated for at least 45 days,wait for author to mark ready,2023-04-17T00:36:00Z
issue,184036,"torch.nn.PairwiseDistance returns +Inf for moderately large p (overflow in fp32, no log-space rewrite)",https://github.com/pytorch/pytorch/issues/184036,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: distance functions, module: edge cases, module: linear algebra",scan when time allows,2026-06-22T22:12:43Z
issue,173225,[RFC] Renaming MPS backend as Metal backend,https://github.com/pytorch/pytorch/issues/173225,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: mps,scan when time allows,2026-05-30T13:50:29Z
issue,177082,Stable Linux binaries of LibTorch have broken soname and versioning in libgomp,https://github.com/pytorch/pytorch/issues/177082,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: build",scan when time allows,2026-05-09T16:44:06Z
issue,111950,Operators that return dynamic-shape outputs that require_grad choke in AOTAutograd / unbacked SymInts are reallocated,https://github.com/pytorch/pytorch/issues/111950,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: aotdispatch, module: dynamic shapes, module: pt2-dispatcher, oncall: pt2",scan when time allows,2026-03-31T23:22:16Z
issue,94652,dynamo: handle contiguous graph breaks,https://github.com/pytorch/pytorch/issues/94652,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamo, module: graph breaks, oncall: pt2",scan when time allows,2026-03-11T22:17:58Z
issue,102898,After dynamo minifier generates repros that don't entirely match what we minified over,https://github.com/pytorch/pytorch/issues/102898,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dynamo,scan when time allows,2026-03-11T20:36:15Z
issue,174986,`torch.istft` returns an unhelpful internal error,https://github.com/pytorch/pytorch/issues/174986,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: error checking, module: fft",scan when time allows,2026-03-02T15:33:09Z
issue,169058,[MPS] Inconsistent overflow behavior when casting Float to Int8 compared to CPU/CUDA (Saturation vs Wraparound),https://github.com/pytorch/pytorch/issues/169058,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: edge cases, module: mps",scan when time allows,2026-02-19T22:18:23Z
issue,165537,"`torch.linspace` with `inf` endpoint and `float16` dtype returns all `nan`,inconsistent with NumPy",https://github.com/pytorch/pytorch/issues/165537,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nans and infs,scan when time allows,2026-02-19T18:52:35Z
issue,122290,"In a simple model, exporting onnx loses model operations",https://github.com/pytorch/pytorch/issues/122290,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2026-01-16T15:09:25Z
issue,170573,"Silent Heap Buffer Overflow in standard ops (add, pinv) via implicit view resizing",https://github.com/pytorch/pytorch/issues/170573,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python frontend,scan when time allows,2026-01-10T00:01:07Z
issue,172085,Many test failures when GPU does not support BF16,https://github.com/pytorch/pytorch/issues/172085,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bfloat16, module: cuda, module: tests",scan when time allows,2026-01-09T21:17:56Z
issue,170563,vLLM v1/distributed/test_eagle_dp.py Tests failing on PyTorch 2.10,https://github.com/pytorch/pytorch/issues/170563,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: vllm,scan when time allows,2026-01-09T18:05:14Z
issue,149187,Issue with Shared CUDA Tensor Reference Counting in Multi-Processing,https://github.com/pytorch/pytorch/issues/149187,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: multiprocessing",scan when time allows,2026-01-09T10:18:54Z
issue,149062,Reserve customized modules in torch.compile's dynamo tracer,https://github.com/pytorch/pytorch/issues/149062,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",scan when time allows,2026-01-09T07:44:18Z
issue,171697,"When the pinned_use_background_threads option is first set to True and then set to False, the background threads may not stop",https://github.com/pytorch/pytorch/issues/171697,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2026-01-09T07:06:49Z
issue,172034,Become compatible with CMAKE_CUDA_ARCHITECTURES,https://github.com/pytorch/pytorch/issues/172034,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cpp-extensions, module: cuda",scan when time allows,2026-01-09T00:26:36Z
issue,171534,Add proper FMA support for Pallas backend,https://github.com/pytorch/pytorch/issues/171534,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, module: pallas",scan when time allows,2026-01-08T20:27:58Z
issue,171956,[Inductor] Eliminate triton-related hardcode,https://github.com/pytorch/pytorch/issues/171956,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: privateuse1,scan when time allows,2026-01-08T07:16:24Z
issue,160787,[Intel Arc XPU] NaN Values in Gradients after SDPA with is_causal,https://github.com/pytorch/pytorch/issues/160787,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: xpu,scan when time allows,2026-01-08T07:10:02Z
issue,168935,[ROCm] enabling compile should not cause signficiant accuracy loss on many important models,https://github.com/pytorch/pytorch/issues/168935,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rocm,scan when time allows,2026-01-08T06:04:05Z
issue,166114,[BUG]interpolate operator triggers severe performance degradation when enabling deterministic computation,https://github.com/pytorch/pytorch/issues/166114,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: determinism,scan when time allows,2026-01-07T20:00:36Z
issue,169893,Investigate which submodules in third_party/ can be omitted from stable header hiding,https://github.com/pytorch/pytorch/issues/169893,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: cpp-extensions",scan when time allows,2026-01-07T16:36:19Z
issue,144766,Matmul with int32 parameters on Intel GPU leads to errors,https://github.com/pytorch/pytorch/issues/144766,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: xpu,scan when time allows,2026-01-07T07:32:06Z
issue,151549,multiprocessing.pool.MaybeEncodingError Reason: 'RuntimeError('_share_fd_: only available on CPU')',https://github.com/pytorch/pytorch/issues/151549,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: multiprocessing, module: xpu",scan when time allows,2026-01-07T07:30:08Z
issue,151308,[Intel GPU][UT failure] Depthwise conv related UT failures,https://github.com/pytorch/pytorch/issues/151308,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: xpu,scan when time allows,2026-01-07T07:26:00Z
issue,159232,Support installing Python bindings in CMake,https://github.com/pytorch/pytorch/issues/159232,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2026-01-07T00:16:27Z
issue,171771,[RFC] Optimize CUDA Allocator's Synchronization Behavior During OOM,https://github.com/pytorch/pytorch/issues/171771,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: cudacachingallocator",scan when time allows,2026-01-06T22:56:52Z
issue,169073,Replace deprecated __args__ type access in typing libraries with proper typing.get_args,https://github.com/pytorch/pytorch/issues/169073,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: typing,scan when time allows,2026-01-06T21:00:08Z
issue,168376,`TORCH_SHOW_CPP_STACKTRACES=1` often fails to display all function names,https://github.com/pytorch/pytorch/issues/168376,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: debug-build, module: error checking",scan when time allows,2026-01-06T19:40:01Z
issue,74627,Feature request: INT4 format support,https://github.com/pytorch/pytorch/issues/74627,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2026-01-06T13:15:18Z
issue,95100,"RuntimeError: view_as_complex is only supported for half, float and double tensors, but got a tensor of scalar type: BFloat16",https://github.com/pytorch/pytorch/issues/95100,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bfloat16, module: complex",scan when time allows,2026-01-06T10:09:11Z
issue,149694,torch.det results in nans for torch.func.hessian,https://github.com/pytorch/pytorch/issues/149694,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: functorch, module: numerical-stability",scan when time allows,2026-01-05T23:58:26Z
issue,171302,Make ConcatOptim class for using multiple optimizers as one,https://github.com/pytorch/pytorch/issues/171302,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2026-01-05T18:56:36Z
issue,171263,"[CUDA Graphs] Capture failure leaves generator state corrupted, causing ""Offset increment outside graph capture encountered unexpectedly"" on subsequent RNG operations",https://github.com/pytorch/pytorch/issues/171263,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda graphs,scan when time allows,2026-01-05T18:54:40Z
issue,171401,Feedback about Flatten,https://github.com/pytorch/pytorch/issues/171401,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2026-01-05T18:53:41Z
issue,171323,Segmentation fault when calling `torch.tensor()` after importing xgboost,https://github.com/pytorch/pytorch/issues/171323,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: macos, module: openmp",scan when time allows,2026-01-05T18:30:05Z
issue,171286,Memory Viz Y-Axis Uses SI Units (GB) Instead of Binary Units (GiB),https://github.com/pytorch/pytorch/issues/171286,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, oncall: profiler",scan when time allows,2026-01-05T18:17:07Z
issue,171536,`torch.nn.PixelUnshuffle` throws RuntimeError: Stride calculation overflowed on CUDA with empty input and large downscale factor (CPU succeeds),https://github.com/pytorch/pytorch/issues/171536,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: error checking,scan when time allows,2026-01-05T15:02:18Z
issue,107352,dist.scatter is incompatible with transpose/permute operation,https://github.com/pytorch/pytorch/issues/107352,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2026-01-05T07:54:13Z
issue,168696,Test: TestConvolutionNNDeviceType.test_conv3d_cudnn_broken,https://github.com/pytorch/pytorch/issues/168696,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: rocm",scan when time allows,2026-01-03T16:19:45Z
issue,155992,[feature request] Native checkpointing to/from `s3://` (DCP / `torch.load` / `torch.save`),https://github.com/pytorch/pytorch/issues/155992,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: serialization, oncall: distributed checkpointing",scan when time allows,2026-01-02T11:54:25Z
issue,91965,Proxy/cache server option/hooks for downloading model checkpoints and dataset archive files in cloud environment,https://github.com/pytorch/pytorch/issues/91965,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: hub,scan when time allows,2026-01-02T10:36:49Z
issue,140130,[onnx] [njt] [feature request] Export NJT-enabled SDPA / MHA ops to ORT's PackingMode Attention,https://github.com/pytorch/pytorch/issues/140130,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2025-12-30T21:20:37Z
issue,167210,[RFC] Introduce the Unified Memory Pool and Pluggable Allocator in torch.accelerator,https://github.com/pytorch/pytorch/issues/167210,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: accelerator,scan when time allows,2025-12-30T10:48:36Z
issue,167170,[XPU][Tracker] Enable `torch._scaled_matmul`,https://github.com/pytorch/pytorch/issues/167170,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: xpu,scan when time allows,2025-12-30T04:59:10Z
issue,171179,"NCCL error: /pytorch/torch/csrc/distributed/c10d/NCCLUtils.hpp:268, NCCL version 2.22.3 ncclSystemError: System call (e.g. socket, malloc) or external library call failed or device error. Last error: socketStartConnect: Connect to IP<51423> failed",https://github.com/pytorch/pytorch/issues/171179,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2025-12-30T03:26:14Z
issue,120587,Type hints on nn.Module.__setattr__ value,https://github.com/pytorch/pytorch/issues/120587,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: typing",scan when time allows,2025-12-29T13:37:04Z
issue,11936,torch.bmm doesn't support CUDA uint8 (byte) tensor,https://github.com/pytorch/pytorch/issues/11936,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, module: cuda, module: linear algebra",scan when time allows,2025-12-29T02:39:54Z
issue,169870,Capturing ViewAndMutationMeta for training graphs for PyTorch 2.8,https://github.com/pytorch/pytorch/issues/169870,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: viewing and reshaping, oncall: pt2",scan when time allows,2025-12-28T22:09:55Z
issue,86060,[MPS] Pin_memory on MPS backend raises exception of setting wrong device storage,https://github.com/pytorch/pytorch/issues/86060,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: mps,scan when time allows,2025-12-28T18:36:21Z
issue,169191,dim broadcast in masked tensor between mask and data.,https://github.com/pytorch/pytorch/issues/169191,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: masked operators,scan when time allows,2025-12-28T16:08:23Z
issue,112377,"RuntimeError: NVML_SUCCESS == r INTERNAL ASSERT FAILED at ""/opt/conda/conda-bld/pytorch_1695392020201/work/c10/cuda/CUDACachingAllocator.cpp"":1154, please report a bug to PyTorch.",https://github.com/pytorch/pytorch/issues/112377,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cudacachingallocator,scan when time allows,2025-12-28T11:09:49Z
issue,170496,torch.func.grad does not support FlashAttention autograd function,https://github.com/pytorch/pytorch/issues/170496,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2025-12-28T04:38:40Z
issue,163504,[MPS] Unexpected behavior in torch.nn.functional.one_hot when target values exceed num_classes in MPS,https://github.com/pytorch/pytorch/issues/163504,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: error checking, module: mps",scan when time allows,2025-12-27T18:17:53Z
issue,168722,Test: TestCudaMallocAsync.test_device_memory_used,https://github.com/pytorch/pytorch/issues/168722,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rocm,scan when time allows,2025-12-27T06:24:29Z
issue,1468,avg_pool functions hold input for backward,https://github.com/pytorch/pytorch/issues/1468,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: pooling,scan when time allows,2025-12-26T22:36:28Z
issue,166853,torch.randn on CPU is not accurate with dtype=bfloat16,https://github.com/pytorch/pytorch/issues/166853,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bfloat16, module: correctness (silent), module: random",scan when time allows,2025-12-26T11:59:28Z
issue,88810,Return the attention weights using the Transformer Encoder class.,https://github.com/pytorch/pytorch/issues/88810,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2025-12-26T05:27:02Z
issue,167762,Conan Support,https://github.com/pytorch/pytorch/issues/167762,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp,scan when time allows,2025-12-26T04:21:15Z
issue,81876,dtype mismatch when after using auto mixed precision,https://github.com/pytorch/pytorch/issues/81876,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision),scan when time allows,2025-12-26T02:46:41Z
issue,171184,No hash in CachingAutotuner meta information for some kernels,https://github.com/pytorch/pytorch/issues/171184,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",scan when time allows,2025-12-25T18:52:15Z
issue,69610,[Question] How to extract/expose the complete PyTorch computation graph (forward and backward)?,https://github.com/pytorch/pytorch/issues/69610,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, oncall: visualization",scan when time allows,2025-12-24T06:43:52Z
issue,168774,Test: TestMatmulCuda.test_cublas_addmm,https://github.com/pytorch/pytorch/issues/168774,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: linear algebra, module: rocm",scan when time allows,2025-12-24T06:15:07Z
issue,63138,Sign in slogdet is set to requires_grad = False even when using complex numbers.,https://github.com/pytorch/pytorch/issues/63138,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: complex, module: linear algebra",scan when time allows,2025-12-24T00:43:35Z
issue,77230,fast `gradcheck` fails when outputs that do not require grad precede outputs that do,https://github.com/pytorch/pytorch/issues/77230,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: linear algebra",scan when time allows,2025-12-24T00:40:49Z
issue,167825,extra mem copy for value to assign when dispatching `index_put` to `masked_fill` on PrivateUse1 backend.,https://github.com/pytorch/pytorch/issues/167825,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: advanced indexing, module: privateuse1",scan when time allows,2025-12-23T08:42:09Z
issue,171100,Feedback about torch.bernoulli,https://github.com/pytorch/pytorch/issues/171100,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2025-12-23T05:29:40Z
issue,164041,[vLLM] AssertionError: out= ops aren't yet supported,https://github.com/pytorch/pytorch/issues/164041,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, module: vllm, oncall: pt2",scan when time allows,2025-12-22T21:56:05Z
issue,171039,"When using torch.multiprocessing for CUDA IPC, memory cannot be released.",https://github.com/pytorch/pytorch/issues/171039,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: multiprocessing",scan when time allows,2025-12-22T17:46:19Z
issue,170936,NaN-checking anomaly mode should work with DTensor,https://github.com/pytorch/pytorch/issues/170936,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dtensor,scan when time allows,2025-12-22T17:41:37Z
issue,68966,`torch.ldexp` generated tests fail on call to `torch.mul`,https://github.com/pytorch/pytorch/issues/68966,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2025-12-22T14:09:32Z
issue,168807,Test: TestNNDeviceType.test_upsamplingNearest2d_launch_fail,https://github.com/pytorch/pytorch/issues/168807,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rocm,scan when time allows,2025-12-22T02:06:30Z
issue,170674,Flex attention cannot support page size/block mask size != 128 for prefill.,https://github.com/pytorch/pytorch/issues/170674,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2025-12-20T23:38:01Z
issue,170851,Add forward_by_train and forward_by_eval convenience methods to nn.Module,https://github.com/pytorch/pytorch/issues/170851,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2025-12-20T22:55:45Z
issue,76750,Bug in dataloader iterator found by mypy,https://github.com/pytorch/pytorch/issues/76750,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2025-12-20T22:35:36Z
issue,80256,[complex] dropout and it's variants should support complex tensors,https://github.com/pytorch/pytorch/issues/80256,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: complex, module: nn",scan when time allows,2025-12-20T12:29:37Z
issue,170168,torch.linspace with integer dtype gives inconsistent results on CPU vs CUDA,https://github.com/pytorch/pytorch/issues/170168,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: edge cases, module: python frontend",scan when time allows,2025-12-20T09:21:36Z
issue,170694,Proposal to simplify `DeviceThreadHandlePool`,https://github.com/pytorch/pytorch/issues/170694,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2025-12-20T02:58:29Z
issue,168450,Test: DistTensorOpsTest.test_index,https://github.com/pytorch/pytorch/issues/168450,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ddp, module: rocm",scan when time allows,2025-12-19T21:42:03Z
issue,170542,BuildExtension(no_python_abi_suffix=True).get_ext_filename result is incorrect when ABI part is not present in the filename.,https://github.com/pytorch/pytorch/issues/170542,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp-extensions,scan when time allows,2025-12-19T21:32:38Z
issue,168448,Test: DistMatrixOpsTest.test_grouped_mm,https://github.com/pytorch/pytorch/issues/168448,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ddp, module: rocm",scan when time allows,2025-12-19T21:21:22Z
issue,170495,vLLM Blackwell Fusion and Compile Tests failing PyTorch 2.10,https://github.com/pytorch/pytorch/issues/170495,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: vllm,scan when time allows,2025-12-19T19:48:35Z
issue,169479,Discussion on addition Spin commands that would be useful,https://github.com/pytorch/pytorch/issues/169479,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: devx,scan when time allows,2025-12-19T16:02:52Z
issue,88380,cuDNN error (CUDNN_STATUS_NOT_SUPPORTED) for torch.nn.functional.grid_sample(),https://github.com/pytorch/pytorch/issues/88380,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: cudnn",scan when time allows,2025-12-19T09:28:53Z
issue,147274,PaddleOCR PyTorch conflicts,https://github.com/pytorch/pytorch/issues/147274,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: windows",scan when time allows,2025-12-19T07:08:21Z
issue,170284,selective activation checkpoint compatibility to async activation offload,https://github.com/pytorch/pytorch/issues/170284,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: activation checkpointing,scan when time allows,2025-12-19T05:35:46Z
issue,169146,bad numerics for variable length attention with cudnn,https://github.com/pytorch/pytorch/issues/169146,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cudnn, module: sdpa",scan when time allows,2025-12-19T01:34:06Z
issue,168858,Test: TestTesting.test_cuda_assert_should_not_stop_common_distributed_test_suite,https://github.com/pytorch/pytorch/issues/168858,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rocm,scan when time allows,2025-12-18T23:17:09Z
issue,170691,[Feature Request] Add StateSpaceModel to torch.nn,https://github.com/pytorch/pytorch/issues/170691,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2025-12-18T15:50:59Z
issue,170458,must_in_dispatch takes a long time in torch registered ops,https://github.com/pytorch/pytorch/issues/170458,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: torchbind,scan when time allows,2025-12-18T15:39:31Z
issue,170426,argmax over multiple axis,https://github.com/pytorch/pytorch/issues/170426,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python frontend,scan when time allows,2025-12-18T15:37:31Z
issue,169154,AOTIModelPackageLoader occupies too much disk space when launching multiple instances,https://github.com/pytorch/pytorch/issues/169154,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: aotinductor, oncall: export, oncall: pt2",scan when time allows,2025-12-18T02:26:48Z
issue,167526,Missing documentation for CUTLASS backend,https://github.com/pytorch/pytorch/issues/167526,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: docs",scan when time allows,2025-12-17T15:25:44Z
issue,154606,`torch.cumprod` for complex128 produces inconsistent results between CPU and CUDA,https://github.com/pytorch/pytorch/issues/154606,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: complex, module: edge cases, module: nans and infs",scan when time allows,2025-12-17T12:39:20Z
issue,166512,Enable GitHub Actions app to support ppc64le CI jobs using IBM-hosted runners,https://github.com/pytorch/pytorch/issues/166512,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ci, module: power",scan when time allows,2025-12-17T11:29:42Z
issue,163984,[inductor] Convert complex operations to real-block matrices,https://github.com/pytorch/pytorch/issues/163984,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: complex, oncall: pt2",scan when time allows,2025-12-17T10:58:06Z
issue,170402,Write more test cases for Conv2d to check for 0 steps,https://github.com/pytorch/pytorch/issues/170402,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2025-12-17T08:29:25Z
issue,170536,accept int padding value for F.pad,https://github.com/pytorch/pytorch/issues/170536,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: padding,scan when time allows,2025-12-17T05:27:50Z
issue,170350,[feature request] Checked version of indexing/embedding ops,https://github.com/pytorch/pytorch/issues/170350,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: advanced indexing, module: embedding, module: nn",scan when time allows,2025-12-17T05:09:46Z
issue,170382,Feedback about torch.nn.functional.interpolate,https://github.com/pytorch/pytorch/issues/170382,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2025-12-17T04:40:17Z
issue,162606,Tensorpipe - ROCm support,https://github.com/pytorch/pytorch/issues/162606,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: rocm, module: tensorpipe",scan when time allows,2025-12-17T02:56:09Z
issue,76557,Pip packaging and publishing improvements in pytorch wheels for better integration with poetry,https://github.com/pytorch/pytorch/issues/76557,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: binaries,scan when time allows,2025-12-16T23:15:01Z
issue,164878,Ban and remove plain asserts with no message in our python code,https://github.com/pytorch/pytorch/issues/164878,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: error checking, module: python frontend",scan when time allows,2025-12-16T20:02:43Z
issue,52596,pickle is a security issue,https://github.com/pytorch/pytorch/issues/52596,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: hub, module: pickle, module: serialization",scan when time allows,2025-12-16T16:57:23Z
issue,124018,Fix OOMs during parallel builds,https://github.com/pytorch/pytorch/issues/124018,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2025-12-16T15:17:02Z
issue,170345,[vllm] Pytorch failure lazy_imports.py is failing on torch._inductor.async_compile for release 2.9.0 and 2.10.0 rc1,https://github.com/pytorch/pytorch/issues/170345,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, module: vllm, oncall: pt2",scan when time allows,2025-12-16T00:16:40Z
issue,144810,Support torch.func.grad for Flex Attention,https://github.com/pytorch/pytorch/issues/144810,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: flex attention, module: functorch, module: higher order operators, module: pt2-dispatcher",scan when time allows,2025-12-15T23:21:58Z
issue,136623,The MPS Backend sometimes samples outside of distribution domain with `multinomial`,https://github.com/pytorch/pytorch/issues/136623,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: mps",scan when time allows,2025-12-15T16:09:24Z
issue,170365,[CI][CUDA][GH100] LinAlg Unit Test Failures,https://github.com/pytorch/pytorch/issues/170365,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ci, module: cuda",scan when time allows,2025-12-15T16:05:11Z
issue,165141,"[ROCm][RDNA3/GFX1100] MIOpen Conv3d catastrophically fails on large batches with bfloat16, causing vLLM to extremely slowdown",https://github.com/pytorch/pytorch/issues/165141,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: performance, module: rocm",scan when time allows,2025-12-15T12:51:29Z
issue,71379,Leaky cmake cuda compile options,https://github.com/pytorch/pytorch/issues/71379,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cpp, module: cpp-extensions",scan when time allows,2025-12-15T10:59:19Z
issue,169816,"Cannot load or export TorchVision quantized models on Windows - ""Quantized backend not supported""",https://github.com/pytorch/pytorch/issues/169816,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: onnx, module: windows, oncall: quantization",scan when time allows,2025-12-15T03:35:43Z
issue,168863,Test: TestTorchDeviceType.test_cov,https://github.com/pytorch/pytorch/issues/168863,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rocm,scan when time allows,2025-12-12T19:08:25Z
issue,169078,Custom Non-Integer Indexing,https://github.com/pytorch/pytorch/issues/169078,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2025-12-12T18:11:31Z
issue,89438,Allow `low` and `high` to be tensors in `torch.randint`,https://github.com/pytorch/pytorch/issues/89438,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: random,scan when time allows,2025-12-12T17:51:11Z
issue,170266,pytorchbot does not react to Copilot PRs,https://github.com/pytorch/pytorch/issues/170266,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2025-12-12T16:23:51Z
issue,167447,Memory related APIs in torch.accelerator are not supported on MPS,https://github.com/pytorch/pytorch/issues/167447,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: accelerator, module: mps",scan when time allows,2025-12-12T11:21:14Z
issue,169294,Conv2d CPU performance 2-3x slower on macOS due to missing MKLDNN backend,https://github.com/pytorch/pytorch/issues/169294,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: arm, module: build, module: macos, module: performance",scan when time allows,2025-12-12T10:37:14Z
issue,119215,[c10d] Workaround the ncclCommAbort() hang issues,https://github.com/pytorch/pytorch/issues/119215,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: c10d, oncall: distributed",scan when time allows,2025-12-12T08:52:58Z
issue,81405,Implement shape/size functions for nestedtensor,https://github.com/pytorch/pytorch/issues/81405,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nestedtensor,scan when time allows,2025-12-12T05:01:01Z
issue,151066,[ROCm] QR decomposition is much slower on MI300x than A100,https://github.com/pytorch/pytorch/issues/151066,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: performance, module: rocm",scan when time allows,2025-12-11T17:50:16Z
issue,170175,"[MPS][Inconsistency] nonzero() incorrectly preserves non-contiguous strides from sliced inputs, causing .view() crash on CPU/CUDA,but pass on MPS",https://github.com/pytorch/pytorch/issues/170175,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: mps,scan when time allows,2025-12-11T14:38:11Z
issue,132624,Add support for float8 dtypes for the MPS backend,https://github.com/pytorch/pytorch/issues/132624,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: floatx (formerly float8), module: mps",scan when time allows,2025-12-11T08:47:37Z
issue,144445,NotImplementedError: Output channels > 65536 not supported at the MPS device.,https://github.com/pytorch/pytorch/issues/144445,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: mps",scan when time allows,2025-12-11T07:12:42Z
issue,168743,Test: TestCudaPrimaryCtx.test_set_device_0,https://github.com/pytorch/pytorch/issues/168743,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rocm,scan when time allows,2025-12-11T06:18:30Z
issue,168800,Test: TestNN.test_batchnorm,https://github.com/pytorch/pytorch/issues/168800,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rocm,scan when time allows,2025-12-11T05:11:51Z
issue,136833,Thread safety issue with torch.compile(),https://github.com/pytorch/pytorch/issues/136833,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamo, module: inductor, module: pt2-dispatcher, oncall: pt2",scan when time allows,2025-12-11T00:57:34Z
issue,140322,torch_dispatch x HOP handling,https://github.com/pytorch/pytorch/issues/140322,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dispatch, module: higher order operators, module: pt2-dispatcher, oncall: pt2",scan when time allows,2025-12-10T19:28:29Z
issue,106388,Better export story for autograd.Function?,https://github.com/pytorch/pytorch/issues/106388,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: pt2,scan when time allows,2025-12-10T08:28:37Z
issue,133751,"`torch.nn.LSTM` using cuDNN fails with sequences longer than 65,535",https://github.com/pytorch/pytorch/issues/133751,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: 64-bit, module: cuda, module: cudnn, module: nn",scan when time allows,2025-12-10T01:47:50Z
issue,169955,[Bug] Potential command injection risk (shell=True) in subprocess,https://github.com/pytorch/pytorch/issues/169955,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2025-12-10T00:24:54Z
issue,169934,Autocast cache not updated by grad enabled state,https://github.com/pytorch/pytorch/issues/169934,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision), module: autograd",scan when time allows,2025-12-09T19:08:40Z
issue,169909,[CI] cudaErrorInvalidValue in test_std_cuda_kernel_launch_check_error_show_cpp_stacktraces,https://github.com/pytorch/pytorch/issues/169909,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: tests",scan when time allows,2025-12-09T19:01:18Z
issue,169956,`TestLinalgCUDA.test__int_mm` fails false/false case on A100,https://github.com/pytorch/pytorch/issues/169956,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: linear algebra, module: tests",scan when time allows,2025-12-09T18:53:42Z
issue,169797,"When augment_with_fx_traces=True but the user has misconfigured the FX config, raise an error",https://github.com/pytorch/pytorch/issues/169797,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: fx, oncall: profiler",scan when time allows,2025-12-09T18:23:40Z
issue,116350,RuntimeError: derivative for aten::_scaled_dot_product_flash_attention_backward is not implemented,https://github.com/pytorch/pytorch/issues/116350,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sdpa,scan when time allows,2025-12-09T16:34:09Z
issue,169184,[RFC] Speed-up checkout operations in CI with treeless checkouts,https://github.com/pytorch/pytorch/issues/169184,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2025-12-09T16:23:41Z
issue,36577,Add load_state_dict and state_dict() in C++,https://github.com/pytorch/pytorch/issues/36577,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp,scan when time allows,2025-12-09T14:20:14Z
issue,169738,"[MPS][Inductor] AdaptiveMaxPool{1,2}d produces incorrect results (numerical correctness)",https://github.com/pytorch/pytorch/issues/169738,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: mps, oncall: pt2",scan when time allows,2025-12-09T12:07:58Z
issue,77962,Operating on boolean torch tensor and numpy array casts to `unit8`,https://github.com/pytorch/pytorch/issues/77962,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: boolean tensor, module: numpy, module: type promotion",scan when time allows,2025-12-09T11:04:56Z
issue,120710,Reduce duplication in number of places where we specify dependency pin,https://github.com/pytorch/pytorch/issues/120710,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ci, module: devx",scan when time allows,2025-12-09T09:20:45Z
issue,169763,Multiple torch.load() call on the same BinaryIO,https://github.com/pytorch/pytorch/issues/169763,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: serialization,scan when time allows,2025-12-09T04:47:10Z
issue,169469,AOTI infinite recursion from a scalar buffer,https://github.com/pytorch/pytorch/issues/169469,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: aotinductor, oncall: export, oncall: pt2",scan when time allows,2025-12-09T01:57:41Z
issue,67602,[feature request] Add no-op set_epoch on Sampler class,https://github.com/pytorch/pytorch/issues/67602,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2025-12-09T01:51:09Z
issue,168977,Easier way to correlate DebugMode with FX graph,https://github.com/pytorch/pytorch/issues/168977,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: pt2-dispatcher, oncall: pt2",scan when time allows,2025-12-08T23:52:11Z
issue,157648,RuntimeError: operator torchvision::nms does not exist,https://github.com/pytorch/pytorch/issues/157648,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: hub,scan when time allows,2025-12-08T13:34:20Z
issue,117582,Add forward AD support for nn.functional.grid_sample,https://github.com/pytorch/pytorch/issues/117582,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: forward ad, module: nn",scan when time allows,2025-12-08T12:21:23Z
issue,162346,The batching rule of aten::affine_grid_generator is not implemented,https://github.com/pytorch/pytorch/issues/162346,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: functorch, module: vmap",scan when time allows,2025-12-08T03:43:52Z
issue,144850,Batching rule for aten::_thnn_fused_gru_cell,https://github.com/pytorch/pytorch/issues/144850,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: functorch,scan when time allows,2025-12-08T03:25:29Z
issue,149821,`Aborted` error when using `torch.cuda.memory.caching_allocator_delete`,https://github.com/pytorch/pytorch/issues/149821,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: cudacachingallocator",scan when time allows,2025-12-06T09:14:35Z
issue,169675,Header-hiding when TORCH_TARGET_VERSION is set is broken for editable torch installs,https://github.com/pytorch/pytorch/issues/169675,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cpp-extensions",scan when time allows,2025-12-06T00:25:05Z
issue,102963,torch.svd fails on large matrices,https://github.com/pytorch/pytorch/issues/102963,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cublas, module: cuda, module: linear algebra",scan when time allows,2025-12-05T11:09:55Z
issue,169290,[MPS][Inductor] LayerNorm/BatchNorm/GroupNorm fail to compile with dynamic=True due to invalid Metal codegen (VLA & Min),https://github.com/pytorch/pytorch/issues/169290,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: mps, oncall: pt2",scan when time allows,2025-12-05T02:50:23Z
issue,161105,UED guard check failure,https://github.com/pytorch/pytorch/issues/161105,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",scan when time allows,2025-12-05T01:56:35Z
issue,164116,qta and export onnx error,https://github.com/pytorch/pytorch/issues/164116,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: onnx, oncall: quantization",scan when time allows,2025-12-05T01:53:34Z
issue,163052,CPU dynamic quantization (qint8) cannot be used with multiprocessing,https://github.com/pytorch/pytorch/issues/163052,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpu, oncall: quantization",scan when time allows,2025-12-05T00:53:35Z
issue,160651,`dequantize_per_channel` in `quantized_decomposed` not compatible with Float8 types,https://github.com/pytorch/pytorch/issues/160651,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: floatx (formerly float8), oncall: quantization",scan when time allows,2025-12-05T00:53:05Z
issue,168529,Test: AsyncTPTest.test_optimal_layout,https://github.com/pytorch/pytorch/issues/168529,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: rocm, module: symm_mem",scan when time allows,2025-12-04T21:20:01Z
issue,168528,Test: AsyncTPTest.test_fused_scaled_matmul_reduce_scatter,https://github.com/pytorch/pytorch/issues/168528,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: rocm, module: symm_mem",scan when time allows,2025-12-04T21:19:56Z
issue,168527,Test: AsyncTPTest.test_fused_matmul_reduce_scatter,https://github.com/pytorch/pytorch/issues/168527,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: rocm, module: symm_mem",scan when time allows,2025-12-04T21:19:51Z
issue,168526,Test: AsyncTPTest.test_fused_all_gather_scaled_matmul,https://github.com/pytorch/pytorch/issues/168526,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: rocm, module: symm_mem",scan when time allows,2025-12-04T21:19:45Z
issue,168525,Test: AsyncTPTest.test_multimem_all_gather_matmul,https://github.com/pytorch/pytorch/issues/168525,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: rocm, module: symm_mem",scan when time allows,2025-12-04T21:19:41Z
issue,168524,Test: AsyncTPTest.test_fused_all_gather_matmul_native,https://github.com/pytorch/pytorch/issues/168524,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: rocm, module: symm_mem",scan when time allows,2025-12-04T21:19:36Z
issue,168523,Test: AsyncTPTest.test_fused_all_gather_matmul,https://github.com/pytorch/pytorch/issues/168523,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: rocm, module: symm_mem",scan when time allows,2025-12-04T21:19:31Z
issue,168522,TestModule: test/distributed/test_symmetric_memory.py::AsyncTPTest,https://github.com/pytorch/pytorch/issues/168522,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: rocm, module: symm_mem",scan when time allows,2025-12-04T21:19:26Z
issue,168460,Test: ProcessGroupGlooTest.test_block_current_stream_cuda,https://github.com/pytorch/pytorch/issues/168460,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: c10d, module: rocm",scan when time allows,2025-12-04T21:17:52Z
issue,168459,Test: ProcessGroupGlooTest.test_reduce_stress_cuda,https://github.com/pytorch/pytorch/issues/168459,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: c10d, module: rocm",scan when time allows,2025-12-04T21:17:45Z
issue,168458,Test: ProcessGroupGlooTest.test_allgather_stress_cuda,https://github.com/pytorch/pytorch/issues/168458,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: c10d, module: rocm",scan when time allows,2025-12-04T21:17:38Z
issue,168457,Test: ProcessGroupGlooTest.test_gather_stress_cuda,https://github.com/pytorch/pytorch/issues/168457,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: c10d, module: rocm",scan when time allows,2025-12-04T21:17:29Z
issue,168456,Test: ProcessGroupGlooTest.test_scatter_stress_cuda,https://github.com/pytorch/pytorch/issues/168456,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: c10d, module: rocm",scan when time allows,2025-12-04T21:17:22Z
issue,168455,Test: ProcessGroupGlooTest.test_allreduce_stress_cuda,https://github.com/pytorch/pytorch/issues/168455,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: c10d, module: rocm",scan when time allows,2025-12-04T21:17:16Z
issue,168454,Test: ProcessGroupGlooTest.test_broadcast_stress_cuda,https://github.com/pytorch/pytorch/issues/168454,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: c10d, module: rocm",scan when time allows,2025-12-04T21:17:07Z
issue,168453,TestModule: test/distributed/test_c10d_gloo.py::ProcessGroupGlooTest,https://github.com/pytorch/pytorch/issues/168453,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: c10d, module: rocm",scan when time allows,2025-12-04T21:16:56Z
issue,168415,Test: TestShardedTensorChunked.test_multiple_local_shards,https://github.com/pytorch/pytorch/issues/168415,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ddp, module: rocm",scan when time allows,2025-12-04T21:16:35Z
issue,168414,Test: TestShardedTensorChunked.test_new_group,https://github.com/pytorch/pytorch/issues/168414,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ddp, module: rocm",scan when time allows,2025-12-04T21:16:28Z
issue,168413,Test: TestShardedTensorChunked.test_partial_world_size,https://github.com/pytorch/pytorch/issues/168413,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ddp, module: rocm",scan when time allows,2025-12-04T21:16:20Z
issue,168412,Test: TestShardedTensorChunked.test_complete_world_size,https://github.com/pytorch/pytorch/issues/168412,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ddp, module: rocm",scan when time allows,2025-12-04T21:16:10Z
issue,152308,bizarre behavior with torch module's Attribute Error,https://github.com/pytorch/pytorch/issues/152308,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2025-12-04T21:05:17Z
issue,168431,Test: TestShardedTensorFromLocalShards.test_init_from_local_shards_and_global_metadata,https://github.com/pytorch/pytorch/issues/168431,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ddp, module: rocm",scan when time allows,2025-12-04T20:54:03Z
issue,168430,Test: TestShardedTensorFromLocalShards.test_init_from_local_shards_and_global_metadata_with_local_view,https://github.com/pytorch/pytorch/issues/168430,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ddp, module: rocm",scan when time allows,2025-12-04T20:53:59Z
issue,168429,Test: TestShardedTensorFromLocalShards.test_init_from_local_shards_and_global_metadata_with_all_zeros,https://github.com/pytorch/pytorch/issues/168429,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ddp, module: rocm",scan when time allows,2025-12-04T20:53:54Z
issue,168428,Test: TestShardedTensorFromLocalShards.test_non_rw_sharded_recalc_for_metadata,https://github.com/pytorch/pytorch/issues/168428,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ddp, module: rocm",scan when time allows,2025-12-04T20:53:50Z
issue,168427,Test: TestShardedTensorFromLocalShards.test_init_from_local_shards_with_different_glb_size,https://github.com/pytorch/pytorch/issues/168427,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ddp, module: rocm",scan when time allows,2025-12-04T20:53:46Z
issue,168426,Test: TestShardedTensorFromLocalShards.test_recalc_for_metadata,https://github.com/pytorch/pytorch/issues/168426,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ddp, module: rocm",scan when time allows,2025-12-04T20:53:40Z
issue,168425,Test: TestShardedTensorFromLocalShards.test_init_from_local_shards,https://github.com/pytorch/pytorch/issues/168425,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ddp, module: rocm",scan when time allows,2025-12-04T20:53:35Z
issue,168423,Test: TestShardedTensorFromLocalTensor.test_init_from_local_tensor,https://github.com/pytorch/pytorch/issues/168423,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ddp, module: rocm",scan when time allows,2025-12-04T20:53:07Z
issue,168421,Test: TestShardedTensorEnumerable.test_with_rpc_names,https://github.com/pytorch/pytorch/issues/168421,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ddp, module: rocm",scan when time allows,2025-12-04T20:52:45Z
issue,168420,Test: TestShardedTensorEnumerable.test_multiple_local_shards,https://github.com/pytorch/pytorch/issues/168420,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ddp, module: rocm",scan when time allows,2025-12-04T20:52:40Z
issue,168419,Test: TestShardedTensorEnumerable.test_new_group,https://github.com/pytorch/pytorch/issues/168419,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ddp, module: rocm",scan when time allows,2025-12-04T20:52:36Z
issue,168418,Test: TestShardedTensorEnumerable.test_partial_world_size,https://github.com/pytorch/pytorch/issues/168418,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ddp, module: rocm",scan when time allows,2025-12-04T20:52:33Z
issue,168417,Test: TestShardedTensorEnumerable.test_grid_sharding,https://github.com/pytorch/pytorch/issues/168417,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ddp, module: rocm",scan when time allows,2025-12-04T20:52:29Z
issue,168411,TestModule: test/distributed/_shard/sharded_tensor/test_sharded_tensor.py::TestShardedTensorChunked,https://github.com/pytorch/pytorch/issues/168411,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ddp, module: rocm",scan when time allows,2025-12-04T20:49:30Z
issue,168424,TestModule: test/distributed/_shard/sharded_tensor/test_sharded_tensor.py::TestShardedTensorFromLocalShards,https://github.com/pytorch/pytorch/issues/168424,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ddp, module: rocm",scan when time allows,2025-12-04T20:47:32Z
issue,168422,TestModule: test/distributed/_shard/sharded_tensor/test_sharded_tensor.py::TestShardedTensorFromLocalTensor,https://github.com/pytorch/pytorch/issues/168422,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ddp, module: rocm",scan when time allows,2025-12-04T20:47:23Z
issue,168416,TestModule: test/distributed/_shard/sharded_tensor/test_sharded_tensor.py::TestShardedTensorEnumerable,https://github.com/pytorch/pytorch/issues/168416,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ddp, module: rocm",scan when time allows,2025-12-04T20:47:06Z
issue,161954,[RFC] CUDA code decoupling and directory refactoring,https://github.com/pytorch/pytorch/issues/161954,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: privateuse1,scan when time allows,2025-12-04T20:07:21Z
issue,169429,Error loading `torch\lib\torch_python.dll` due to DLL initialization failure on Windows from torch `2.9.0`,https://github.com/pytorch/pytorch/issues/169429,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: windows",scan when time allows,2025-12-04T08:02:21Z
issue,169119,"`torch.compile(backend=""cudagraphs"", fullgraph=True, cuda)` fails for sparse CSR → COO (`torch.compile does not support sparse Tensors`)",https://github.com/pytorch/pytorch/issues/169119,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: sparse, oncall: pt2",scan when time allows,2025-12-04T02:11:22Z
issue,169371,C++ Generator API is platform dependent,https://github.com/pytorch/pytorch/issues/169371,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: accelerator, module: cpp",scan when time allows,2025-12-04T02:07:26Z
issue,169390,[CI] [CUDA] SIGIOT error with test_cuda.py test_graph_capture_reclaim_4_streams,https://github.com/pytorch/pytorch/issues/169390,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: tests",scan when time allows,2025-12-04T00:18:12Z
issue,169080,Processing & Tiling Massive Images,https://github.com/pytorch/pytorch/issues/169080,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: data, module: dataloader",scan when time allows,2025-12-03T18:46:14Z
issue,166167,"Proposal: include 'extras packages' for supported versions of PyTorch-adjacent tools (e.g., numpy, onnx)",https://github.com/pytorch/pytorch/issues/166167,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2025-12-03T18:27:04Z
issue,169458,[download.pytorch.org] obselete dangling `xformers` index on CPU only branch,https://github.com/pytorch/pytorch/issues/169458,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: infra,scan when time allows,2025-12-03T16:01:47Z
issue,168624,Test: MinifierIsolateTests.test_after_aot_gpu_runtime_error,https://github.com/pytorch/pytorch/issues/168624,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, module: rocm, oncall: pt2",scan when time allows,2025-12-03T14:21:42Z
issue,169237,Large discrepancy between CPU and GPU norms for identical float32 tensor,https://github.com/pytorch/pytorch/issues/169237,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: norms and normalization",scan when time allows,2025-12-03T12:20:46Z
issue,169163,Silent failure of torch.stack,https://github.com/pytorch/pytorch/issues/169163,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: 64-bit, module: correctness (silent), module: cuda",scan when time allows,2025-12-03T12:20:22Z
issue,169266,Translation units with thrust::pair should include thrust/pair.h,https://github.com/pytorch/pytorch/issues/169266,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: sparse",scan when time allows,2025-12-03T12:19:52Z
issue,169378,[rocm triton 3.6] rocm-periodic-dynamo-benchmarks-test : convnextv2_nano.fcmae_ft_in22k_in1k fail_accuracy,https://github.com/pytorch/pytorch/issues/169378,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rocm,scan when time allows,2025-12-03T12:17:42Z
issue,169133,vLLM tlparse sometimes missing compilation_metrics,https://github.com/pytorch/pytorch/issues/169133,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: vllm,scan when time allows,2025-12-03T12:17:10Z
issue,132132,Tests stopped working after merging #126905 on Arm64,https://github.com/pytorch/pytorch/issues/132132,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: arm, module: ci, module: tests, oncall: pt2",scan when time allows,2025-12-03T01:31:21Z
issue,121967,[PT2.2] PyTorch dispacthing non core ATen ops in .compile mode for torch ops.,https://github.com/pytorch/pytorch/issues/121967,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: pt2-dispatcher, oncall: pt2",scan when time allows,2025-12-02T23:39:31Z
issue,159013,"[RFC] Support slicing of submesh and more flexible operations to one device mesh such as reshape/split, etc.",https://github.com/pytorch/pytorch/issues/159013,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: devicemesh, oncall: distributed",scan when time allows,2025-12-02T17:47:19Z
issue,138280,"Refactor FlexibleLayout to separate out ""this stride can be changed"" and ""how this buffer is allocated can be changed""",https://github.com/pytorch/pytorch/issues/138280,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",scan when time allows,2025-12-02T17:11:15Z
issue,151934,`torch.bmm` is slow on non-contiguous BF16 CPU tensors,https://github.com/pytorch/pytorch/issues/151934,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bfloat16, module: cpu, module: half, module: performance",scan when time allows,2025-12-02T07:30:13Z
issue,133629,RuntimeError: free_upper_bound + pytorch_used_bytes[device],https://github.com/pytorch/pytorch/issues/133629,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: cudacachingallocator",scan when time allows,2025-12-02T06:09:49Z
issue,168976,DebugMode doesn't track input/output relationships,https://github.com/pytorch/pytorch/issues/168976,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: pt2-dispatcher, oncall: pt2",scan when time allows,2025-12-02T04:10:43Z
issue,137767,[CPU]Tracking complex datatype issues,https://github.com/pytorch/pytorch/issues/137767,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: complex, module: cpu",scan when time allows,2025-12-02T01:40:00Z
issue,169213,CPU bad-free in `torch.nn.utils.rnn.pad_packed_sequence` / `unpack_sequence`,https://github.com/pytorch/pytorch/issues/169213,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rnn,scan when time allows,2025-12-02T01:32:55Z
issue,169318,[debug mode] disable debug mode temporarily,https://github.com/pytorch/pytorch/issues/169318,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2025-12-02T01:29:09Z
issue,169152,[debug mode] support pipeline parallelism,https://github.com/pytorch/pytorch/issues/169152,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: pt2-dispatcher, oncall: pt2",scan when time allows,2025-12-01T22:47:55Z
issue,169055,`[feature request][discussion][cuBLASLt]` fused backward support for `addmm` with epilogue fusions.,https://github.com/pytorch/pytorch/issues/169055,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: cuda",scan when time allows,2025-12-01T19:27:51Z
issue,109098,Can't initializa NVML,https://github.com/pytorch/pytorch/issues/109098,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2025-12-01T15:28:14Z
issue,35713,libtorch for Windows. MNIST example does no work.,https://github.com/pytorch/pytorch/issues/35713,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp,scan when time allows,2025-12-01T03:42:46Z
issue,151110,Update epsilon logic to improve numerical stability,https://github.com/pytorch/pytorch/issues/151110,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bc-breaking, module: numerical-stability",scan when time allows,2025-11-30T18:15:48Z
issue,139065,Cannot compile with latest LLVM-19,https://github.com/pytorch/pytorch/issues/139065,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cuda",scan when time allows,2025-11-30T13:46:11Z
issue,147317,Tensorboard `add_video()` broken for `moviepy>=2.0`,https://github.com/pytorch/pytorch/issues/147317,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tensorboard,scan when time allows,2025-11-29T03:08:34Z
issue,97395,Expanded weights tests broken,https://github.com/pytorch/pytorch/issues/97395,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: functorch,scan when time allows,2025-11-28T09:41:26Z
issue,152731,Inconsistent float16 overflow behavior between CPU and CUDA devices,https://github.com/pytorch/pytorch/issues/152731,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: edge cases, module: half",scan when time allows,2025-11-28T07:17:51Z
issue,168967,implement __str__ for Schedular classes,https://github.com/pytorch/pytorch/issues/168967,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2025-11-27T20:53:45Z
issue,169170,Optimizer __repr__ Performance & Stability Improvement,https://github.com/pytorch/pytorch/issues/169170,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2025-11-27T20:29:19Z
issue,169072,Properly fix destructive hardware interference warnings,https://github.com/pytorch/pytorch/issues/169072,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2025-11-27T20:11:45Z
issue,15821,No test coverage for kwargs of AvgPool2d and AvgPool3d,https://github.com/pytorch/pytorch/issues/15821,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, module: nn, module: pooling",scan when time allows,2025-11-27T17:48:47Z
issue,78656,Allow batch_norm_backward_elemt and batch_norm_gather_stats_with_counts handle 0 counts,https://github.com/pytorch/pytorch/issues/78656,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: norms and normalization,scan when time allows,2025-11-27T10:33:43Z
issue,166131,Return a view from tensor(requires_grad=False) in autograd function may cause incorrect requires_grad attribute.,https://github.com/pytorch/pytorch/issues/166131,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2025-11-27T09:06:44Z
issue,163400,"Add Flash Dynamic Mask Attention (FDMA) as an SDPA backend for sparse, extremely long-context attention",https://github.com/pytorch/pytorch/issues/163400,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: sdpa, module: sparse",scan when time allows,2025-11-27T05:06:00Z
issue,165612,RFC: Optionally accept NumPy dtypes in all APIs where torch dtypes are accepted,https://github.com/pytorch/pytorch/issues/165612,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: floatx (formerly float8), module: python frontend",scan when time allows,2025-11-27T00:42:21Z
issue,166378,Multihead Attention's in_proj_weight is initialized inconsistently,https://github.com/pytorch/pytorch/issues/166378,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: sdpa",scan when time allows,2025-11-26T22:59:57Z
issue,65151,Registering a global fallback for all operators that defaults us to assuming that autograd is not implemented,https://github.com/pytorch/pytorch/issues/65151,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: bc-breaking, module: custom-operators, module: molly-guard",scan when time allows,2025-11-26T22:32:04Z
issue,167721,"Minimal, comprehensive test suite",https://github.com/pytorch/pytorch/issues/167721,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: infra, module: testing",scan when time allows,2025-11-26T21:55:31Z
issue,49284,TensorBoard SummaryWriter to remote storage unreasonably slow,https://github.com/pytorch/pytorch/issues/49284,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tensorboard,scan when time allows,2025-11-26T18:29:36Z
issue,167656,Lack of tensor identity tracking in DebugMode makes it difficult to interpret results,https://github.com/pytorch/pytorch/issues/167656,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: oncall: distributed, oncall: pt2",scan when time allows,2025-11-26T15:03:13Z
issue,168403,Test: TestFullyShardOverlap.test_fully_shard_training_overlap,https://github.com/pytorch/pytorch/issues/168403,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: fsdp, module: rocm",scan when time allows,2025-11-25T19:58:08Z
issue,168405,Test: TestFullyShardNDTraining.test_2d_mlp_with_nd_mesh,https://github.com/pytorch/pytorch/issues/168405,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: fsdp, module: rocm",scan when time allows,2025-11-25T19:58:07Z
issue,168407,Test: TestFullyShard2DTraining.test_train_parity_2d_mlp,https://github.com/pytorch/pytorch/issues/168407,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: fsdp, module: rocm",scan when time allows,2025-11-25T19:58:03Z
issue,168476,Test: ComposabilityTest.test_pp_fsdp,https://github.com/pytorch/pytorch/issues/168476,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: fsdp, module: rocm",scan when time allows,2025-11-25T19:53:13Z
issue,168452,Test: TestComputeCommReorderingMultiProc.test_grouped_scheduler_node,https://github.com/pytorch/pytorch/issues/168452,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ddp, module: rocm",scan when time allows,2025-11-25T19:52:39Z
issue,168477,Test: ComposabilityTest.test_pp_fsdp_unshard_reshard_runtime,https://github.com/pytorch/pytorch/issues/168477,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: fsdp, module: rocm",scan when time allows,2025-11-25T19:51:55Z
issue,168537,Test: LoweringTest.test_lowering_one_shot_all_reduce,https://github.com/pytorch/pytorch/issues/168537,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: rocm, module: symm_mem",scan when time allows,2025-11-25T19:49:20Z
issue,168568,Test: AOTInductorTestsTemplate.test__weight_int4pack_mm_with_scales_and_zeros,https://github.com/pytorch/pytorch/issues/168568,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: aotinductor, module: rocm, oncall: export, oncall: pt2",scan when time allows,2025-11-25T19:46:09Z
issue,168567,Test: AOTInductorTestsTemplate.test__weight_int4pack_mm,https://github.com/pytorch/pytorch/issues/168567,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: aotinductor, module: rocm, oncall: export, oncall: pt2",scan when time allows,2025-11-25T19:46:08Z
issue,168563,TestModule: test/inductor/test_aot_inductor.py::AOTInductorTestsTemplate,https://github.com/pytorch/pytorch/issues/168563,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: aotinductor, module: rocm, oncall: export, oncall: pt2",scan when time allows,2025-11-25T19:46:03Z
issue,168623,TestModule: test/inductor/test_minifier_isolate.py::MinifierIsolateTests,https://github.com/pytorch/pytorch/issues/168623,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, module: rocm, oncall: pt2",scan when time allows,2025-11-25T19:43:20Z
issue,168646,Test: MultiKernelTest.test_triton_relu_fused_gemm,https://github.com/pytorch/pytorch/issues/168646,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, module: rocm, oncall: pt2",scan when time allows,2025-11-25T19:42:26Z
issue,168645,Test: MultiKernelTest.test_triton_gemm,https://github.com/pytorch/pytorch/issues/168645,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, module: rocm, oncall: pt2",scan when time allows,2025-11-25T19:42:25Z
issue,168644,TestModule: test/inductor/test_multi_kernel.py::MultiKernelTest,https://github.com/pytorch/pytorch/issues/168644,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, module: rocm, oncall: pt2",scan when time allows,2025-11-25T19:42:24Z
issue,168651,TestModule: test/inductor/test_select_algorithm.py::TestSelectAlgorithm,https://github.com/pytorch/pytorch/issues/168651,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, module: rocm, oncall: pt2",scan when time allows,2025-11-25T19:42:06Z
issue,168692,TestModule: test/nn/test_convolution.py::TestConvolutionNNDeviceType,https://github.com/pytorch/pytorch/issues/168692,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: rocm",scan when time allows,2025-11-25T19:37:37Z
issue,168706,TestModule: test/test_cpp_extensions_aot.py::TestCppExtensionAOT,https://github.com/pytorch/pytorch/issues/168706,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: aotinductor, module: rocm, oncall: export, oncall: pt2",scan when time allows,2025-11-25T19:37:06Z
issue,168707,Test: TestCppExtensionAOT.test_cublas_extension,https://github.com/pytorch/pytorch/issues/168707,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: aotinductor, module: rocm, oncall: export, oncall: pt2",scan when time allows,2025-11-25T19:37:00Z
issue,168708,Test: TestCppExtensionAOT.test_cusolver_extension,https://github.com/pytorch/pytorch/issues/168708,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: aotinductor, module: rocm, oncall: export, oncall: pt2",scan when time allows,2025-11-25T19:36:58Z
issue,168709,Test: TestCppExtensionAOT.test_cuda_dlink_libs,https://github.com/pytorch/pytorch/issues/168709,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: aotinductor, module: rocm, oncall: export, oncall: pt2",scan when time allows,2025-11-25T19:36:55Z
issue,168819,TestModule: test/test_scaled_matmul_cuda.py::TestFP8Matmul,https://github.com/pytorch/pytorch/issues/168819,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: floatx (formerly float8), module: linear algebra, module: rocm",scan when time allows,2025-11-25T19:34:06Z
issue,168756,TestModule: test/test_linalg.py::TestLinalg,https://github.com/pytorch/pytorch/issues/168756,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: linear algebra, module: rocm",scan when time allows,2025-11-25T19:31:24Z
issue,168766,Test: TestLinalg.test__dyn_quant_pack_4bit_weight,https://github.com/pytorch/pytorch/issues/168766,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: linear algebra, module: rocm",scan when time allows,2025-11-25T19:31:11Z
issue,168767,Test: TestLinalg.test__dyn_quant_matmul_4bit,https://github.com/pytorch/pytorch/issues/168767,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: linear algebra, module: rocm",scan when time allows,2025-11-25T19:31:10Z
issue,168768,Test: TestLinalg.test_compile_dyn_quant_matmul_4bit,https://github.com/pytorch/pytorch/issues/168768,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: linear algebra, module: rocm",scan when time allows,2025-11-25T19:31:06Z
issue,168771,Test: TestLinalg.test_ldl_solve,https://github.com/pytorch/pytorch/issues/168771,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: linear algebra, module: rocm",scan when time allows,2025-11-25T19:30:55Z
issue,168772,Test: TestLinalg.test_ck_blas_library,https://github.com/pytorch/pytorch/issues/168772,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: linear algebra, module: rocm",scan when time allows,2025-11-25T19:30:52Z
issue,168773,TestModule: test/test_matmul_cuda.py::TestMatmulCuda,https://github.com/pytorch/pytorch/issues/168773,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: linear algebra, module: rocm",scan when time allows,2025-11-25T19:30:44Z
issue,168777,Test: TestMatmulCuda.test_grouped_gemm_compiled,https://github.com/pytorch/pytorch/issues/168777,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: linear algebra, module: rocm",scan when time allows,2025-11-25T19:30:42Z
issue,168778,Test: TestMatmulCuda.test_mm_bmm_dtype_overload,https://github.com/pytorch/pytorch/issues/168778,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: linear algebra, module: rocm",scan when time allows,2025-11-25T19:30:39Z
issue,168776,Test: TestMatmulCuda.test_cublas_batch_invariance_blackwell,https://github.com/pytorch/pytorch/issues/168776,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: linear algebra, module: rocm",scan when time allows,2025-11-25T19:30:39Z
issue,168780,TestModule: test/test_matmul_cuda.py::TestMixedDtypesLinearCuda,https://github.com/pytorch/pytorch/issues/168780,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: linear algebra, module: rocm",scan when time allows,2025-11-25T19:30:22Z
issue,168781,Test: TestMixedDtypesLinearCuda.test_mixed_dtypes_linear,https://github.com/pytorch/pytorch/issues/168781,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: linear algebra, module: rocm",scan when time allows,2025-11-25T19:30:17Z
issue,168831,TestModule: test/test_sparse_semi_structured.py::SparseSemiStructuredTensorCompileTest,https://github.com/pytorch/pytorch/issues/168831,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: rocm, module: sparse",scan when time allows,2025-11-25T19:22:52Z
issue,168832,Test: SparseSemiStructuredTensorCompileTest.test_mlp_contiguous_relu_compile_cusparselt,https://github.com/pytorch/pytorch/issues/168832,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: rocm, module: sparse",scan when time allows,2025-11-25T19:22:51Z
issue,168833,Test: SparseSemiStructuredTensorCompileTest.test_mlp_contiguous_relu_compile_cutlass,https://github.com/pytorch/pytorch/issues/168833,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: rocm, module: sparse",scan when time allows,2025-11-25T19:22:50Z
issue,168834,Test: SparseSemiStructuredTensorCompileTest.test_sp24_compile,https://github.com/pytorch/pytorch/issues/168834,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: rocm, module: sparse",scan when time allows,2025-11-25T19:22:49Z
issue,168835,TestModule: test/test_sparse_semi_structured.py::TestSparseSemiStructuredTraining,https://github.com/pytorch/pytorch/issues/168835,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: rocm, module: sparse",scan when time allows,2025-11-25T19:22:45Z
issue,168836,Test: TestSparseSemiStructuredTraining.test_prune_dense_static_sort,https://github.com/pytorch/pytorch/issues/168836,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: rocm, module: sparse",scan when time allows,2025-11-25T19:22:36Z
issue,168837,Test: TestSparseSemiStructuredTraining.test_pruning_algo_largest_abs_values_greedy,https://github.com/pytorch/pytorch/issues/168837,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: rocm, module: sparse",scan when time allows,2025-11-25T19:22:35Z
issue,168838,Test: TestSparseSemiStructuredTraining.test_pack_both_ways_meta_correctness,https://github.com/pytorch/pytorch/issues/168838,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: rocm, module: sparse",scan when time allows,2025-11-25T19:22:27Z
issue,168840,Test: TestSparseSemiStructuredTraining.test_pack_both_ways_edge_case1,https://github.com/pytorch/pytorch/issues/168840,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: rocm, module: sparse",scan when time allows,2025-11-25T19:22:26Z
issue,168839,Test: TestSparseSemiStructuredTraining.test_pack_both_ways_id,https://github.com/pytorch/pytorch/issues/168839,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: rocm, module: sparse",scan when time allows,2025-11-25T19:22:26Z
issue,168842,Test: TestSparseSemiStructuredTraining.test_sp24_apply_dense,https://github.com/pytorch/pytorch/issues/168842,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: rocm, module: sparse",scan when time allows,2025-11-25T19:22:25Z
issue,168841,Test: TestSparseSemiStructuredTraining.test_sp24_apply,https://github.com/pytorch/pytorch/issues/168841,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: rocm, module: sparse",scan when time allows,2025-11-25T19:22:25Z
issue,168843,Test: TestSparseSemiStructuredTraining.test_sp24_matmuls,https://github.com/pytorch/pytorch/issues/168843,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: rocm, module: sparse",scan when time allows,2025-11-25T19:22:24Z
issue,168844,Test: TestSparseSemiStructuredTraining.test_sp24_matmuls_mat_vec,https://github.com/pytorch/pytorch/issues/168844,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: rocm, module: sparse",scan when time allows,2025-11-25T19:22:23Z
issue,168845,Test: TestSparseSemiStructuredTraining.test_sp24_matmuls_bmm,https://github.com/pytorch/pytorch/issues/168845,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: rocm, module: sparse",scan when time allows,2025-11-25T19:22:20Z
issue,168846,TestModule: test/test_sparse_semi_structured.py::TestSparseSemiStructuredCUTLASS,https://github.com/pytorch/pytorch/issues/168846,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: rocm, module: sparse",scan when time allows,2025-11-25T19:22:19Z
issue,168848,Test: TestSparseSemiStructuredCUTLASS.test_sparse_semi_structured_ops_cutlass,https://github.com/pytorch/pytorch/issues/168848,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: rocm, module: sparse",scan when time allows,2025-11-25T19:22:16Z
issue,168847,Test: TestSparseSemiStructuredCUTLASS.test_linear_cutlass,https://github.com/pytorch/pytorch/issues/168847,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: rocm, module: sparse",scan when time allows,2025-11-25T19:22:16Z
issue,168850,Test: TestSparseSemiStructuredCUTLASS.test_conversions_all_patterns,https://github.com/pytorch/pytorch/issues/168850,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: rocm, module: sparse",scan when time allows,2025-11-25T19:22:15Z
issue,168849,Test: TestSparseSemiStructuredCUTLASS.test_conversions,https://github.com/pytorch/pytorch/issues/168849,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: rocm, module: sparse",scan when time allows,2025-11-25T19:22:15Z
issue,168851,TestModule: test/test_sparse_semi_structured.py::TestSparseSemiStructuredCUSPARSELT,https://github.com/pytorch/pytorch/issues/168851,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: rocm, module: sparse",scan when time allows,2025-11-25T19:21:35Z
issue,168852,Test: TestSparseSemiStructuredCUSPARSELT.test_cslt_sparse_mm_alpha,https://github.com/pytorch/pytorch/issues/168852,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: rocm, module: sparse",scan when time allows,2025-11-25T19:21:32Z
issue,168853,Test: TestSparseSemiStructuredCUSPARSELT.test_cslt_sparse_mm_alpha_compile_autotune,https://github.com/pytorch/pytorch/issues/168853,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: rocm, module: sparse",scan when time allows,2025-11-25T19:21:31Z
issue,168854,Test: TestSparseSemiStructuredCUSPARSELT.test_cslt_sparse_mm_alpha_mixed_dtype,https://github.com/pytorch/pytorch/issues/168854,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: rocm, module: sparse",scan when time allows,2025-11-25T19:21:29Z
issue,168446,Test: RingAttentionTest.test_ring_attention_sdpa,https://github.com/pytorch/pytorch/issues/168446,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: rocm, module: sdpa",scan when time allows,2025-11-25T19:17:40Z
issue,168445,TestModule: test/distributed/tensor/test_attention.py::RingAttentionTest,https://github.com/pytorch/pytorch/issues/168445,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ddp, module: rocm, module: sdpa",scan when time allows,2025-11-25T19:17:37Z
issue,168788,Test: TestNestedTensorSubclass.test_sdpa_backwards,https://github.com/pytorch/pytorch/issues/168788,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: rocm, module: sdpa",scan when time allows,2025-11-25T19:15:05Z
issue,168789,Test: TestNestedTensorSubclass.test_sdpa_autocast,https://github.com/pytorch/pytorch/issues/168789,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: rocm, module: sdpa",scan when time allows,2025-11-25T19:15:04Z
issue,168876,Test: TestSDPACudaOnly.test_fused_kernels_nested_broadcasting,https://github.com/pytorch/pytorch/issues/168876,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: rocm, module: sdpa",scan when time allows,2025-11-25T19:13:02Z
issue,168875,Test: TestSDPACudaOnly.test_fused_backwards_throws_determinism_warning,https://github.com/pytorch/pytorch/issues/168875,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: rocm, module: sdpa",scan when time allows,2025-11-25T19:13:02Z
issue,168877,Test: TestSDPACudaOnly.test_fused_kernels_nested_broadcasting_query_dense,https://github.com/pytorch/pytorch/issues/168877,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: rocm, module: sdpa",scan when time allows,2025-11-25T19:13:01Z
issue,168481,Test: NCCLSymmetricMemoryTest.test_nccl_symmem_alloc,https://github.com/pytorch/pytorch/issues/168481,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nccl, module: rocm, module: symm_mem",scan when time allows,2025-11-25T18:52:28Z
issue,168533,Test: SymmMemNegativeTest.test_barrier_timeout,https://github.com/pytorch/pytorch/issues/168533,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: rocm, module: symm_mem",scan when time allows,2025-11-25T18:52:18Z
issue,168534,Test: SymmMemNegativeTest.test_put_signal_timeout,https://github.com/pytorch/pytorch/issues/168534,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: rocm, module: symm_mem",scan when time allows,2025-11-25T18:52:17Z
issue,168535,Test: SymmMemNegativeTest.test_wait_signal_timeout,https://github.com/pytorch/pytorch/issues/168535,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: rocm, module: symm_mem",scan when time allows,2025-11-25T18:52:15Z
issue,168480,TestModule: test/distributed/test_nccl.py::NCCLSymmetricMemoryTest,https://github.com/pytorch/pytorch/issues/168480,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nccl, module: rocm, module: symm_mem",scan when time allows,2025-11-25T18:45:58Z
issue,168447,TestModule: test/distributed/tensor/test_matrix_ops.py::DistMatrixOpsTest,https://github.com/pytorch/pytorch/issues/168447,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ddp, module: rocm",scan when time allows,2025-11-25T18:40:20Z
issue,168449,TestModule: test/distributed/tensor/test_tensor_ops.py::DistTensorOpsTest,https://github.com/pytorch/pytorch/issues/168449,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ddp, module: rocm",scan when time allows,2025-11-25T18:40:04Z
issue,168451,TestModule: test/distributed/test_aten_comm_compute_reordering.py::TestComputeCommReorderingMultiProc,https://github.com/pytorch/pytorch/issues/168451,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ddp, module: rocm",scan when time allows,2025-11-25T18:40:00Z
issue,168475,TestModule: test/distributed/test_composability.py::ComposabilityTest,https://github.com/pytorch/pytorch/issues/168475,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ddp, module: rocm",scan when time allows,2025-11-25T18:39:32Z
issue,168532,TestModule: test/distributed/test_symmetric_memory.py::SymmMemNegativeTest,https://github.com/pytorch/pytorch/issues/168532,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: rocm, module: symm_mem",scan when time allows,2025-11-25T18:36:21Z
issue,168536,TestModule: test/distributed/test_symmetric_memory.py::LoweringTest,https://github.com/pytorch/pytorch/issues/168536,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: rocm, module: symm_mem",scan when time allows,2025-11-25T18:36:17Z
issue,168503,TestModule: test/distributed/test_nvshmem_triton.py::NVSHMEMTritonTest,https://github.com/pytorch/pytorch/issues/168503,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: rocm, module: symm_mem",scan when time allows,2025-11-25T18:33:45Z
issue,168515,Test: NVSHMEMTritonTest.test_triton_alltoall,https://github.com/pytorch/pytorch/issues/168515,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: rocm, module: symm_mem",scan when time allows,2025-11-25T18:33:35Z
issue,168516,Test: NVSHMEMTritonTest.test_triton_broadcast,https://github.com/pytorch/pytorch/issues/168516,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: rocm, module: symm_mem",scan when time allows,2025-11-25T18:33:34Z
issue,168518,Test: NVSHMEMTritonTest.test_triton_minmax_reduce,https://github.com/pytorch/pytorch/issues/168518,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: rocm, module: symm_mem",scan when time allows,2025-11-25T18:33:33Z
issue,168517,Test: NVSHMEMTritonTest.test_triton_sum_reduce,https://github.com/pytorch/pytorch/issues/168517,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: rocm, module: symm_mem",scan when time allows,2025-11-25T18:33:33Z
issue,168519,Test: NVSHMEMTritonTest.test_triton_prod_reduce,https://github.com/pytorch/pytorch/issues/168519,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: rocm, module: symm_mem",scan when time allows,2025-11-25T18:33:26Z
issue,153187,Inconsistent overflow handling in `torch.clamp_min` between CPU and CUDA for float16 tensors,https://github.com/pytorch/pytorch/issues/153187,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: edge cases,scan when time allows,2025-11-25T18:24:43Z
issue,168435,Test: DistributedUtilTest.test_port_already_in_use_on_worker,https://github.com/pytorch/pytorch/issues/168435,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ddp, module: rocm",scan when time allows,2025-11-25T18:22:27Z
issue,168434,TestModule: test/distributed/elastic/utils/distributed_test.py::DistributedUtilTest,https://github.com/pytorch/pytorch/issues/168434,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ddp, module: rocm",scan when time allows,2025-11-25T18:22:18Z
issue,168406,TestModule: test/distributed/_composable/test_composability/test_2d_composability.py::TestFullyShard2DTraining,https://github.com/pytorch/pytorch/issues/168406,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: fsdp, module: rocm",scan when time allows,2025-11-25T18:20:16Z
issue,168404,TestModule: test/distributed/_composable/fsdp/test_fully_shard_training.py::TestFullyShardNDTraining,https://github.com/pytorch/pytorch/issues/168404,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: fsdp, module: rocm",scan when time allows,2025-11-25T01:03:31Z
issue,168402,TestModule: test/distributed/_composable/fsdp/test_fully_shard_overlap.py::TestFullyShardOverlap,https://github.com/pytorch/pytorch/issues/168402,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: fsdp, module: rocm",scan when time allows,2025-11-25T01:03:19Z
issue,168861,TestModule: test/test_torch.py::TestTorchDeviceType,https://github.com/pytorch/pytorch/issues/168861,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rocm,scan when time allows,2025-11-25T00:51:42Z
issue,168855,TestModule: test/test_testing.py::TestTesting,https://github.com/pytorch/pytorch/issues/168855,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rocm,scan when time allows,2025-11-25T00:51:34Z
issue,168806,Test: TestNNDeviceType.test_upsampling_64bit_indexing_channels_last,https://github.com/pytorch/pytorch/issues/168806,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rocm,scan when time allows,2025-11-25T00:50:29Z
issue,168801,TestModule: test/test_nn.py::TestNNDeviceType,https://github.com/pytorch/pytorch/issues/168801,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rocm,scan when time allows,2025-11-25T00:50:21Z
issue,168798,TestModule: test/test_nn.py::TestNN,https://github.com/pytorch/pytorch/issues/168798,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rocm,scan when time allows,2025-11-25T00:50:17Z
issue,168792,Test: TestNestedTensorSubclass.test_compile_preserves_metadata_cache,https://github.com/pytorch/pytorch/issues/168792,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rocm,scan when time allows,2025-11-25T00:50:08Z
issue,168791,Test: TestNestedTensorSubclass.test_dummy_mha_with_nt,https://github.com/pytorch/pytorch/issues/168791,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rocm,scan when time allows,2025-11-25T00:50:07Z
issue,168784,TestModule: test/test_nestedtensor.py::TestNestedTensorSubclass,https://github.com/pytorch/pytorch/issues/168784,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rocm,scan when time allows,2025-11-25T00:49:57Z
issue,168742,TestModule: test/test_cuda_primary_ctx.py::TestCudaPrimaryCtx,https://github.com/pytorch/pytorch/issues/168742,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rocm,scan when time allows,2025-11-25T00:48:58Z
issue,168721,Test: TestCudaMallocAsync.test_memory_profiler_viz,https://github.com/pytorch/pytorch/issues/168721,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rocm,scan when time allows,2025-11-25T00:48:29Z
issue,168720,TestModule: test/test_cuda.py::TestCudaMallocAsync,https://github.com/pytorch/pytorch/issues/168720,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rocm,scan when time allows,2025-11-25T00:48:28Z
issue,149016,DISABLED test_var_mean_tile_reduction_True_dynamic_shapes_cuda (__main__.DynamicShapesGPUTests),https://github.com/pytorch/pytorch/issues/149016,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: flaky-tests, module: inductor, module: rocm, oncall: pt2",scan when time allows,2025-11-25T00:38:52Z
issue,103495,DISABLED test_mem_get_info (__main__.TestCuda),https://github.com/pytorch/pytorch/issues/103495,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: flaky-tests, module: rocm",scan when time allows,2025-11-25T00:37:55Z
issue,167970,Thread Sanitizer support is clunky right now.,https://github.com/pytorch/pytorch/issues/167970,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: sanitizers",scan when time allows,2025-11-24T18:56:32Z
issue,149990,[Async TP] Fuse all-gather-matmuls for float8 rowwise training,https://github.com/pytorch/pytorch/issues/149990,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2025-11-24T18:35:21Z
issue,148748,export deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B failed,https://github.com/pytorch/pytorch/issues/148748,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: oncall: export, oncall: pt2",scan when time allows,2025-11-24T17:22:52Z
issue,166409,`torch.polygamma` produces inconsistent results between CPU and CUDA for moderate orders,https://github.com/pytorch/pytorch/issues/166409,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: nans and infs",scan when time allows,2025-11-24T15:37:25Z
issue,168304,Finish pytree flattening path in create_native_op_schema,https://github.com/pytorch/pytorch/issues/168304,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dtensor,scan when time allows,2025-11-24T13:40:21Z
issue,119968,"Error on trying to run detectron2 on Mac M1 with device set to ""mps"" (Metal Performance Shaders)",https://github.com/pytorch/pytorch/issues/119968,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: mps,scan when time allows,2025-11-24T13:12:38Z
issue,125718,`torch.compile` and complex numbers,https://github.com/pytorch/pytorch/issues/125718,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: complex, module: dynamo, oncall: pt2",scan when time allows,2025-11-24T09:00:00Z
issue,168158,A more efficient alternative to the Tensor.record_stream(...) api is needed,https://github.com/pytorch/pytorch/issues/168158,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2025-11-24T07:50:02Z
issue,162489,BatchNorm1d exhibits NaN outputs on GPU due to extreme input values causing numerical instability.,https://github.com/pytorch/pytorch/issues/162489,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: edge cases, module: nn",scan when time allows,2025-11-24T05:28:09Z
issue,166234,"!handles_.at(i) INTERNAL ASSERT FAILED at ""/pytorch/c10/cuda/CUDACachingAllocator.cpp"":430",https://github.com/pytorch/pytorch/issues/166234,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: windows, module: wsl",scan when time allows,2025-11-23T19:59:51Z
issue,168907,make lint fails on Mac,https://github.com/pytorch/pytorch/issues/168907,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: lint, module: macos",scan when time allows,2025-11-23T19:01:05Z
issue,150989,Add `pad_to_multiple_of` to `torch.nn.utils.rnn.pad_sequence`,https://github.com/pytorch/pytorch/issues/150989,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2025-11-22T17:20:41Z
issue,168253,nestedtensor inconsistency in `torch.masked_select`,https://github.com/pytorch/pytorch/issues/168253,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nestedtensor,scan when time allows,2025-11-21T17:47:04Z
issue,122030,[MPS] Wrong Calculation on MPS,https://github.com/pytorch/pytorch/issues/122030,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: mps",scan when time allows,2025-11-21T17:34:58Z
issue,166651,Make `_grouped_mm` public and add out_dtype,https://github.com/pytorch/pytorch/issues/166651,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: linear algebra",scan when time allows,2025-11-21T01:05:34Z
issue,167979,"Jupyter kernel restart during eval after first epoch (MNIST, CPU and CUDA, PyTorch 2.10.0.dev20251114+cu128 on Windows 11 + RTX 5080)",https://github.com/pytorch/pytorch/issues/167979,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: openmp, module: windows",scan when time allows,2025-11-20T18:17:04Z
issue,165999,[RFC] Improve the performance of `associative_scan()` with `cuda.compute`,https://github.com/pytorch/pytorch/issues/165999,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: higher order operators, module: pt2-dispatcher, oncall: pt2",scan when time allows,2025-11-20T14:37:30Z
issue,161302,Support for nn.SyncBatchNorm in torch.compile,https://github.com/pytorch/pytorch/issues/161302,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: distributed, oncall: pt2",scan when time allows,2025-11-19T23:55:45Z
issue,85604,"Multiple GPUs get ""errno: 98 - Address already in use""",https://github.com/pytorch/pytorch/issues/85604,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: c10d, oncall: distributed",scan when time allows,2025-11-19T23:04:35Z
issue,164493,[CI][CUDA][SymmMem] B200 Periodical Job Failure with test_triton_sum_reduce_int64,https://github.com/pytorch/pytorch/issues/164493,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ci, module: cuda",scan when time allows,2025-11-19T18:29:40Z
issue,167950,Insufficient documentation about the batching logic of `torch.linalg.solve`,https://github.com/pytorch/pytorch/issues/167950,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: linear algebra",scan when time allows,2025-11-19T16:44:07Z
issue,139124,shift operators not supporting integral types,https://github.com/pytorch/pytorch/issues/139124,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: unsigned int,scan when time allows,2025-11-19T06:52:30Z
issue,168128,"[proposal] Make docs webpage layout leaner, reduce number of menu/nav bars",https://github.com/pytorch/pytorch/issues/168128,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2025-11-19T04:54:20Z
issue,16706,Very poor Uniform() sampling near floating 0.0,https://github.com/pytorch/pytorch/issues/16706,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: random,scan when time allows,2025-11-19T00:11:12Z
issue,168010,Misleading error message for device mismatch in `conv2d`,https://github.com/pytorch/pytorch/issues/168010,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2025-11-18T21:58:44Z
issue,105448,Will nn.unfold support non-4D-tensor input in future version?,https://github.com/pytorch/pytorch/issues/105448,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2025-11-18T19:36:46Z
issue,167645,`torch.export` incompatible with `parametrize.cached()` - model 25x slower than expected,https://github.com/pytorch/pytorch/issues/167645,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: oncall: export, oncall: pt2",scan when time allows,2025-11-18T19:18:50Z
issue,167895,stable-diffusion 2 gone from huggingface,https://github.com/pytorch/pytorch/issues/167895,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: pt2,scan when time allows,2025-11-18T19:10:39Z
issue,167063,`torch.onnx.export` with `dynamo` breaks `FakeQuantize` exporting,https://github.com/pytorch/pytorch/issues/167063,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2025-11-18T14:43:06Z
issue,156792,backprop fails in HF SegformerForSemanticSegmentation on MPS backend with torch > 2.4,https://github.com/pytorch/pytorch/issues/156792,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: mps,scan when time allows,2025-11-17T19:20:02Z
issue,150517,Pinned memory doubles memory usage for tensors slightly over 128MB,https://github.com/pytorch/pytorch/issues/150517,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: memory usage",scan when time allows,2025-11-17T17:30:47Z
issue,145837,torch.nested.narrow() or torch.nested.to_padded_tensor() breaks backwards pass - invalid gradient,https://github.com/pytorch/pytorch/issues/145837,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nestedtensor,scan when time allows,2025-11-17T13:41:41Z
issue,153536,`torch.utils.data.default_collate` raises misleading warning for read-only NumPy arrays,https://github.com/pytorch/pytorch/issues/153536,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: data,scan when time allows,2025-11-17T04:44:10Z
issue,43259,RuntimeError: Expected to have finished reduction in the prior iteration before starting a new one.,https://github.com/pytorch/pytorch/issues/43259,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: data parallel, oncall: distributed",scan when time allows,2025-11-16T22:14:14Z
issue,162591,"SequentialLR scheduler, bug with last_epoch = N",https://github.com/pytorch/pytorch/issues/162591,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: lrscheduler, module: optimizer",scan when time allows,2025-11-16T17:46:25Z
issue,165906,"`torch.bmm` returns wrong results for large tensors with torch.set_float32_matmul_precision(""medium"")",https://github.com/pytorch/pytorch/issues/165906,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: numerical-stability,scan when time allows,2025-11-15T23:10:20Z
issue,167724,CUDA error: invalid configuration argument for tensor operation,https://github.com/pytorch/pytorch/issues/167724,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2025-11-15T03:56:00Z
issue,167765,[RFC] Extend MemPool support to SymmMem CUDA backend,https://github.com/pytorch/pytorch/issues/167765,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: symm_mem,scan when time allows,2025-11-15T03:55:04Z
issue,167754,Getting fallback when using scaled_dot_product_attention() despite CUDA kernels being installed and available,https://github.com/pytorch/pytorch/issues/167754,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sdpa,scan when time allows,2025-11-15T03:54:51Z
issue,167283,[RFC] Pytorch compute platform quality levels proposal,https://github.com/pytorch/pytorch/issues/167283,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: accelerator, module: backend",scan when time allows,2025-11-14T22:28:35Z
issue,122250,Proposal to make `torch.nn.modules.utils._pair` public,https://github.com/pytorch/pytorch/issues/122250,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2025-11-14T18:33:55Z
issue,146371,Torchrun does not handle worker failure gracefully,https://github.com/pytorch/pytorch/issues/146371,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2025-11-14T04:31:31Z
issue,167746,[xpu] Balance xpu ci test jobs workload,https://github.com/pytorch/pytorch/issues/167746,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ci, module: xpu",scan when time allows,2025-11-13T22:33:25Z
issue,152406,[DTensor] Calling .item() on DTensor with Partial placement results in local value,https://github.com/pytorch/pytorch/issues/152406,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2025-11-13T19:35:24Z
issue,124686,Unbacked SymInts: Should backwards graph with unbacked SymInts be recompiled with hints,https://github.com/pytorch/pytorch/issues/124686,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamic shapes, oncall: pt2",scan when time allows,2025-11-13T06:41:14Z
issue,167670,PyreFly results are dependent on local python version,https://github.com/pytorch/pytorch/issues/167670,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ci, module: lint",scan when time allows,2025-11-12T22:37:06Z
issue,117762,PyEval_AcquireThread can crash process if called from another thread at interpreter shutdown,https://github.com/pytorch/pytorch/issues/117762,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python frontend,scan when time allows,2025-11-12T16:26:29Z
issue,162592,Multi-Batching in F.scaled_dot_product_attention does not work.,https://github.com/pytorch/pytorch/issues/162592,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sdpa,scan when time allows,2025-11-12T10:36:10Z
issue,117949,aten::slow_conv3d_forward still missing from MPS,https://github.com/pytorch/pytorch/issues/117949,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: mps,scan when time allows,2025-11-12T06:57:07Z
issue,167558,per_page=1000 doesn't work in hud.pytorch.org,https://github.com/pytorch/pytorch/issues/167558,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: devx,scan when time allows,2025-11-11T19:30:25Z
issue,102334,There is a memory leak in torch.load,https://github.com/pytorch/pytorch/issues/102334,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: memory usage, module: serialization",scan when time allows,2025-11-11T18:52:27Z
issue,167349,AOTAutograd cache bypasses on DTensor.to_local(),https://github.com/pytorch/pytorch/issues/167349,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: oncall: distributed, oncall: pt2",scan when time allows,2025-11-11T18:30:10Z
issue,167027,combine compiled vectorized function without recompiling already compiled part,https://github.com/pytorch/pytorch/issues/167027,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",scan when time allows,2025-11-11T18:15:06Z
issue,167254,Fine control of fp32 precison in backward phase,https://github.com/pytorch/pytorch/issues/167254,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2025-11-11T18:00:37Z
issue,166646,Allow kwargs in torch.utils.checkpoint.checkpoint().,https://github.com/pytorch/pytorch/issues/166646,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: activation checkpointing,scan when time allows,2025-11-11T15:27:56Z
issue,156701,Add generator argument to torch.rand_like,https://github.com/pytorch/pytorch/issues/156701,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python frontend,scan when time allows,2025-11-11T09:07:50Z
issue,167520,improve selective decomposition tracing time,https://github.com/pytorch/pytorch/issues/167520,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",scan when time allows,2025-11-11T06:34:47Z
issue,167420,conv3d with dilation and bias = severe performance degradation in bf16/fp16,https://github.com/pytorch/pytorch/issues/167420,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: cudnn, module: performance",scan when time allows,2025-11-10T23:34:05Z
issue,109774,[DDP + Dynamo] Tracing DDP AllReduce (Compiled DDP),https://github.com/pytorch/pytorch/issues/109774,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ddp, oncall: distributed, oncall: pt2",scan when time allows,2025-11-10T21:36:37Z
issue,150799,FSDP in hybrid mode throws _saved_grad_shard error when backward is called on cross-rank all-gathered loss,https://github.com/pytorch/pytorch/issues/150799,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: fsdp, oncall: distributed",scan when time allows,2025-11-10T18:28:28Z
issue,167409,FFT operations hang indefinitely on RTX 5000 Ada (driver 570.86.15),https://github.com/pytorch/pytorch/issues/167409,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: fft",scan when time allows,2025-11-10T18:05:33Z
issue,107800,[feature request] [discussion] Include basic `ctypes` bindings for `cudart`/`cublasLt`/`cublas`/`nvrtc`/`cudnn` with stock PyTorch,https://github.com/pytorch/pytorch/issues/107800,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cublas, module: cuda",scan when time allows,2025-11-10T16:56:43Z
issue,157015,[WIP][RFC] Compilable flex_attention + Context Parallel,https://github.com/pytorch/pytorch/issues/157015,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: flex attention, module: higher order operators, module: pt2-dispatcher, oncall: pt2",scan when time allows,2025-11-10T01:19:21Z
issue,108948,Add Lambert W function as torch.special.lambertw,https://github.com/pytorch/pytorch/issues/108948,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: special,scan when time allows,2025-11-09T09:29:53Z
issue,147613,torch.nn.AvgPool2d fails with stride >= 2^31 on CUDA,https://github.com/pytorch/pytorch/issues/147613,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: 64-bit, module: cuda, module: nn, module: pooling",scan when time allows,2025-11-08T22:53:11Z
issue,163425,`torch._grouped_gemm` has poor performance for large batch sizes with dtype BF16,https://github.com/pytorch/pytorch/issues/163425,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bfloat16, module: cuda, module: linear algebra, module: performance",scan when time allows,2025-11-08T20:14:42Z
issue,108175,Enable FlashAttentionV2 on Windows,https://github.com/pytorch/pytorch/issues/108175,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sdpa,scan when time allows,2025-11-08T17:38:08Z
issue,117321,[C++-polyfill] Pt2 - Discussion around user defined type->behavior dispatching,https://github.com/pytorch/pytorch/issues/117321,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",scan when time allows,2025-11-07T21:32:22Z
issue,111674,Dynamo Compile samples should record file/line that raised exception,https://github.com/pytorch/pytorch/issues/111674,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",scan when time allows,2025-11-07T21:30:57Z
issue,102063,[inlined-inbuilt-nn-modules][dynamo][BE] Revisit call_method of NNModuleVariable,https://github.com/pytorch/pytorch/issues/102063,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",scan when time allows,2025-11-07T21:21:21Z
issue,167351,move to gcc11 for CUDA 13 CI tests,https://github.com/pytorch/pytorch/issues/167351,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ci, module: cuda",scan when time allows,2025-11-07T21:08:38Z
issue,118966,"[hierarchical compilation] A way to designate a portion of torch.compile as a noinline block that is compiled/guarded separately, but less disruptive than a graph break (e.g., for loops)",https://github.com/pytorch/pytorch/issues/118966,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: compile-time, module: dynamo, oncall: pt2",scan when time allows,2025-11-07T15:43:19Z
issue,65161,error: ‘_mm512_set_epi8’ was not declared in this scope when build from source on AMD 3600 CPU,https://github.com/pytorch/pytorch/issues/65161,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: vectorization",scan when time allows,2025-11-07T15:29:47Z
issue,106608,ROCm & Windows Support,https://github.com/pytorch/pytorch/issues/106608,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rocm,scan when time allows,2025-11-07T03:45:44Z
issue,167229,"The quality of button ""Ask AI"" is very low. I propose removing it. The button gets too much area of webpage, made it inconvenient.",https://github.com/pytorch/pytorch/issues/167229,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2025-11-07T02:16:14Z
issue,167257,Support cuda event input/output in graph with streams,https://github.com/pytorch/pytorch/issues/167257,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",scan when time allows,2025-11-06T22:16:35Z
issue,159226,Add support sdpa kernels on gfx1101 (AMD RX7800XT or W7700),https://github.com/pytorch/pytorch/issues/159226,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rocm,scan when time allows,2025-11-06T22:13:46Z
issue,126154,Optimizer classes not `dill` picklable after using `torch.compile`,https://github.com/pytorch/pytorch/issues/126154,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: serialization, oncall: pt2",scan when time allows,2025-11-06T21:57:54Z
issue,118884,[Export] [TorchVision] Export Support for TorchVision Detection Models,https://github.com/pytorch/pytorch/issues/118884,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: export, oncall: pt2",scan when time allows,2025-11-06T21:40:46Z
issue,106130,torch._subclasses.fake_tensor.DynamicOutputShapeException: aten.nonzero.default,https://github.com/pytorch/pytorch/issues/106130,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: faketensor,scan when time allows,2025-11-06T15:09:57Z
issue,130282,[custom_ops] We should be able to mark an operator as non-differentiable.,https://github.com/pytorch/pytorch/issues/130282,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: custom-operators, module: pt2-dispatcher, oncall: pt2",scan when time allows,2025-11-06T14:37:08Z
issue,133268,Adafactor compile support tracker,https://github.com/pytorch/pytorch/issues/133268,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: optimizer, oncall: pt2",scan when time allows,2025-11-06T05:12:14Z
issue,126566,`torch.compiler.allow_in_graph` does not create a `call_module` op in fx.Graph in torch 2.3.0,https://github.com/pytorch/pytorch/issues/126566,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",scan when time allows,2025-11-06T04:45:53Z
issue,137128,Context manager to skip checks inside decompositions,https://github.com/pytorch/pytorch/issues/137128,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: decompositions, module: dynamic shapes, module: dynamo, oncall: pt2",scan when time allows,2025-11-06T02:47:19Z
issue,119128,[dynamo] Refactor switch statements to improve compile times,https://github.com/pytorch/pytorch/issues/119128,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: compile-time, module: dynamo, oncall: pt2",scan when time allows,2025-11-06T01:40:33Z
issue,97155,Custom recurrent network takes very long to compile for long sequences,https://github.com/pytorch/pytorch/issues/97155,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: compile-time, module: dynamo, module: python dispatcher, oncall: pt2",scan when time allows,2025-11-06T00:40:50Z
issue,167119,Pyrefly may regress type coverage - consider reenabling mypy,https://github.com/pytorch/pytorch/issues/167119,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: typing,scan when time allows,2025-11-05T21:18:55Z
issue,138215,Dynamo x autograd.Function: leverage compiled autograd on graph break in backwards,https://github.com/pytorch/pytorch/issues/138215,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: compiled autograd, oncall: pt2",scan when time allows,2025-11-05T21:08:03Z
issue,126846,propagate_real_tensors can cause deadlocks for programs with collectives,https://github.com/pytorch/pytorch/issues/126846,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamic shapes, oncall: pt2",scan when time allows,2025-11-05T21:07:07Z
issue,118758,Selectively disable dynamic shapes in an inner compile region,https://github.com/pytorch/pytorch/issues/118758,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamic shapes, module: dynamo, oncall: pt2",scan when time allows,2025-11-05T21:05:01Z
issue,58734,"Support for `uint16`, `uint32`, and `uint64`",https://github.com/pytorch/pytorch/issues/58734,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: python array api, oncall: pt2",scan when time allows,2025-11-05T19:48:10Z
issue,109074,torch.compile/triton holding GIL during compilation and CompiledKernel call results in deadlocks during distributed training,https://github.com/pytorch/pytorch/issues/109074,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: compile-time, module: inductor, oncall: pt2",scan when time allows,2025-11-05T19:20:37Z
issue,103891,Docker images: faster linker for `torch.compile`,https://github.com/pytorch/pytorch/issues/103891,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docker, oncall: pt2",scan when time allows,2025-11-05T19:14:20Z
issue,118573,[Inductor] Unite all possible way to track mutation,https://github.com/pytorch/pytorch/issues/118573,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",scan when time allows,2025-11-05T19:02:22Z
issue,124427,"Find a common home for decompositions, perhaps outside of the obliquely named _refs directory",https://github.com/pytorch/pytorch/issues/124427,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: pt2,scan when time allows,2025-11-05T19:01:18Z
issue,93923,Dynamo uses CONSTANT_MATCH guards for string inputs,https://github.com/pytorch/pytorch/issues/93923,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",scan when time allows,2025-11-05T17:54:33Z
issue,124245,[RFC] Add new CPP builder for inductor on pytorch Windows,https://github.com/pytorch/pytorch/issues/124245,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, module: windows, oncall: cpu inductor, oncall: pt2",scan when time allows,2025-11-05T17:37:58Z
issue,163990,Add CI job to track typing changes made to PyTorch,https://github.com/pytorch/pytorch/issues/163990,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: infra, module: typing",scan when time allows,2025-11-05T15:45:27Z
issue,149301,Unexpected results w/ LayerNorm -- suspecting possible memory issue?,https://github.com/pytorch/pytorch/issues/149301,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: 64-bit, module: cuda, module: norms and normalization, module: numerical-stability",scan when time allows,2025-11-05T14:02:34Z
issue,135860,Lobpcg for complex gives wrong results.,https://github.com/pytorch/pytorch/issues/135860,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: complex, module: correctness (silent), module: linear algebra",scan when time allows,2025-11-05T02:23:17Z
issue,166611,CuTe DSL NVFP4 GEMM/Grouped GEMM kernels,https://github.com/pytorch/pytorch/issues/166611,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: linear algebra",scan when time allows,2025-11-04T19:25:35Z
issue,166511,RuntimeError when finetuning Gemma 3 using Unsloth on Databricks,https://github.com/pytorch/pytorch/issues/166511,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: functorch, module: pt2-dispatcher, oncall: pt2",scan when time allows,2025-11-04T09:42:40Z
issue,166696,`torch.accelerator.is_available()` fails when privateuseone is used,https://github.com/pytorch/pytorch/issues/166696,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: accelerator,scan when time allows,2025-11-04T07:54:51Z
issue,160420,CachingHostAllocator does not take into account NUMA affinity of GPU's when reusing freed blocks,https://github.com/pytorch/pytorch/issues/160420,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cudacachingallocator,scan when time allows,2025-11-04T07:03:52Z
issue,166823,Change tab order of docs to have main content before left sidebar,https://github.com/pytorch/pytorch/issues/166823,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2025-11-04T01:39:41Z
issue,153863,"cmake: add USE_SYSTEM_{KLEIDI,CUDNN_FRONTEND,CUTLASS,FMT} options to USE_SYSTEM_LIBS",https://github.com/pytorch/pytorch/issues/153863,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2025-11-03T14:00:56Z
issue,165730,"autocast('cuda', enabled=False) silently ignored on NPU without warning, causing precision issues",https://github.com/pytorch/pytorch/issues/165730,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision), module: error checking, module: privateuse1",scan when time allows,2025-11-03T03:08:33Z
issue,164707,Significant difference in performance between CUSPARSELT and CUTLASS when doing SparseSemiStructuredTensor matmul,https://github.com/pytorch/pytorch/issues/164707,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2025-11-02T06:37:18Z
issue,166733,benchmark and advertise torch.compile's ability to reduce triton kernel launch overhead,https://github.com/pytorch/pytorch/issues/166733,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: vllm, oncall: pt2",scan when time allows,2025-11-01T03:33:05Z
issue,164891,Better API or GuardRails for Distributed Optimizations with non-SPMD graphs,https://github.com/pytorch/pytorch/issues/164891,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dtensor, module: dynamo, oncall: pt2",scan when time allows,2025-11-01T02:31:01Z
issue,166401,torch.masked.argmax returns different results on CPU vs CUDA when multiple maximum values exist,https://github.com/pytorch/pytorch/issues/166401,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: masked operators,scan when time allows,2025-11-01T00:53:45Z
issue,152595,"ROCm, 7900 XTX (gfx1100): Pytorch FLASH_ATTENTION SDPA is 2.5x slower than MATH (fp16, head_dim 256, seqlen 4360, 12 heads)",https://github.com/pytorch/pytorch/issues/152595,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: performance, module: rocm, module: sdpa",scan when time allows,2025-10-31T23:00:32Z
issue,164860,Driver update on H100 and A100 instances,https://github.com/pytorch/pytorch/issues/164860,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2025-10-31T17:55:33Z
issue,166686,"pytorch can not load libtorch trained pt pth model ,",https://github.com/pytorch/pytorch/issues/166686,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: serialization,scan when time allows,2025-10-31T17:27:31Z
issue,166557,Two conflicting definitions of `torch.backends.cudnn.rnn` since #125888,https://github.com/pytorch/pytorch/issues/166557,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cudnn,scan when time allows,2025-10-31T15:39:09Z
issue,123505,RuntimeError: Unrecognized CachingAllocator option: expandable_segments,https://github.com/pytorch/pytorch/issues/123505,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2025-10-30T11:18:58Z
issue,147449,gradient checkpointing with use_reentrant=False cannot reduce peak memory,https://github.com/pytorch/pytorch/issues/147449,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: activation checkpointing,scan when time allows,2025-10-30T05:31:36Z
issue,164813,flex_attention not functional in NVIDIA RTX 2080Ti,https://github.com/pytorch/pytorch/issues/164813,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",scan when time allows,2025-10-30T02:34:12Z
issue,165409,[torch.compile][fullgraph][functionalization] In-place op on overlapped view from unfold fails: “tensor being mutated has internal overlap” (eager OK),https://github.com/pytorch/pytorch/issues/165409,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: functionalization, module: pt2-dispatcher, oncall: pt2",scan when time allows,2025-10-30T02:17:03Z
issue,163785,Revisit guarding on unbacked inputs !,https://github.com/pytorch/pytorch/issues/163785,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamic shapes, oncall: pt2",scan when time allows,2025-10-29T22:58:35Z
issue,148681,foreach CUDA tests flaky on CUDA 12.6+ due to flaky profiler results,https://github.com/pytorch/pytorch/issues/148681,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2025-10-29T17:12:53Z
issue,129843,Concatenating CSR matrices fails,https://github.com/pytorch/pytorch/issues/129843,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2025-10-29T14:24:34Z
issue,160248,"PyTorch 2.8: gloo-cuda linking issue, undefined symbol",https://github.com/pytorch/pytorch/issues/160248,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cuda",scan when time allows,2025-10-29T06:35:22Z
issue,162790,torch.tensor and torch.as_tensor flatten (concatenates) list of 1d tensors instead of nesting them,https://github.com/pytorch/pytorch/issues/162790,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python frontend,scan when time allows,2025-10-29T04:41:11Z
issue,166116,[CCA] CUDACachingAllocator always release physical memory handle when the expandable segment unmaps.,https://github.com/pytorch/pytorch/issues/166116,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cudacachingallocator,scan when time allows,2025-10-29T02:57:00Z
issue,154074,UNSTABLE Build manywheel docker images for s390x / build-docker-cpu-s390x,https://github.com/pytorch/pytorch/issues/154074,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2025-10-29T02:37:31Z
issue,165231,_scaled_mm errors with row-major rhs matrix,https://github.com/pytorch/pytorch/issues/165231,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: floatx (formerly float8), module: linear algebra",scan when time allows,2025-10-28T18:50:40Z
issue,144564,Tabulate not official dependency of PyTorch but needed by features like FlopCounterMode,https://github.com/pytorch/pytorch/issues/144564,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: flop counter,scan when time allows,2025-10-28T12:53:08Z
issue,153996,UR Error when calling grid_sample,https://github.com/pytorch/pytorch/issues/153996,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: windows, module: xpu",scan when time allows,2025-10-28T12:42:45Z
issue,166186,We should have an env var to turn off all dynamic shapes,https://github.com/pytorch/pytorch/issues/166186,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamic shapes, module: dynamo, oncall: pt2",scan when time allows,2025-10-28T03:54:22Z
issue,166233,"license: Is it possible to stop using Conda in the Dockerfile? Due to Conda’s licensing issues, many companies have already received legal warning letters.",https://github.com/pytorch/pytorch/issues/166233,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: docker",scan when time allows,2025-10-28T03:42:37Z
issue,166243,[BugReport] Exponential distribution may generate zero values on CPU backend,https://github.com/pytorch/pytorch/issues/166243,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distributions,scan when time allows,2025-10-28T00:54:41Z
issue,156434,Suggest to use the torch cmake target instead of ${TORCH_LIBRARIES} in the c++ docs,https://github.com/pytorch/pytorch/issues/156434,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2025-10-27T17:42:34Z
issue,29814,SGD fails on sparse matrix,https://github.com/pytorch/pytorch/issues/29814,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: optimizer, module: sparse",scan when time allows,2025-10-27T17:34:16Z
issue,165811,[RFC] A Python backend registration API,https://github.com/pytorch/pytorch/issues/165811,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: backend, module: python frontend",scan when time allows,2025-10-27T17:28:37Z
issue,165741,Ctx or at least Thread level switch for fp32 IEEE precision tolerance,https://github.com/pytorch/pytorch/issues/165741,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: docs, module: multithreading",scan when time allows,2025-10-27T17:23:56Z
issue,64932,[RFC] TorchStore - A Shared-Memory Tensor Store,https://github.com/pytorch/pytorch/issues/64932,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2025-10-27T16:56:02Z
issue,165324,How to enable Bfloat16 when using torch.func.jvp,https://github.com/pytorch/pytorch/issues/165324,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision),scan when time allows,2025-10-27T15:23:35Z
issue,166080,Feature Request: masked convolution,https://github.com/pytorch/pytorch/issues/166080,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2025-10-27T15:08:49Z
issue,152285,Error after successful build: No module named 'torch._C._distributed_c10d',https://github.com/pytorch/pytorch/issues/152285,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, oncall: distributed",scan when time allows,2025-10-27T09:03:18Z
issue,160849,Add new shuffle mechanism for SGD/Adam --- CorgiPile without full data shuffle,https://github.com/pytorch/pytorch/issues/160849,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: data, module: dataloader",scan when time allows,2025-10-27T03:19:59Z
issue,123465,lintrunner-noclang job may diverge from `lintrunner` locally,https://github.com/pytorch/pytorch/issues/123465,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: lint, module: typing",scan when time allows,2025-10-27T03:04:02Z
issue,127271,ImportError: cannot import name 'triton_key' from 'triton.compiler.compiler' when using triton 2.3.x,https://github.com/pytorch/pytorch/issues/127271,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",scan when time allows,2025-10-27T02:44:20Z
issue,165626,Code Style Cleanup: Replace python3 -mpip with the more readable python3 -m pip,https://github.com/pytorch/pytorch/issues/165626,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2025-10-26T11:21:15Z
issue,157218,[Feature] Taylor expansion pruning,https://github.com/pytorch/pytorch/issues/157218,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: pruning,scan when time allows,2025-10-26T08:05:36Z
issue,166219,Why are there so many warnings when building the C++ libtorch project? How to resolve it?,https://github.com/pytorch/pytorch/issues/166219,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp-extensions, module: windows",scan when time allows,2025-10-25T15:39:20Z
issue,73359,"Pytorch Typing, for Tensor type annotations",https://github.com/pytorch/pytorch/issues/73359,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: typing,scan when time allows,2025-10-25T05:41:04Z
issue,139494,Mutating custom ops slower than non-mutating custom ops.,https://github.com/pytorch/pytorch/issues/139494,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: custom-operators, module: pt2-dispatcher, module: vllm, oncall: pt2",scan when time allows,2025-10-24T17:30:15Z
issue,133689,[torchbind x PT2] Handle vllm's ScalarType,https://github.com/pytorch/pytorch/issues/133689,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: torchbind, module: vllm",scan when time allows,2025-10-24T16:38:55Z
issue,150488,Lightweight CUDAGraph backend,https://github.com/pytorch/pytorch/issues/150488,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda graphs, module: inductor, module: vllm, oncall: pt2",scan when time allows,2025-10-24T16:38:40Z
issue,150484,API to specify cudagraph sizes,https://github.com/pytorch/pytorch/issues/150484,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda graphs, module: inductor, module: vllm, oncall: pt2",scan when time allows,2025-10-24T16:38:40Z
issue,150482,API to specify static input indices for cudagraph,https://github.com/pytorch/pytorch/issues/150482,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda graphs, module: inductor, module: vllm, oncall: pt2",scan when time allows,2025-10-24T16:38:39Z
issue,150190,"auto_functionalize HOPs are cacheable, also we should check that we allow for manual overrides for cacheability",https://github.com/pytorch/pytorch/issues/150190,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: vllm,scan when time allows,2025-10-24T16:38:38Z
issue,148983,inference_mode Tensors do not always need to be guarded on,https://github.com/pytorch/pytorch/issues/148983,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: vllm,scan when time allows,2025-10-24T16:38:36Z
issue,163428,[vLLM] Crash during inferece in torch dynamo on 2xRTX2080Ti,https://github.com/pytorch/pytorch/issues/163428,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: flex attention, module: higher order operators, module: pt2-dispatcher, module: vllm",scan when time allows,2025-10-24T16:38:07Z
issue,162402,Issue with torch.compile - sm110a(Nvidia Jetson Thor),https://github.com/pytorch/pytorch/issues/162402,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: vllm, oncall: pt2",scan when time allows,2025-10-24T16:38:06Z
issue,158699,reports of torch.compile'd rmsnorm numerics different from eager,https://github.com/pytorch/pytorch/issues/158699,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, module: vllm, oncall: pt2",scan when time allows,2025-10-24T16:38:00Z
issue,161500,nn.TransformerEncoder NaNs with float mask and float src_key_padding_mask only when in both eval() and torch.no_grad(),https://github.com/pytorch/pytorch/issues/161500,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nans and infs, module: nn",scan when time allows,2025-10-24T09:51:45Z
issue,166101,Compiling libtorch requires GLIBC_2.35,https://github.com/pytorch/pytorch/issues/166101,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: abi, module: rocm",scan when time allows,2025-10-24T05:27:17Z
issue,139349,torch.nansum does not work with complex numbers on CPU,https://github.com/pytorch/pytorch/issues/139349,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: complex, module: nans and infs",scan when time allows,2025-10-23T18:54:10Z
issue,165982,LELU Activation Function: Proposal for PyTorch,https://github.com/pytorch/pytorch/issues/165982,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2025-10-23T12:33:36Z
issue,101666,Unexpected behavior of fmod op in some float32 input,https://github.com/pytorch/pytorch/issues/101666,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: numerical-stability,scan when time allows,2025-10-23T06:37:14Z
issue,165992,AttributeError: module 'triton.language.math' has no attribute 'tanh',https://github.com/pytorch/pytorch/issues/165992,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2025-10-23T01:33:28Z
issue,166054,Add XLA to `at::accelerator::getAccelerator` and `at::accelerator::isAccelerator`,https://github.com/pytorch/pytorch/issues/166054,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: accelerator,scan when time allows,2025-10-22T23:47:56Z
issue,165912,Runtime assert on unbacked input inside torch.cond improperly gets lifted to top level,https://github.com/pytorch/pytorch/issues/165912,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dynamic shapes,scan when time allows,2025-10-22T23:44:19Z
issue,85791,nn.CrossEntropyLoss overflow with FP16 and minibatch,https://github.com/pytorch/pytorch/issues/85791,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: edge cases, module: loss, module: nn",scan when time allows,2025-10-22T20:20:36Z
issue,166064,Add submodul for `six`,https://github.com/pytorch/pytorch/issues/166064,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: third_party",scan when time allows,2025-10-22T16:16:30Z
issue,166072,UNSTABLE trunk / linux-jammy-py3-clang12-executorch / test (executorch),https://github.com/pytorch/pytorch/issues/166072,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2025-10-22T16:15:20Z
issue,105912,Batching rule for aten::bincount.,https://github.com/pytorch/pytorch/issues/105912,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: functorch,scan when time allows,2025-10-22T11:28:36Z
issue,128785,"Any plans for a ""torch.minmax"" (min-max normalization) function?",https://github.com/pytorch/pytorch/issues/128785,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python frontend,scan when time allows,2025-10-22T08:02:04Z
issue,166011,cusparse deprecation warning in CUDA 13.0 inductor build,https://github.com/pytorch/pytorch/issues/166011,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ci, module: sparse",scan when time allows,2025-10-21T19:28:25Z
issue,166014,Make Inductor Fallback Nodes Less Reliant on Invariants from Functionalization / AOT Autograd,https://github.com/pytorch/pytorch/issues/166014,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",scan when time allows,2025-10-21T18:59:31Z
issue,165976,Internal Assert Failure with Specific Large Padding Value in torch.nn.Conv3d,https://github.com/pytorch/pytorch/issues/165976,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: error checking",scan when time allows,2025-10-21T14:27:55Z
issue,165826,“Request: Add CUDA support for NVIDIA RTX 5070 Ti (Ada Lovelace Refresh)”,https://github.com/pytorch/pytorch/issues/165826,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2025-10-20T15:30:01Z
issue,165880,torch.vmap doesn't work on most of the models,https://github.com/pytorch/pytorch/issues/165880,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: activation checkpointing, module: functorch",scan when time allows,2025-10-20T14:29:08Z
issue,165902,torchcodec in pytorch url,https://github.com/pytorch/pytorch/issues/165902,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: binaries,scan when time allows,2025-10-20T14:27:16Z
issue,162235,Inconsistent handling of `-0.0` between CPU and CUDA for many operators,https://github.com/pytorch/pytorch/issues/162235,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: edge cases, module: nans and infs",scan when time allows,2025-10-20T01:36:26Z
issue,165421,[torch.compile][fullgraph][dynamic shapes] aten.masked_select.default rejected as “Dynamic shape operator” (eager OK),https://github.com/pytorch/pytorch/issues/165421,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamic shapes, oncall: pt2",scan when time allows,2025-10-19T23:10:31Z
issue,165799,`torch.where` does not accept scalar argument when `out=` is passed,https://github.com/pytorch/pytorch/issues/165799,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python frontend,scan when time allows,2025-10-19T19:21:34Z
issue,148552,Return type annotation of `Tensor.long()` etc is not narrowed down to dtype-specific names `LongTensor` etc,https://github.com/pytorch/pytorch/issues/148552,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: typing,scan when time allows,2025-10-19T15:10:34Z
issue,145337,[libTorch] Model initialization on multi-device is slow. It seems to run sequentially in multi-thread,https://github.com/pytorch/pytorch/issues/145337,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp,scan when time allows,2025-10-19T07:50:16Z
issue,165715,Missing spans in CPU section of profile likely caused by memory allocation with expandable segments,https://github.com/pytorch/pytorch/issues/165715,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, oncall: profiler",scan when time allows,2025-10-19T03:23:28Z
issue,165170,"A distributed test ""test_cupy_as_tensor"" fails on B200.",https://github.com/pytorch/pytorch/issues/165170,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: tests",scan when time allows,2025-10-17T23:58:12Z
issue,58779,C++ Extensions that use ninja should enable colors by default,https://github.com/pytorch/pytorch/issues/58779,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp-extensions,scan when time allows,2025-10-17T18:06:14Z
issue,65083,THPVariable_subclass_dealloc (i.e. tp_dealloc for torch.Tensor) should handle Python error state gracefully,https://github.com/pytorch/pytorch/issues/65083,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: internals",scan when time allows,2025-10-17T18:06:00Z
issue,64185,PyTorch public API cleanup,https://github.com/pytorch/pytorch/issues/64185,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2025-10-17T17:48:54Z
issue,29137,"torch.sum(tensor, dim=()) is different from np.sum(arr, axis=())",https://github.com/pytorch/pytorch/issues/29137,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bc-breaking, module: deprecation, module: numpy, module: reductions",scan when time allows,2025-10-17T17:48:03Z
issue,27780,torch.cuda.default_generators documentation referenced but don't exist.,https://github.com/pytorch/pytorch/issues/27780,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: docs",scan when time allows,2025-10-17T17:42:29Z
issue,3619,Multiprocessing with torch.solve hangs,https://github.com/pytorch/pytorch/issues/3619,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: linear algebra, module: multiprocessing",scan when time allows,2025-10-17T17:38:50Z
issue,165251,out_dtype for einsum,https://github.com/pytorch/pytorch/issues/165251,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: linear algebra, module: type promotion",scan when time allows,2025-10-17T15:55:36Z
issue,165744,Inconsistency in `conv_transpose2d` for float16 with inf weights between CPU and CUDA,https://github.com/pytorch/pytorch/issues/165744,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: nans and infs",scan when time allows,2025-10-17T14:42:16Z
issue,165739,Inconsistent in `torch.fft.irfft` and `torch.fft.ifftn` for complex64 inputs containing `Inf`,https://github.com/pytorch/pytorch/issues/165739,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: fft, module: nans and infs",scan when time allows,2025-10-17T14:41:15Z
issue,165685,"`PYTORCH_CUDA_ALLOC_CONF=""expandable_segments:True""` triggers `RuntimeError: pidfd_getfd: Operation not permitted` for CUDA IPC transfer",https://github.com/pytorch/pytorch/issues/165685,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: cudacachingallocator",scan when time allows,2025-10-17T04:32:48Z
issue,165634,make_fx pre_dispatch=True produces incorrect graph when BatchedFallback is triggered,https://github.com/pytorch/pytorch/issues/165634,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2025-10-17T04:30:22Z
issue,165578,Out of tree backend documentation does not seem accurate,https://github.com/pytorch/pytorch/issues/165578,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: privateuse1",scan when time allows,2025-10-17T04:30:01Z
issue,165570,Add support for element-wise operations to flop counter,https://github.com/pytorch/pytorch/issues/165570,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: flop counter,scan when time allows,2025-10-17T04:29:30Z
issue,114629,[DeviceMesh] `init_device_mesh` does not support CPU-only,https://github.com/pytorch/pytorch/issues/114629,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: devicemesh, oncall: distributed",scan when time allows,2025-10-17T00:42:43Z
issue,165684,Python version in test CI job doesn't align with the python version in the docker build log,https://github.com/pytorch/pytorch/issues/165684,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ci, module: docker, module: infra",scan when time allows,2025-10-16T22:51:16Z
issue,160838,backward on 5090(sm_120) much slower than on 4090,https://github.com/pytorch/pytorch/issues/160838,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: performance",scan when time allows,2025-10-16T14:45:17Z
issue,165621,Feedback about RNN,https://github.com/pytorch/pytorch/issues/165621,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: rnn",scan when time allows,2025-10-16T14:18:23Z
issue,155671,torch.clamp throws overflow error on CPU but not on CUDA,https://github.com/pytorch/pytorch/issues/155671,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: error checking",scan when time allows,2025-10-15T23:21:38Z
issue,114605,vmap: Transform single-element tensor to integer,https://github.com/pytorch/pytorch/issues/114605,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: functorch,scan when time allows,2025-10-15T19:03:16Z
issue,107873,"[BE] Consolidation of SymNode methods constant_int, maybe_as_int, etc",https://github.com/pytorch/pytorch/issues/107873,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamic shapes, oncall: pt2",scan when time allows,2025-10-15T18:54:57Z
issue,106614,Case study of torch.compile / cpp inductor on CPU: min_sum / mul_sum with 1d / matmul-like with static / dynamic shapes,https://github.com/pytorch/pytorch/issues/106614,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: custom-operators, module: inductor, oncall: pt2",scan when time allows,2025-10-15T18:35:31Z
issue,94167,"Faster `pad_sequence` and `tensor_split` function with CUDA kernel, are they possible?",https://github.com/pytorch/pytorch/issues/94167,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, module: rnn, oncall: pt2",scan when time allows,2025-10-15T18:28:10Z
issue,80372,[BE] Update ProcessGroupWrapper tests to test other collective message,https://github.com/pytorch/pytorch/issues/80372,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: c10d, oncall: distributed",scan when time allows,2025-10-15T16:52:25Z
issue,48959,[Poll] Support DistributedDataParallel (DDP) in PyTorch C++ API (libtorch),https://github.com/pytorch/pytorch/issues/48959,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ddp, oncall: distributed",scan when time allows,2025-10-15T15:49:55Z
issue,165038,Print float tensors on devices that do not support fp64,https://github.com/pytorch/pytorch/issues/165038,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: backend, module: privateuse1, module: python frontend",scan when time allows,2025-10-15T07:16:53Z
issue,147044,Clean up backend_type_map from distributed_c10d,https://github.com/pytorch/pytorch/issues/147044,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: c10d, oncall: distributed",scan when time allows,2025-10-15T06:28:35Z
issue,129649,Low GPU precision,https://github.com/pytorch/pytorch/issues/129649,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: numerical-stability",scan when time allows,2025-10-15T05:34:55Z
issue,122274,CUDA: Performance issues with `torch.unique`,https://github.com/pytorch/pytorch/issues/122274,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: performance",scan when time allows,2025-10-15T01:53:39Z
issue,146817,BF16 linear(matmul) operator 100x slower on odd matrix dimension sizes on A100,https://github.com/pytorch/pytorch/issues/146817,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cublas, module: cuda, module: linear algebra, module: performance",scan when time allows,2025-10-15T01:40:57Z
issue,147134,`torch.mul` uses `OpMathType` for computation.,https://github.com/pytorch/pytorch/issues/147134,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bfloat16, module: cuda, module: half, module: numerical-stability",scan when time allows,2025-10-14T23:48:00Z
issue,151433,The compare interface of the memory snapshot visualization tool “_memory_viz.py” has a bug,https://github.com/pytorch/pytorch/issues/151433,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: memory usage, oncall: profiler",scan when time allows,2025-10-14T23:45:04Z
issue,165425,Support FFT operations in FLOP counter,https://github.com/pytorch/pytorch/issues/165425,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: fft, module: flop counter",scan when time allows,2025-10-14T23:15:49Z
issue,164732,Fake tensor leakage in aot_autograd by itself,https://github.com/pytorch/pytorch/issues/164732,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: faketensor, module: functorch, module: pt2-dispatcher, oncall: pt2",scan when time allows,2025-10-14T18:21:11Z
issue,165244,torch.tensor slicing with negative step raises ValueError,https://github.com/pytorch/pytorch/issues/165244,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: python array api, module: python frontend",scan when time allows,2025-10-14T15:05:25Z
issue,165277,"autowrap_modules should be tuple[ModuleType, ...]",https://github.com/pytorch/pytorch/issues/165277,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2025-10-14T13:56:12Z
issue,165312,TransformerEncoder crashes with IndexError when num_layers=0,https://github.com/pytorch/pytorch/issues/165312,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2025-10-14T13:34:02Z
issue,119574,Cannot configure dist_timeout when using device_mesh,https://github.com/pytorch/pytorch/issues/119574,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: devicemesh, oncall: distributed",scan when time allows,2025-10-14T08:42:59Z
issue,154226,NotImplementedError when computing JVP of Attention,https://github.com/pytorch/pytorch/issues/154226,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: forward ad, module: sdpa",scan when time allows,2025-10-14T08:16:43Z
issue,156062,torch.nn.functional.conv_transpose3d return inconsistent results when weight containing inf between CPU and GPU,https://github.com/pytorch/pytorch/issues/156062,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2025-10-14T08:12:55Z
issue,165319,Memory leak when converting from numpy array,https://github.com/pytorch/pytorch/issues/165319,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: memory usage, module: numpy",scan when time allows,2025-10-14T08:06:37Z
issue,164925,add support for distributed backends to specify default backend in multiple backends format,https://github.com/pytorch/pytorch/issues/164925,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2025-10-14T03:17:00Z
issue,165122,RPC: abort when RRef from remote() is mistakenly called with wait() due to unhandled exception in thread pool,https://github.com/pytorch/pytorch/issues/165122,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2025-10-13T17:33:50Z
issue,164799,Check behavior of multicast object release when using clang to compile PyTorch,https://github.com/pytorch/pytorch/issues/164799,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, oncall: distributed",scan when time allows,2025-10-13T17:19:21Z
issue,164843,`set_model_state_dict()` causes `AssertionError: found no DeviceMesh from dtensor args for c10d.broadcast_.default!`,https://github.com/pytorch/pytorch/issues/164843,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2025-10-13T17:19:01Z
issue,164122,[CUDA][Sparse][Linalg] torch.sparse.spsolve is unavailable,https://github.com/pytorch/pytorch/issues/164122,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: cuda, module: linear algebra, module: sparse",scan when time allows,2025-10-13T13:38:08Z
issue,132292,Can we avoid RecordStream for P2P ops?,https://github.com/pytorch/pytorch/issues/132292,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2025-10-13T08:21:44Z
issue,130073,☂️ TORCH_INTERNAL_ASSERT failures while running CI,https://github.com/pytorch/pytorch/issues/130073,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: assert failure, module: ci, module: dynamic shapes",scan when time allows,2025-10-13T06:34:57Z
issue,126738,FakeTensor `forward()` on `bert` attention when using FakeTensor returns a DataDependentOutputException,https://github.com/pytorch/pytorch/issues/126738,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamic shapes, module: faketensor",scan when time allows,2025-10-13T05:48:52Z
issue,160438,torch.autocast with no_grad() on aliased model disables gradients for subsequent forward on same instance,https://github.com/pytorch/pytorch/issues/160438,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision),scan when time allows,2025-10-13T00:52:18Z
issue,82053,Inconsistent implementation of quant_utils:: ChooseQuantizationParams compared with fbgemm:: ChooseQuantizationParams,https://github.com/pytorch/pytorch/issues/82053,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2025-10-12T22:40:48Z
issue,31394,torch.scatter_logsumexp,https://github.com/pytorch/pytorch/issues/31394,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: scatter & gather ops,scan when time allows,2025-10-12T21:32:22Z
issue,140696,inconsistency in ```torch.nn.functional.adaptive_avg_pool3d``` on CPU and GPU,https://github.com/pytorch/pytorch/issues/140696,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: nn",scan when time allows,2025-10-12T20:31:18Z
issue,114569,inconsistency on torch.clamp,https://github.com/pytorch/pytorch/issues/114569,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: nans and infs",scan when time allows,2025-10-12T02:09:49Z
issue,142232,torch.tensor.to(‘cuda’) so slow in jetson orin,https://github.com/pytorch/pytorch/issues/142232,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: jetson",scan when time allows,2025-10-11T23:45:29Z
issue,153497,NotImplementedError: Could not run 'aten::log' with arguments from the 'SparseCUDA' backend.,https://github.com/pytorch/pytorch/issues/153497,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2025-10-11T23:35:58Z
issue,161381,[XPU] ​​Incorrect free_memory reported by torch.xpu.mem_get_info()​,https://github.com/pytorch/pytorch/issues/161381,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: xpu,scan when time allows,2025-10-11T01:58:32Z
issue,165173,Add CUDA 13.0 in CI,https://github.com/pytorch/pytorch/issues/165173,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ci, module: cuda",scan when time allows,2025-10-10T23:52:46Z
issue,150021,[ued] I want to see the full sizes/strides in TORCH_LOGS=recompiles,https://github.com/pytorch/pytorch/issues/150021,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: compile ux, module: dynamo, oncall: pt2",scan when time allows,2025-10-10T16:23:21Z
issue,121286,"Assertion error (`assert op == ""==""`) in symbolic_shapes.prettify_results for operator `<=`",https://github.com/pytorch/pytorch/issues/121286,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamic shapes, oncall: export, oncall: pt2",scan when time allows,2025-10-10T16:09:38Z
issue,114414,[BE] remove SmallVector optimization in PyInterpreter.cpp when storing custom sizes,https://github.com/pytorch/pytorch/issues/114414,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: __torch_dispatch__, module: dynamic shapes",scan when time allows,2025-10-10T15:43:00Z
issue,165100,Header files not found during build,https://github.com/pytorch/pytorch/issues/165100,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2025-10-10T13:43:50Z
issue,165042,constant ones/zeros tensor,https://github.com/pytorch/pytorch/issues/165042,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2025-10-10T13:42:00Z
issue,144824,[MPS] Indexing Returns 0 if OOB,https://github.com/pytorch/pytorch/issues/144824,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: error checking, module: mps",scan when time allows,2025-10-10T12:43:08Z
issue,153327,FPE when calling `torch.pixel_shuffle()` with empty tensors and large upscale_factor,https://github.com/pytorch/pytorch/issues/153327,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: empty tensor, module: error checking",scan when time allows,2025-10-10T09:51:26Z
issue,123119,`nn.Embedding` behaves differently on CPU (fails) and MPS (fails not),https://github.com/pytorch/pytorch/issues/123119,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: embedding, module: mps, module: nn",scan when time allows,2025-10-10T09:31:20Z
issue,165040,DISABLED rocm / linux-jammy-rocm-py3_10-gfx1100 / test (default),https://github.com/pytorch/pytorch/issues/165040,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ci, module: rocm",scan when time allows,2025-10-09T23:59:25Z
issue,132067,Pickle loads torch tensor memory leak,https://github.com/pytorch/pytorch/issues/132067,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: serialization,scan when time allows,2025-10-09T21:31:20Z
issue,164302,Mixing cuda graph and non cuda graph code works poorly in the presence of autograd.,https://github.com/pytorch/pytorch/issues/164302,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: cuda graphs",scan when time allows,2025-10-09T18:08:40Z
issue,131283,"`input` argument of `nan_to_num()` works with `complex` type but `nan`, `posinf` and `neginf` argument don't work with `complex` type",https://github.com/pytorch/pytorch/issues/131283,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: complex,scan when time allows,2025-10-09T13:27:09Z
issue,67158,Make streams used for NCCL operations configurable,https://github.com/pytorch/pytorch/issues/67158,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nccl, oncall: distributed",scan when time allows,2025-10-09T03:32:19Z
issue,164062,[strict-export][install_free_tensors] Tracker for issues,https://github.com/pytorch/pytorch/issues/164062,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: export, oncall: pt2",scan when time allows,2025-10-08T23:54:49Z
issue,163013,Compiled HOPS do not support compilation when run under a custom DispatchMode,https://github.com/pytorch/pytorch/issues/163013,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: __torch_dispatch__, module: higher order operators, module: pt2-dispatcher, oncall: pt2",scan when time allows,2025-10-08T20:09:19Z
issue,164872,Tensor subclass slice during `inference_mode` fails,https://github.com/pytorch/pytorch/issues/164872,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dispatch, oncall: quantization",scan when time allows,2025-10-08T19:33:54Z
issue,164951,Docker checkouts take 30+ min on H100 runners,https://github.com/pytorch/pytorch/issues/164951,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2025-10-08T17:12:25Z
issue,164759,Bug : Side effect in th.func.vjp of th.func.jvp,https://github.com/pytorch/pytorch/issues/164759,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: functorch",scan when time allows,2025-10-07T21:34:08Z
issue,164787,Add static build for NCCL in CI,https://github.com/pytorch/pytorch/issues/164787,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ci, module: cuda",scan when time allows,2025-10-07T21:30:38Z
issue,164862,Fix incomplete exception handling in validate_input_col function,https://github.com/pytorch/pytorch/issues/164862,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: data,scan when time allows,2025-10-07T21:30:05Z
issue,158782,[BE] linter for base docker image configurations with docker_builds.yml,https://github.com/pytorch/pytorch/issues/158782,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ci, module: docker, module: lint",scan when time allows,2025-10-07T20:33:52Z
issue,118842,CUDA Extension Bug for Pytorch 2.2.0,https://github.com/pytorch/pytorch/issues/118842,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp-extensions, module: cuda",scan when time allows,2025-10-07T20:03:43Z
issue,95408,Parallel Associative Scan,https://github.com/pytorch/pytorch/issues/95408,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: functorch, module: pt2-dispatcher, oncall: pt2",scan when time allows,2025-10-07T19:21:22Z
issue,154311,torch.inf.to(torch.int32) produces different values on CPU vs CUDA,https://github.com/pytorch/pytorch/issues/154311,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: edge cases, module: nans and infs",scan when time allows,2025-10-07T18:17:05Z
issue,164756,[PP] Unshard and reshard not working with FSDP,https://github.com/pytorch/pytorch/issues/164756,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: pipelining, oncall: distributed",scan when time allows,2025-10-07T16:57:43Z
issue,25570,Remote memory access similar to MPI one-sided in pytorch,https://github.com/pytorch/pytorch/issues/25570,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: mpi, oncall: distributed",scan when time allows,2025-10-07T16:18:04Z
issue,98836,PyTorch's packaged libgomp causes significant performance penalties on CPU when used together with other Python packages,https://github.com/pytorch/pytorch/issues/98836,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: multithreading",scan when time allows,2025-10-07T15:27:59Z
issue,164786,How should we handle PyTorch build flags in torch/headeronly for custom ops?,https://github.com/pytorch/pytorch/issues/164786,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: custom-operators, module: pt2-dispatcher, oncall: pt2",scan when time allows,2025-10-07T15:26:28Z
issue,100249,"[discussion] ""TensorList"" as first-class python-world abstraction (to mirror existing dispatcher's `TensorList` and `torch::autograd::tensor_list`) and as key for dispatch for merging `torch._foreach_*` into regular `torch.*` functions",https://github.com/pytorch/pytorch/issues/100249,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nestedtensor,scan when time allows,2025-10-07T14:02:40Z
issue,164331,Anomaly mode (detect_anomaly) should populate backward stacks with torch.profiler.profile(with_stack=True),https://github.com/pytorch/pytorch/issues/164331,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, oncall: profiler",scan when time allows,2025-10-06T19:32:59Z
issue,137392,segfault when using DTensor with nonblocking nccl comm,https://github.com/pytorch/pytorch/issues/137392,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dtensor, module: nccl, oncall: distributed",scan when time allows,2025-10-06T10:28:38Z
issue,152717,Cannot mask a DTensor,https://github.com/pytorch/pytorch/issues/152717,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dtensor, oncall: distributed",scan when time allows,2025-10-05T13:01:54Z
issue,47052,Activation functions for complex tensors,https://github.com/pytorch/pytorch/issues/47052,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: complex, module: nn",scan when time allows,2025-10-04T12:24:51Z
issue,77049,Peak GPU-memory usage extremely huge when sorting with torch.sort,https://github.com/pytorch/pytorch/issues/77049,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: memory usage, module: sorting and selection",scan when time allows,2025-10-03T17:16:47Z
issue,155472,[RFC][c10d] Make it easier to act on group,https://github.com/pytorch/pytorch/issues/155472,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2025-10-03T01:41:09Z
issue,125984,`torch.export` code that includes `torch.autograd.grad` RuntimeError: element 0 of tensors does not require grad and does not have a grad_fn,https://github.com/pytorch/pytorch/issues/125984,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: onnx, oncall: export, oncall: pt2",scan when time allows,2025-10-02T21:29:48Z
issue,164249,Accuracy issue with sum-reduction on A100,https://github.com/pytorch/pytorch/issues/164249,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: numerical-stability, module: tests",scan when time allows,2025-10-02T14:57:43Z
issue,162719,linalg.eig does not get parallelized on CPU,https://github.com/pytorch/pytorch/issues/162719,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: linear algebra, module: performance",scan when time allows,2025-10-02T12:03:49Z
issue,101241,Different results with vmap when using torch.jit.script,https://github.com/pytorch/pytorch/issues/101241,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: functorch, oncall: jit",scan when time allows,2025-10-02T12:03:26Z
issue,153976,MPS (Metal) CTC Loss Implementation (Gist link),https://github.com/pytorch/pytorch/issues/153976,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: mps,scan when time allows,2025-10-02T02:58:52Z
issue,77724,FSDP: enhanced shared parameter support,https://github.com/pytorch/pytorch/issues/77724,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: fsdp, oncall: distributed",scan when time allows,2025-10-01T17:44:24Z
issue,163709,Inconsistent handling of backend floating point precision settings (related to PR #125888),https://github.com/pytorch/pytorch/issues/163709,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: performance, module: python frontend",scan when time allows,2025-10-01T17:40:17Z
issue,102109,Can't vmap over a slice expression,https://github.com/pytorch/pytorch/issues/102109,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: functorch,scan when time allows,2025-09-30T20:44:12Z
issue,158127,"A ""Merge-when-ready"" label when ci finishes?",https://github.com/pytorch/pytorch/issues/158127,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2025-09-30T01:27:08Z
issue,71465,torch.nn.LayerNorm support for arbitrary axis in order to allow NCHW application,https://github.com/pytorch/pytorch/issues/71465,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2025-09-29T23:18:37Z
issue,27647,🚀 Graceful RPCAgent termination in multi-driver scenario,https://github.com/pytorch/pytorch/issues/27647,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rpc,scan when time allows,2025-09-29T23:16:37Z
issue,61417,Reductions tracking issue,https://github.com/pytorch/pytorch/issues/61417,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: numpy, module: reductions",scan when time allows,2025-09-29T23:15:54Z
issue,22402,PyTorch Tensor subclasses and protocols for NumPy interoperability,https://github.com/pytorch/pytorch/issues/22402,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: numpy,scan when time allows,2025-09-29T23:14:13Z
issue,22815,TensorImpl de-virtualization,https://github.com/pytorch/pytorch/issues/22815,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dependency bug, module: internals",scan when time allows,2025-09-29T23:12:08Z
issue,25785,[Feature request] modified Cholesky decomposition,https://github.com/pytorch/pytorch/issues/25785,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: linear algebra,scan when time allows,2025-09-29T23:11:08Z
issue,31895,"torch.masked_select out argument can easily be misused, because output shape is dynamically computed",https://github.com/pytorch/pytorch/issues/31895,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: numpy, module: safe resize",scan when time allows,2025-09-29T23:10:54Z
issue,24243,hasSideEffects INTERNAL ASSERT FAILED when using .split method with JIT,https://github.com/pytorch/pytorch/issues/24243,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2025-09-29T23:04:37Z
issue,20367,Overhead performance regression over time umbrella issue.,https://github.com/pytorch/pytorch/issues/20367,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: cuda, module: internals, module: performance",scan when time allows,2025-09-29T23:02:34Z
issue,18631,FP32 depthwise convolution is slow in GPU,https://github.com/pytorch/pytorch/issues/18631,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: cuda, module: cudnn, module: dependency bug",scan when time allows,2025-09-29T23:01:25Z
issue,18182,Update weight initialisations to current best practices,https://github.com/pytorch/pytorch/issues/18182,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bc-breaking, module: nn",scan when time allows,2025-09-29T23:01:08Z
issue,11390,Views created in no_grad block still have requires_grad=True,https://github.com/pytorch/pytorch/issues/11390,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: docs",scan when time allows,2025-09-29T23:00:48Z
issue,15829,computing entropy of a tensor,https://github.com/pytorch/pytorch/issues/15829,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2025-09-29T22:56:42Z
issue,158428,Inconsistent torch.cholesky_inverse results on complex128 between CPU and CUDA,https://github.com/pytorch/pytorch/issues/158428,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: linear algebra,scan when time allows,2025-09-29T21:44:15Z
issue,163950,ImportError: cannot import name 'skip_code' from 'torch._C._dynamo.eval_frame' (unknown location),https://github.com/pytorch/pytorch/issues/163950,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, oncall: pt2",scan when time allows,2025-09-29T16:48:15Z
issue,74540,No factory functions for strided quantized tensors,https://github.com/pytorch/pytorch/issues/74540,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2025-09-29T16:00:42Z
issue,164061,GPU Memory Leak due to distributions,https://github.com/pytorch/pytorch/issues/164061,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distributions,scan when time allows,2025-09-29T14:54:42Z
issue,163821,Support for nn.BitLinear layer,https://github.com/pytorch/pytorch/issues/163821,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2025-09-29T13:55:57Z
issue,154842,`torch.linalg.solve` does not raise an error for singular matrix on CPU.,https://github.com/pytorch/pytorch/issues/154842,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: linear algebra,scan when time allows,2025-09-29T13:48:33Z
issue,164089,Failing Operator Benchmarks,https://github.com/pytorch/pytorch/issues/164089,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2025-09-29T13:46:11Z
issue,67470,"integer (and possibly float16) matmul fails test_noncontiguous_samples on CPU (and all backward dtypes, too?)",https://github.com/pytorch/pytorch/issues/67470,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: linear algebra",scan when time allows,2025-09-27T03:31:41Z
issue,53178,Exceedingly different F.conv2d outputs on cuda/cpu,https://github.com/pytorch/pytorch/issues/53178,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: correctness (silent), module: cuda, module: cudnn",scan when time allows,2025-09-27T03:30:44Z
issue,48900,addmm with out= argument returns incorrect result,https://github.com/pytorch/pytorch/issues/48900,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: docs, module: linear algebra, module: partial aliasing",scan when time allows,2025-09-27T03:30:07Z
issue,68070,C++ at::Tensor's pinned_memory status is not printing out correctly.,https://github.com/pytorch/pytorch/issues/68070,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: cpp",scan when time allows,2025-09-27T03:27:19Z
issue,56794,No tensor operations allowed inside at::parallel_for,https://github.com/pytorch/pytorch/issues/56794,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: multithreading,scan when time allows,2025-09-26T23:50:42Z
issue,64412,Memory leak in multi-thread inference,https://github.com/pytorch/pytorch/issues/64412,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: memory usage, module: multithreading, module: performance",scan when time allows,2025-09-26T23:49:45Z
issue,68174,RFC: “What’s in a (NumPY) name?” — Changing PyTorch Operator Names,https://github.com/pytorch/pytorch/issues/68174,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: numpy,scan when time allows,2025-09-26T23:49:06Z
issue,71774,matmul returns uninitialized memory for int64 tensors with inner dimension of zero,https://github.com/pytorch/pytorch/issues/71774,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: linear algebra",scan when time allows,2025-09-26T23:46:21Z
issue,75432,Rollup: Top forward-over-reverse formulas,https://github.com/pytorch/pytorch/issues/75432,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: forward ad",scan when time allows,2025-09-26T23:45:15Z
issue,74613,OpInfo request for `nn.functional` and `unbind`,https://github.com/pytorch/pytorch/issues/74613,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tests,scan when time allows,2025-09-26T23:44:05Z
issue,98456,README could use link to governance,https://github.com/pytorch/pytorch/issues/98456,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2025-09-26T23:37:12Z
issue,111384,[quant][pt2] Default batchnorm aten op has poor numerics during QAT,https://github.com/pytorch/pytorch/issues/111384,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2025-09-26T23:35:18Z
issue,117586,PyTorch 2.1.0 High memory utilization compared to PyTorch 2.0.1,https://github.com/pytorch/pytorch/issues/117586,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2025-09-26T23:33:59Z
issue,163789,[docs] instructions to locally build docs are underspecified,https://github.com/pytorch/pytorch/issues/163789,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2025-09-26T22:34:16Z
issue,163982,Need to update Magma version in Pytorch,https://github.com/pytorch/pytorch/issues/163982,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2025-09-26T19:23:09Z
issue,163902,Add banner for docker sevs to hud,https://github.com/pytorch/pytorch/issues/163902,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ci, module: docker",scan when time allows,2025-09-26T02:30:09Z
issue,161713,[Proposal] Improving Type Signatures for Types Exported by torch,https://github.com/pytorch/pytorch/issues/161713,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: typing,scan when time allows,2025-09-25T13:52:43Z
issue,163062,GIL is not released when creating a device tensor using a shape depending on data from another device tensor,https://github.com/pytorch/pytorch/issues/163062,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: python frontend",scan when time allows,2025-09-25T02:23:29Z
issue,163659,Allow double in native_functions.yaml as a schema type,https://github.com/pytorch/pytorch/issues/163659,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp-extensions, module: dispatch, module: library, module: pt2-dispatcher",scan when time allows,2025-09-24T18:45:19Z
issue,158831,Severe Performance Degradation in Batched SVD on A100-40G,https://github.com/pytorch/pytorch/issues/158831,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: linear algebra, module: performance",scan when time allows,2025-09-24T09:42:01Z
issue,163048,[compile/aot-autograd] `torch.autograd.grad` w.r.t. an intermediate tensor works in eager but fails under `torch.compile` (“tensor not used in the graph”),https://github.com/pytorch/pytorch/issues/163048,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: aotdispatch, module: pt2-dispatcher, oncall: pt2",scan when time allows,2025-09-24T03:34:55Z
issue,124645,Registering function that takes `const SymInt&` to op that accepts `SymInt` leads to cryptic error,https://github.com/pytorch/pytorch/issues/124645,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: custom-operators, module: dynamic shapes, module: pt2-dispatcher",scan when time allows,2025-09-24T02:12:03Z
issue,162971,[CD] Reasonable time constraint for binary builds,https://github.com/pytorch/pytorch/issues/162971,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: ci",scan when time allows,2025-09-23T20:23:03Z
issue,163302,[vLLM CI] Periodically run B200 tests,https://github.com/pytorch/pytorch/issues/163302,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2025-09-23T20:13:26Z
issue,163577,Lintrunner failures on files that only manifest on CI,https://github.com/pytorch/pytorch/issues/163577,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2025-09-23T20:05:06Z
issue,159674,OOM when using use_mem_pool due to restriction on retry,https://github.com/pytorch/pytorch/issues/159674,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2025-09-23T17:58:14Z
issue,163657,UNSTABLE periodic / linux-jammy-cuda12.4-py3.10-gcc11 / test (legacy_nvidia_driver),https://github.com/pytorch/pytorch/issues/163657,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2025-09-23T16:18:42Z
issue,163624,[aoti] [xpu] [null-pointer-deference] potential npt issue in `sycl_runtime_wrappers.h`,https://github.com/pytorch/pytorch/issues/163624,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: aotinductor, oncall: export, oncall: pt2",scan when time allows,2025-09-23T16:02:18Z
issue,163574,Bfloat16 large numerical discrepancy on Blackwell,https://github.com/pytorch/pytorch/issues/163574,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bfloat16, module: correctness (silent), module: cuda, module: numerical-stability",scan when time allows,2025-09-23T15:50:27Z
issue,144464,"[ONNX] Trying to export QAT model with torch.onnx.export, running into Problems that workarounds cant fix. Is ONNX export planned for QAT?",https://github.com/pytorch/pytorch/issues/144464,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2025-09-23T13:34:16Z
issue,140520,Enable autograd for padded dense forward / backward operators.,https://github.com/pytorch/pytorch/issues/140520,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: nestedtensor",scan when time allows,2025-09-22T22:27:06Z
issue,162820,[CI][CUDA][Distributed] test_ring_flex_attention failed on 8xB200 Runner,https://github.com/pytorch/pytorch/issues/162820,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ci, module: flex attention, module: higher order operators, module: pt2-dispatcher",scan when time allows,2025-09-22T22:23:48Z
issue,163502,at::Tensor::to_sparse_csr() does not produce correct result in Debug configuration when tensor is big sparse CSC on CUDA device in LibTorch 2.8.0,https://github.com/pytorch/pytorch/issues/163502,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: cuda, module: debug-build, module: sparse",scan when time allows,2025-09-22T19:45:59Z
issue,90607,"Export to ONNX of `as_strided()` hard codes stride in the graph, although it should be dynamic",https://github.com/pytorch/pytorch/issues/90607,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2025-09-22T18:39:40Z
issue,162449,optimum-cli export onnx for model intfloat/multilingual-e5-large-instruct fails with sufficient accuracy,https://github.com/pytorch/pytorch/issues/162449,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2025-09-22T18:38:48Z
issue,126774,[ONNX] Implement aten._trilinear and aten.bilinear.default,https://github.com/pytorch/pytorch/issues/126774,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2025-09-22T18:38:24Z
issue,159380,`torch.nested.nested_tensor_from_jagged` does not automatically compute `max_seqlen`,https://github.com/pytorch/pytorch/issues/159380,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nestedtensor,scan when time allows,2025-09-22T17:53:31Z
issue,162918,`torch.embedding` should accept `indices` of dtype `torch.uint8`,https://github.com/pytorch/pytorch/issues/162918,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: embedding, module: nn",scan when time allows,2025-09-22T16:29:45Z
issue,119108,[MPS] Tracking issue for ModuleInfo failures when enabling testing for `torch.float16`,https://github.com/pytorch/pytorch/issues/119108,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: mps,scan when time allows,2025-09-22T14:41:09Z
issue,162178,[CUDA] Umbrella Issues/Failures on B200 Runner,https://github.com/pytorch/pytorch/issues/162178,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2025-09-22T00:29:33Z
issue,153737,SystemError: PY_SSIZE_T_CLEAN macro must be defined for '#' formats - PyTorch compile fails with Python 3.12,https://github.com/pytorch/pytorch/issues/153737,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, module: python version, oncall: pt2",scan when time allows,2025-09-21T19:16:22Z
issue,161330,better tool to repro issues in 1 dynamo graph,https://github.com/pytorch/pytorch/issues/161330,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",scan when time allows,2025-09-20T01:35:49Z
issue,162782,Is `torch.nn.functional.gumbel_softmax` going to be deprecated?,https://github.com/pytorch/pytorch/issues/162782,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: deprecation, module: nn",scan when time allows,2025-09-19T17:36:22Z
issue,40319,Too many labels in the repo,https://github.com/pytorch/pytorch/issues/40319,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: infra,scan when time allows,2025-09-19T01:25:50Z
issue,163142,`compile_kernel`: fast inline compilation with nvrtc tracker,https://github.com/pytorch/pytorch/issues/163142,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2025-09-18T22:43:32Z
issue,163290,Tracker for migrating xformers to torch stable ABI,https://github.com/pytorch/pytorch/issues/163290,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: abi,scan when time allows,2025-09-18T20:43:43Z
issue,160816,torch._scaled_mm crash with cublas error when out_dtype is not specified,https://github.com/pytorch/pytorch/issues/160816,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: floatx (formerly float8)",scan when time allows,2025-09-18T14:38:51Z
issue,29961,Add support for integer matrix multiplication (particularly for dtype = torch.int8 ),https://github.com/pytorch/pytorch/issues/29961,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2025-09-18T13:50:01Z
issue,157714,AssertHandler::printMessage On Intel GPU,https://github.com/pytorch/pytorch/issues/157714,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: xpu,scan when time allows,2025-09-18T07:12:23Z
issue,90585,[threaded pg] All threads share one Random Number Generator,https://github.com/pytorch/pytorch/issues/90585,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dtensor, module: random, module: threaded pg",scan when time allows,2025-09-18T03:15:02Z
issue,162964,[feat] OpInfo-like concept for subclass edge conditions to handle,https://github.com/pytorch/pytorch/issues/162964,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: testing,scan when time allows,2025-09-18T01:14:03Z
issue,162786,[MSVC] LINK : fatal error LNK1181: cannot open input file 'sleef\src\libm\CMakeFiles\mkrename.dir\mkrename.c.obj',https://github.com/pytorch/pytorch/issues/162786,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: windows,scan when time allows,2025-09-18T01:12:51Z
issue,153502,Add MXFP8 Support to scaled_grouped_gemm,https://github.com/pytorch/pytorch/issues/153502,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: floatx (formerly float8)",scan when time allows,2025-09-17T20:55:53Z
issue,162907,Poor bf16 precision in backward torch.asin,https://github.com/pytorch/pytorch/issues/162907,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: cpu",scan when time allows,2025-09-17T20:30:36Z
issue,162911,`torch._foreach_...` functions error on zero-length inputs,https://github.com/pytorch/pytorch/issues/162911,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: mta,scan when time allows,2025-09-17T20:28:54Z
issue,102894,"BCELoss and BCEWithLogitsLoss differ when one of the input logits is float(""inf"")",https://github.com/pytorch/pytorch/issues/102894,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: edge cases, module: nn",scan when time allows,2025-09-17T18:13:47Z
issue,140514,Error with fused AdamW and FSDP,https://github.com/pytorch/pytorch/issues/140514,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: fsdp, module: optimizer",scan when time allows,2025-09-17T16:00:32Z
issue,103580,Support ByteTensor and ShortTensor for nn.Embedding and nn.EmbeddingBag,https://github.com/pytorch/pytorch/issues/103580,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2025-09-17T05:46:10Z
issue,69532,[feature request] A rank-revealing SVD for better stability in backward.,https://github.com/pytorch/pytorch/issues/69532,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: linear algebra, module: numerical-stability",scan when time allows,2025-09-17T04:42:36Z
issue,106667,[docs] Idea collection of examples of custom ops / inline torch extensions,https://github.com/pytorch/pytorch/issues/106667,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: custom-operators, module: docs",scan when time allows,2025-09-17T04:40:01Z
issue,36524,"Drop _stacklevel from argspecs of F.softmax, F.softmin, F.log_softmax (for implicit dim has been long deprecated)",https://github.com/pytorch/pytorch/issues/36524,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2025-09-17T03:51:03Z
issue,83163,torch.nn.functional.log_softmax parameter '_stacklevel' undocumented,https://github.com/pytorch/pytorch/issues/83163,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2025-09-17T03:48:59Z
issue,156618,THPStorage_shareCuda zeroes CUDA tensor,https://github.com/pytorch/pytorch/issues/156618,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: dependency bug, module: multiprocessing, module: wsl",scan when time allows,2025-09-16T21:53:17Z
issue,40497,Mixed precision causes NaN loss,https://github.com/pytorch/pytorch/issues/40497,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision), module: nans and infs",scan when time allows,2025-09-16T13:41:35Z
issue,162870,[RFC] library function with 64+ arguments,https://github.com/pytorch/pytorch/issues/162870,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: custom-operators, module: library",scan when time allows,2025-09-15T23:25:57Z
issue,151030,Unexpected memory usage in FSDP 2 Hybrid Sharding (HSDP),https://github.com/pytorch/pytorch/issues/151030,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: fsdp, oncall: distributed, oncall: distributed checkpointing",scan when time allows,2025-09-15T21:44:30Z
issue,162700,"RuntimeError: The tensor has a non-zero number of elements, but its data is not allocated yet during torch.backward with InternVL",https://github.com/pytorch/pytorch/issues/162700,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: pt2,scan when time allows,2025-09-15T20:45:48Z
issue,162154,Linear algebra operations hang indefinitely doing nothing when run in subprocesses,https://github.com/pytorch/pytorch/issues/162154,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: deadlock, module: linear algebra, module: multiprocessing",scan when time allows,2025-09-15T15:12:54Z
issue,158861,Different behavior between sparse and dense tensors with broadcasting multiplication.,https://github.com/pytorch/pytorch/issues/158861,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2025-09-15T06:07:44Z
issue,94294,CUBLAS_STATUS_NOT_SUPPORTED when calling cublasDgemv,https://github.com/pytorch/pytorch/issues/94294,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cublas, module: cuda",scan when time allows,2025-09-14T06:20:10Z
issue,162837,XPU backend is falling back to run on the CPU.,https://github.com/pytorch/pytorch/issues/162837,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: xpu,scan when time allows,2025-09-14T00:21:50Z
issue,58054,PyTorch master failed to build with glog 0.5.0.,https://github.com/pytorch/pytorch/issues/58054,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2025-09-13T00:17:24Z
issue,124505,"[feature request] allow torch.compile calls with compiler options, but without modifying global dynamo options",https://github.com/pytorch/pytorch/issues/124505,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: compile ux, module: dynamo, oncall: pt2",scan when time allows,2025-09-12T21:28:50Z
issue,131263,An API to re-enable dynamo-disabled functions,https://github.com/pytorch/pytorch/issues/131263,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: compile ux, module: dynamo, oncall: pt2",scan when time allows,2025-09-12T21:26:01Z
issue,133944,"torch.compile ""run at compile time"" decorator.",https://github.com/pytorch/pytorch/issues/133944,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: compile ux, module: dynamo, oncall: pt2",scan when time allows,2025-09-12T21:19:44Z
issue,156308,Composition of nested `torch.compile` calls is not well defined,https://github.com/pytorch/pytorch/issues/156308,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: compile ux, module: dynamo, oncall: pt2",scan when time allows,2025-09-12T20:11:51Z
issue,161886,Directly monkey patching torch.Tensor.__getitem__ breaks tensor construction,https://github.com/pytorch/pytorch/issues/161886,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python frontend,scan when time allows,2025-09-12T16:35:13Z
issue,136622,torch.onnx.errors.UnsupportedOperatorError: Exporting the operator 'aten::linalg_inv' to ONNX opset version 18 is not supported,https://github.com/pytorch/pytorch/issues/136622,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2025-09-12T14:03:48Z
issue,126881,[DCP] DCP does not support objects which are lazy initialized.,https://github.com/pytorch/pytorch/issues/126881,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed checkpointing,scan when time allows,2025-09-12T03:41:48Z
issue,90560,"[discussion, idea] Batched, vectorized base64 decoding / encoding + maybe RLE decoding / encoding",https://github.com/pytorch/pytorch/issues/90560,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nestedtensor, module: vision",scan when time allows,2025-09-12T01:20:45Z
issue,162711,`scatter` / `gather` index type unclear,https://github.com/pytorch/pytorch/issues/162711,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2025-09-11T22:04:48Z
issue,162712,FPE in torch.range,https://github.com/pytorch/pytorch/issues/162712,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: error checking,scan when time allows,2025-09-11T21:44:42Z
issue,127500,Avoid Having to Register Op For ExternKernelChoice of Aten Refs,https://github.com/pytorch/pytorch/issues/127500,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",scan when time allows,2025-09-11T17:34:03Z
issue,162352,Hiding C/C++ symbols in `version_script.lds`,https://github.com/pytorch/pytorch/issues/162352,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2025-09-11T15:39:54Z
issue,162362,the result is wrong for sdpa with sliding window attention,https://github.com/pytorch/pytorch/issues/162362,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: sdpa",scan when time allows,2025-09-11T15:36:15Z
issue,162482,[sparse] `torch._sparse_broadcast_to` creates false-positive coalesced tensors with inputs of shape `()`,https://github.com/pytorch/pytorch/issues/162482,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2025-09-11T15:34:52Z
issue,162582,calling _record_memory_history(enabled=None) before _dump_snapshot would clear the timestamp information,https://github.com/pytorch/pytorch/issues/162582,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: cudacachingallocator",scan when time allows,2025-09-11T15:33:53Z
issue,162685,Prevent py::class_ with no pointer type from being used on structs that inherit from intrusive_ptr_target,https://github.com/pytorch/pytorch/issues/162685,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: pybind,scan when time allows,2025-09-11T15:14:22Z
issue,162674,"NotImplementedError: ""addmm_sparse_cuda"" not implemented for 'Half'",https://github.com/pytorch/pytorch/issues/162674,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: half, module: sparse",scan when time allows,2025-09-11T14:52:17Z
issue,77737,"torchrun leads to `ModuleNotFoundError: No module named 'tensorboard'`, but python -m torch.distributed.launch is ok",https://github.com/pytorch/pytorch/issues/77737,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: r2p,scan when time allows,2025-09-11T07:04:50Z
issue,160697,[DTensor] nn.Embedding Compile - Empty Masked Buffer from AOT Autograd,https://github.com/pytorch/pytorch/issues/160697,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dtensor, module: dynamo, oncall: distributed",scan when time allows,2025-09-10T22:09:46Z
issue,160240,Migrate test.sh into python based tests,https://github.com/pytorch/pytorch/issues/160240,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2025-09-10T17:03:21Z
issue,152335,pin_memory crashes for big tensors and leaks page locked memory,https://github.com/pytorch/pytorch/issues/152335,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: memory usage,scan when time allows,2025-09-10T15:31:34Z
issue,162375,`torch.compile` fails with `pack_padded_sequence` + `pad_packed_sequence` (FakeTensor error),https://github.com/pytorch/pytorch/issues/162375,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",scan when time allows,2025-09-10T07:56:36Z
issue,145656,Stable C bindings for libtorch,https://github.com/pytorch/pytorch/issues/145656,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp,scan when time allows,2025-09-10T03:55:11Z
issue,162458,torch.float8_e5m2 unique() functionality support,https://github.com/pytorch/pytorch/issues/162458,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: floatx (formerly float8),scan when time allows,2025-09-09T21:01:37Z
issue,162182,CI jobs for x86_64 and x86_zen (AMD CPU) are overwriting each other,https://github.com/pytorch/pytorch/issues/162182,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ci, module: correctness (silent)",scan when time allows,2025-09-09T20:20:22Z
issue,141984,CI feature mirroring or caching for Docker Hub images,https://github.com/pytorch/pytorch/issues/141984,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2025-09-09T17:44:24Z
issue,141235,Enable CI and Build Support for PyTorch on PPC64LE Architecture,https://github.com/pytorch/pytorch/issues/141235,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: power",scan when time allows,2025-09-09T13:20:34Z
issue,113222,EPOCH_DEPRECATION_WARNING in ChainedScheduler.step,https://github.com/pytorch/pytorch/issues/113222,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: lrscheduler,scan when time allows,2025-09-09T02:47:47Z
issue,67958,The `SequentialLR` scheduler uses a deprecated pattern,https://github.com/pytorch/pytorch/issues/67958,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: lrscheduler, module: optimizer, module: tests",scan when time allows,2025-09-09T02:45:31Z
issue,162398,Unexpected variable type when parsing YAML in torchgen,https://github.com/pytorch/pytorch/issues/162398,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: codegen,scan when time allows,2025-09-08T23:19:13Z
issue,162280,libcufile symbol mismatch error in CUDA 13.0 libtorch build,https://github.com/pytorch/pytorch/issues/162280,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cuda",scan when time allows,2025-09-08T22:50:36Z
issue,101930,[feature request] [minor] Inplace torch.flip_,https://github.com/pytorch/pytorch/issues/101930,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python frontend,scan when time allows,2025-09-08T20:50:13Z
issue,162259,[Triton] Support multi-GPU with different types of GPUs (4090s + 5090s on same machine),https://github.com/pytorch/pytorch/issues/162259,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: pt2,scan when time allows,2025-09-08T19:07:46Z
issue,156846,Can't alter the learning rate though loading optimizer states,https://github.com/pytorch/pytorch/issues/156846,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: lrscheduler, module: optimizer",scan when time allows,2025-09-08T14:34:38Z
issue,152992,conv2d with int8 on CUDA: GET was unable to find an engine to execute this computation,https://github.com/pytorch/pytorch/issues/152992,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: cuda",scan when time allows,2025-09-08T14:17:20Z
issue,2569,Add a [strong] hash function for tensor data,https://github.com/pytorch/pytorch/issues/2569,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2025-09-08T05:33:39Z
issue,69687,Performance improvement in Autograd Forward AD using ZeroTensors,https://github.com/pytorch/pytorch/issues/69687,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: forward ad, module: performance",scan when time allows,2025-09-07T19:54:13Z
issue,161970,Confusion behavior of PyTorch when combining boolean indexing with integer indexing,https://github.com/pytorch/pytorch/issues/161970,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: advanced indexing, module: numpy",scan when time allows,2025-09-07T05:08:10Z
issue,160937,Module produced by export.unflatten is not runnable for Qwen3 and Llama3,https://github.com/pytorch/pytorch/issues/160937,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: oncall: export, oncall: pt2",scan when time allows,2025-09-06T04:43:33Z
issue,162283,"Failure after building torch with ""pip install -e . -v --no-build-isolation""",https://github.com/pytorch/pytorch/issues/162283,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: functorch",scan when time allows,2025-09-05T19:53:40Z
issue,157139,Distribute ARM Images,https://github.com/pytorch/pytorch/issues/157139,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: arm, module: binaries, module: cuda, module: docker",scan when time allows,2025-09-05T18:24:06Z
issue,24005,Using `torch.utils.checkpoint.checkpoint_sequential` and `torch.autograd.grad` breaks when used in combination with `DistributedDataParallel`,https://github.com/pytorch/pytorch/issues/24005,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: activation checkpointing, oncall: distributed",scan when time allows,2025-09-05T08:20:54Z
issue,162127,[CUDA] Need support for Cuda 13.0,https://github.com/pytorch/pytorch/issues/162127,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2025-09-04T23:44:10Z
issue,159667,"Non-tensor auxiliary types don't work in torch.func.{grad, value_and_grad}",https://github.com/pytorch/pytorch/issues/159667,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: functional ux, module: functorch",scan when time allows,2025-09-04T19:21:02Z
issue,162198,[Feature Request][Inductor] Support ops.store in ops.masked,https://github.com/pytorch/pytorch/issues/162198,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",scan when time allows,2025-09-04T18:53:30Z
issue,161621,evaluation forward() of Transformer model is slower than training forward() while autocast,https://github.com/pytorch/pytorch/issues/161621,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision), module: cuda, module: performance",scan when time allows,2025-09-04T18:39:28Z
issue,161937,AOTAutograd runtime alias reconstruction applies view on non‑contiguous base under torch.compile (inductor + DDP): “Use .reshape(...) instead”. pre_grad_custom_pass and joint_custom_post_pass can’t override,https://github.com/pytorch/pytorch/issues/161937,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: pt2-dispatcher, oncall: distributed, oncall: pt2",scan when time allows,2025-09-04T18:06:59Z
issue,161509,Update install commands in c++ Debugging,https://github.com/pytorch/pytorch/issues/161509,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2025-09-04T13:25:01Z
issue,161778,RuntimeError: ptrDesc->finalize() with conv1d,https://github.com/pytorch/pytorch/issues/161778,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: cuda",scan when time allows,2025-09-04T13:22:06Z
issue,161869,Add LazyGroupNorm,https://github.com/pytorch/pytorch/issues/161869,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: norms and normalization",scan when time allows,2025-09-04T13:19:38Z
issue,162004,UBSAN appears to be active for ATen but not c10,https://github.com/pytorch/pytorch/issues/162004,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tests,scan when time allows,2025-09-04T13:19:11Z
issue,139761,[RFC] PyTorch - PyPi PEP-759 proposal (wheel-next),https://github.com/pytorch/pytorch/issues/139761,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: binaries,scan when time allows,2025-09-03T15:41:24Z
issue,144931,"Error loading ""torch\lib\aoti_custom_ops.dll"" or one of its dependencies, when importing Torch, when building from Source on Windows 11 with cuDNN.",https://github.com/pytorch/pytorch/issues/144931,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: windows",scan when time allows,2025-09-03T03:19:40Z
issue,161960,WeightsUnpickler error: Trying to call reduce for unrecognized function <class 'code'>,https://github.com/pytorch/pytorch/issues/161960,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: serialization,scan when time allows,2025-09-02T23:31:27Z
issue,155332,compiler cache not work,https://github.com/pytorch/pytorch/issues/155332,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: pt2,scan when time allows,2025-09-02T15:18:12Z
issue,158793,[RFC]: Iterations of DeviceMesh based on recent user requests/feedback,https://github.com/pytorch/pytorch/issues/158793,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: devicemesh, oncall: distributed",scan when time allows,2025-09-02T13:21:10Z
issue,159015,[RFC] Support Fake PG mode for the device mesh and provide an option to reuse PG under the hood,https://github.com/pytorch/pytorch/issues/159015,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: devicemesh, oncall: distributed",scan when time allows,2025-09-02T13:21:08Z
issue,159014,[RFC] Make it easy to subclass DeviceMesh and also working in multi-threading setup,https://github.com/pytorch/pytorch/issues/159014,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: devicemesh, oncall: distributed",scan when time allows,2025-09-02T13:21:08Z
issue,159017,[RFC] Support the bookkeeping of different PGs beyond World_ and group_name,https://github.com/pytorch/pytorch/issues/159017,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: devicemesh, oncall: distributed",scan when time allows,2025-09-02T13:21:07Z
issue,159018,[RFC] Support more flexible pg option when it comes to submesh creation,https://github.com/pytorch/pytorch/issues/159018,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: devicemesh, oncall: distributed",scan when time allows,2025-09-02T13:21:06Z
issue,159019,[RFC] Support shorter print form for non-debugging mode,https://github.com/pytorch/pytorch/issues/159019,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: devicemesh, oncall: distributed",scan when time allows,2025-09-02T13:21:05Z
issue,155798,"`nn.RNN(...).to('cuda')` fails with `cuDNN error: CUDNN_STATUS_BAD_PARAM` on GPU, but works on CPU",https://github.com/pytorch/pytorch/issues/155798,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: cudnn",scan when time allows,2025-09-02T06:27:58Z
issue,161853,torch.nn.FractionalMaxPool3d occurs bug with big kernel_size,https://github.com/pytorch/pytorch/issues/161853,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: edge cases, module: error checking",scan when time allows,2025-09-01T14:01:57Z
issue,131196,Missing float8 storage,https://github.com/pytorch/pytorch/issues/131196,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: floatx (formerly float8),scan when time allows,2025-09-01T03:39:04Z
issue,142840,Document spectral norm non-support on non-diagonalizable inputs.,https://github.com/pytorch/pytorch/issues/142840,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: nn, module: nn.utils.parametrize",scan when time allows,2025-08-31T18:09:06Z
issue,159870,`torch.nn.functional.sigmoid` produces inconsistent results on different device for complex inputs,https://github.com/pytorch/pytorch/issues/159870,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: complex, module: cpu, module: nans and infs, module: nn",scan when time allows,2025-08-31T15:47:52Z
issue,119420,torch.stft sometimes raises RuntimeError: cuFFT error: CUFFT_INTERNAL_ERROR on low free memory,https://github.com/pytorch/pytorch/issues/119420,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: fft, module: memory usage",scan when time allows,2025-08-31T02:19:50Z
issue,63291,[typing] new_ones has wrong signature,https://github.com/pytorch/pytorch/issues/63291,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, module: typing",scan when time allows,2025-08-30T23:01:20Z
issue,63305,Alternate Model Loading - Android,https://github.com/pytorch/pytorch/issues/63305,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: android, oncall: mobile",scan when time allows,2025-08-29T14:27:31Z
issue,58393,"In Android, the memory used by the tensor or model cannot be recycled. Is there any way to solve it?",https://github.com/pytorch/pytorch/issues/58393,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: android, oncall: mobile",scan when time allows,2025-08-29T14:27:19Z
issue,129358,Pytorch build fail with GCC 14.1.0 due to third_party/fbgemm/src/UtilsAvx512.cc:970:35: error: ‘r’ may be used uninitialized [-Werror=maybe-uninitialized],https://github.com/pytorch/pytorch/issues/129358,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2025-08-29T12:01:51Z
issue,161283,Implement half-vs-float option for MPS output match tests,https://github.com/pytorch/pytorch/issues/161283,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: mps, module: tests",scan when time allows,2025-08-28T20:02:12Z
issue,114299,[RFC] Per-Parameter-Sharding FSDP,https://github.com/pytorch/pytorch/issues/114299,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fsdp,scan when time allows,2025-08-28T02:35:24Z
issue,147001,Preload CUDA fails if CUDA libs in different PYTHONPATH,https://github.com/pytorch/pytorch/issues/147001,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cuda, module: third_party",scan when time allows,2025-08-27T17:11:29Z
issue,161357,`reshape` support for sparse tensors (needed for `einsum`),https://github.com/pytorch/pytorch/issues/161357,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2025-08-27T16:20:37Z
issue,159906,New at::HostAllocator interface prevents using more than one allocator implementation for a device type,https://github.com/pytorch/pytorch/issues/159906,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cudacachingallocator,scan when time allows,2025-08-26T23:22:14Z
issue,161281,There should be an easy way in CI to disable individual failing inductor-periodic tests,https://github.com/pytorch/pytorch/issues/161281,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ci, oncall: pt2",scan when time allows,2025-08-26T20:16:40Z
issue,60466,PyTorch unfold could be faster,https://github.com/pytorch/pytorch/issues/60466,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: performance,scan when time allows,2025-08-26T19:17:21Z
issue,161108,AoTInductor gives segmentation fault with torchbind registration,https://github.com/pytorch/pytorch/issues/161108,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: aotinductor, module: torchbind, oncall: export, oncall: pt2",scan when time allows,2025-08-26T18:15:02Z
issue,161524,boolean index_add: Discrepancy between CPU and MPS backend,https://github.com/pytorch/pytorch/issues/161524,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: edge cases, module: mps",scan when time allows,2025-08-26T18:11:35Z
issue,161287,Add partial shape inference support when creating a `view()` of a Nested Jagged Tensor,https://github.com/pytorch/pytorch/issues/161287,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nestedtensor,scan when time allows,2025-08-26T13:18:59Z
issue,161163,Dropout performance degradation after version 2.6.0,https://github.com/pytorch/pytorch/issues/161163,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: intel, module: performance, module: random",scan when time allows,2025-08-26T07:13:29Z
issue,161296,Make `torch.func.functional_call` compatible with params and buffer input as `tensordict.TensorDict`,https://github.com/pytorch/pytorch/issues/161296,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: functorch,scan when time allows,2025-08-26T01:54:04Z
issue,161365,orch.onnx.errors.UnsupportedOperatorError: Exporting the operator 'aten::linalg_inv' to ONNX opset version 16 is not supported.,https://github.com/pytorch/pytorch/issues/161365,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2025-08-25T22:48:07Z
issue,131947,`__torch_function__` does not work for functions that called within other overrided functions,https://github.com/pytorch/pytorch/issues/131947,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: __torch_function__,scan when time allows,2025-08-24T08:17:58Z
issue,160071,[TorchDynamo] excessive stack use: stack is 3286 deep on A100 with AMD EPYC CPU,https://github.com/pytorch/pytorch/issues/160071,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",scan when time allows,2025-08-23T13:30:03Z
issue,93982,`linalg.lstsq` fails the gradient computation in forward mode,https://github.com/pytorch/pytorch/issues/93982,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: forward ad,scan when time allows,2025-08-23T09:52:20Z
issue,160508,`torch.compile` + `torch.vmap` + `torch.func.jacfwd` results in an error,https://github.com/pytorch/pytorch/issues/160508,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamo, module: functorch, module: vmap, oncall: pt2",scan when time allows,2025-08-23T05:48:59Z
issue,159914,`test_non_contiguous_input_mm_plus_mm` looks broken on B200,https://github.com/pytorch/pytorch/issues/159914,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: tf32",scan when time allows,2025-08-22T23:02:14Z
issue,108567,Allow slicing of Nested Tensors along constant dimensions,https://github.com/pytorch/pytorch/issues/108567,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nestedtensor,scan when time allows,2025-08-22T19:17:04Z
issue,99556,torch.func.jacrev fails if model contains full_backward_hook,https://github.com/pytorch/pytorch/issues/99556,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: functorch",scan when time allows,2025-08-22T14:43:15Z
issue,143466,Support Dict Parameter Type for custom_op,https://github.com/pytorch/pytorch/issues/143466,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: custom-operators, module: pt2-dispatcher, oncall: pt2",scan when time allows,2025-08-22T13:40:07Z
issue,126913,Alternative access order for the same buffer can bring big perf win,https://github.com/pytorch/pytorch/issues/126913,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",scan when time allows,2025-08-22T13:17:24Z
issue,160359,Use `tf32x3` through Inductor,https://github.com/pytorch/pytorch/issues/160359,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",scan when time allows,2025-08-22T04:04:52Z
issue,161068,[ATen] Linkage error with TensorBase::data_ptr<T> and TensorBase::mutable_data_ptr<T>,https://github.com/pytorch/pytorch/issues/161068,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp-extensions,scan when time allows,2025-08-21T03:38:10Z
issue,160891,Please add `torch._grouped_mm` on SM_120 (RTX 6000 Blackwell) hardware,https://github.com/pytorch/pytorch/issues/160891,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2025-08-20T22:47:35Z
issue,160629,torch.arange not implemented for complex numbers,https://github.com/pytorch/pytorch/issues/160629,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: complex,scan when time allows,2025-08-20T19:51:40Z
issue,161029,MPS does not support deterministic index operations,https://github.com/pytorch/pytorch/issues/161029,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: determinism, module: mps",scan when time allows,2025-08-20T02:36:16Z
issue,161001,Add some way to associate functional custom operators and their in-place out variants,https://github.com/pytorch/pytorch/issues/161001,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: custom-operators, module: inductor, module: pt2-dispatcher, oncall: pt2",scan when time allows,2025-08-20T02:33:44Z
issue,160856,Discrepancy between PyTorch and TensorRT when casting from FP32 to BF16 causes significant accuracy mismatch,https://github.com/pytorch/pytorch/issues/160856,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: edge cases, module: onnx",scan when time allows,2025-08-19T20:39:49Z
issue,160167,"libtorch: CMake 4.1: Error evaluating generator expression: TARGET_PROPERTY:torch,INTERFACE_LINK_LIBRARIES",https://github.com/pytorch/pytorch/issues/160167,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cpp-extensions, module: windows",scan when time allows,2025-08-19T17:58:01Z
issue,157627,Regression: torch.distributed.gather_object segfaults,https://github.com/pytorch/pytorch/issues/157627,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2025-08-19T13:28:48Z
issue,125070,2.2.0+ regresses SDPA performance on Windows,https://github.com/pytorch/pytorch/issues/125070,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: sdpa, module: windows",scan when time allows,2025-08-19T11:41:20Z
issue,158444,Proposal to introduce CUDA streams in BatchLinearAlgebraLib.cpp,https://github.com/pytorch/pytorch/issues/158444,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: linear algebra",scan when time allows,2025-08-19T07:26:09Z
issue,160337,"[RFC] Simplify bookkeeping of DeviceMesh slicing, (un)flattening, ...",https://github.com/pytorch/pytorch/issues/160337,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: devicemesh, oncall: distributed",scan when time allows,2025-08-18T21:02:17Z
issue,159769,Inductor codegen for float8 dynamic quantization ops for scaled_grouped_mm backward pass is slow,https://github.com/pytorch/pytorch/issues/159769,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",scan when time allows,2025-08-18T19:39:25Z
issue,145471,Is there a PyTorch version that can work properly on the Thor platform based on the Blackwell architecture?,https://github.com/pytorch/pytorch/issues/145471,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2025-08-18T07:31:00Z
issue,160517,Enhance output layouts of all-sparse matrix multiplications,https://github.com/pytorch/pytorch/issues/160517,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2025-08-16T11:34:08Z
issue,156873,[ONNX] Missing repeat_interleave for dynamo=True,https://github.com/pytorch/pytorch/issues/156873,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2025-08-16T00:38:23Z
issue,156745,[v.2.8.0] Release Tracker,https://github.com/pytorch/pytorch/issues/156745,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2025-08-14T17:56:02Z
issue,160405,[Expandable block] how to get the best-fit free block,https://github.com/pytorch/pytorch/issues/160405,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cudacachingallocator,scan when time allows,2025-08-14T05:24:01Z
issue,159781,MacOS Python 3.13t wheel is build for macOS 14.0+ ARM64 while Py3.9-3.13 wheels are for macOS 11.0+ ARM64,https://github.com/pytorch/pytorch/issues/159781,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: macos, oncall: releng",scan when time allows,2025-08-13T18:24:11Z
issue,153919,FPE in `torch.remainder`,https://github.com/pytorch/pytorch/issues/153919,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: edge cases, module: python frontend",scan when time allows,2025-08-13T18:00:16Z
issue,128323,UnsupportedOperatorError: Exporting the operator 'quantized::batch_norm2d' to ONNX opset version 14 is not supported.,https://github.com/pytorch/pytorch/issues/128323,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2025-08-13T10:15:19Z
issue,159784,Immutable assignment (akin to `array.at[...]` in JAX),https://github.com/pytorch/pytorch/issues/159784,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: functionalization, module: functorch, module: pt2-dispatcher, oncall: pt2",scan when time allows,2025-08-13T07:15:51Z
issue,159947,Add official statically linked libtorch libraries,https://github.com/pytorch/pytorch/issues/159947,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: docs, module: nativert",scan when time allows,2025-08-13T07:06:24Z
issue,117884,Foreach Where support,https://github.com/pytorch/pytorch/issues/117884,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: mta, module: optimizer, module: pt2 optimizer, oncall: pt2",scan when time allows,2025-08-13T03:12:20Z
issue,160308,Reserved memory is much bigger than allocated memory in multi-stream scenario,https://github.com/pytorch/pytorch/issues/160308,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: cudacachingallocator",scan when time allows,2025-08-13T02:53:47Z
issue,159932,Cannot view dtype of tensor which is contiguous due to numel=1 but has non-standard striding,https://github.com/pytorch/pytorch/issues/159932,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2025-08-13T02:52:38Z
issue,160387,DISABLED RecordDebugHandles.Basic (__main__.test_jit),https://github.com/pytorch/pytorch/issues/160387,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: flaky-tests, module: unknown",scan when time allows,2025-08-13T02:48:03Z
issue,160122,Add tensor-aware put_signal & wait_signal Triton wrappers to NVSHMEM Triton Kernels,https://github.com/pytorch/pytorch/issues/160122,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2025-08-13T02:47:28Z
issue,159866,BE: enable auto reformat List to list and similar typing,https://github.com/pytorch/pytorch/issues/159866,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ci, module: lint",scan when time allows,2025-08-12T20:26:31Z
issue,159680,[Benchmark Dashboard] Kernel-level benchmark,https://github.com/pytorch/pytorch/issues/159680,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: benchmark,scan when time allows,2025-08-12T20:15:59Z
issue,159676,[Benchmark Dashboard] Track quarterly change,https://github.com/pytorch/pytorch/issues/159676,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: benchmark,scan when time allows,2025-08-12T20:14:54Z
issue,152134,[RFC] Proposed Changes to Feature Tracking & Classification for PyTorch Releases starting Release 2.8,https://github.com/pytorch/pytorch/issues/152134,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2025-08-12T19:35:59Z
issue,160415,[MPS] Conv3d type mismatch with AMP,https://github.com/pytorch/pytorch/issues/160415,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision), module: mps",scan when time allows,2025-08-12T19:10:28Z
issue,155011,Please support libtorch for XPU,https://github.com/pytorch/pytorch/issues/155011,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: xpu",scan when time allows,2025-08-12T06:37:25Z
issue,72775,[vulkan] Vulkan backend fails creating tensor on x86_64 Linux,https://github.com/pytorch/pytorch/issues/72775,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: vulkan,scan when time allows,2025-08-11T22:50:50Z
issue,102319,Re-enable Low Memory Dropout,https://github.com/pytorch/pytorch/issues/102319,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",scan when time allows,2025-08-11T16:53:28Z
issue,160142,"Add a ""max-performance"" mode to torch.compile for aggressive optimization",https://github.com/pytorch/pytorch/issues/160142,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",scan when time allows,2025-08-11T15:37:35Z
issue,159942,Is is possible for pytorch pipeline parallelism to support dynamic shapes(input/output),https://github.com/pytorch/pytorch/issues/159942,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamic shapes, oncall: distributed, oncall: pt2",scan when time allows,2025-08-11T15:34:54Z
issue,160281,Improve CUDAGraph Tree heuristics on starting new generation,https://github.com/pytorch/pytorch/issues/160281,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda graphs, module: inductor, oncall: pt2",scan when time allows,2025-08-10T22:06:27Z
issue,93859,Bitwise-perfect method for (de)serializing tensors in base64,https://github.com/pytorch/pytorch/issues/93859,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: serialization,scan when time allows,2025-08-10T18:53:10Z
issue,60832,State of Torch Named Tensors,https://github.com/pytorch/pytorch/issues/60832,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: named tensor,scan when time allows,2025-08-10T13:24:29Z
issue,119194,torch_function objects passed as non-Tensor args should trigger overrides,https://github.com/pytorch/pytorch/issues/119194,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: __torch_function__, module: dynamic shapes",scan when time allows,2025-08-10T13:23:18Z
issue,156959,Gross mismatch in PDF between CUDA and CPU for multivariate Gaussian mixture models,https://github.com/pytorch/pytorch/issues/156959,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: cuda, module: distributions",scan when time allows,2025-08-09T05:25:38Z
issue,68959,Deterministic implementation for grid_sampler_2d_backward_cuda,https://github.com/pytorch/pytorch/issues/68959,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: determinism",scan when time allows,2025-08-08T22:13:24Z
issue,160225,Building cuda kernels with debug information causes some cuda kernel launches to fail due to capacity constraints,https://github.com/pytorch/pytorch/issues/160225,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cuda, module: debug-build, module: third_party",scan when time allows,2025-08-08T21:54:45Z
issue,84530,`tensordot` not working for dtype int32 and lower when there is only 1 element in the given axis,https://github.com/pytorch/pytorch/issues/84530,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: linear algebra,scan when time allows,2025-08-08T08:40:20Z
issue,101188,Weird dataloader performance degradation caused by torch and numpy import order,https://github.com/pytorch/pytorch/issues/101188,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: openmp,scan when time allows,2025-08-08T05:35:00Z
issue,137389,non-deterministic issue of torch.einsum function on different GPU.,https://github.com/pytorch/pytorch/issues/137389,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cublas, module: numerical-reproducibility",scan when time allows,2025-08-07T21:49:21Z
issue,159852,`torch.cond()` behaves inconsistently when using symbolic predicate,https://github.com/pytorch/pytorch/issues/159852,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: higher order operators, module: pt2-dispatcher, oncall: pt2",scan when time allows,2025-08-07T17:03:06Z
issue,159848,"`tree_flatten(some_pytree, lambda x: isinstance(x, Tensor))` yields `[None], _`",https://github.com/pytorch/pytorch/issues/159848,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: pytree,scan when time allows,2025-08-07T13:57:50Z
issue,57185,Segmentation fault with ITIMER_REAL,https://github.com/pytorch/pytorch/issues/57185,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: macos,scan when time allows,2025-08-07T05:09:56Z
issue,121969,"which Pythorch suports ""TORCH_USE_CUDA_DSA=1"" from the shell environment?",https://github.com/pytorch/pytorch/issues/121969,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2025-08-06T20:44:03Z
issue,159622,Performance bug on `inplace` of `nn.ELU`,https://github.com/pytorch/pytorch/issues/159622,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: memory usage, module: nn",scan when time allows,2025-08-06T19:19:14Z
issue,138831,Add mean and var operation for Nested Tensors,https://github.com/pytorch/pytorch/issues/138831,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nestedtensor,scan when time allows,2025-08-06T17:40:31Z
issue,157950,[RFC]: PyTorch Low-Precision GEMMs Public API,https://github.com/pytorch/pytorch/issues/157950,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: floatx (formerly float8), module: rocm",scan when time allows,2025-08-06T17:39:27Z
issue,159886,torch._dynamo documentation,https://github.com/pytorch/pytorch/issues/159886,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: dynamo, oncall: pt2",scan when time allows,2025-08-06T15:10:43Z
issue,159838,C++23 forward declaration error :`invalid application of 'sizeof' to an incomplete type 'torch::_export::Graph'`,https://github.com/pytorch/pytorch/issues/159838,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: nativert",scan when time allows,2025-08-06T14:13:32Z
issue,158783,[BE] linter to detect unused docker images,https://github.com/pytorch/pytorch/issues/158783,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: lint,scan when time allows,2025-08-05T20:08:22Z
issue,153238,[NJT] can only chunk if the 2nd dimension is ragged,https://github.com/pytorch/pytorch/issues/153238,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nestedtensor,scan when time allows,2025-08-05T18:37:20Z
issue,159739,torch.empty should consider np.bool while parsing args,https://github.com/pytorch/pytorch/issues/159739,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python frontend,scan when time allows,2025-08-05T18:09:56Z
issue,145094,Tracking issue: Incorrect Meta Strides / Turn On PyDispatcher in FakeTensor Mode,https://github.com/pytorch/pytorch/issues/145094,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: decompositions, module: faketensor, module: pt2-dispatcher, oncall: pt2",scan when time allows,2025-08-05T16:46:30Z
issue,157445,DDP+TP composition does not work as expected,https://github.com/pytorch/pytorch/issues/157445,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2025-08-05T14:18:47Z
issue,155997,"ONNX export via Dynamo sets `dft_length = 1` in `DFT`, breaking shape-inference for `torch.fft.rfft`",https://github.com/pytorch/pytorch/issues/155997,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2025-08-05T09:29:11Z
issue,16375,Python-bound C++ frontend modules don't handle attributes well,https://github.com/pytorch/pytorch/issues/16375,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp,scan when time allows,2025-08-05T00:03:30Z
issue,117394,[RFC] Dynamo Single Step Graph,https://github.com/pytorch/pytorch/issues/117394,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, oncall: pt2",scan when time allows,2025-08-04T23:21:04Z
issue,127973,Torch compile OpenMP error with simple nn.Module,https://github.com/pytorch/pytorch/issues/127973,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: macos, module: openmp, oncall: pt2",scan when time allows,2025-08-04T19:41:52Z
issue,124749,Improve Dynamo (and other) flaky tests,https://github.com/pytorch/pytorch/issues/124749,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamo, module: tests, oncall: pt2",scan when time allows,2025-08-04T18:39:20Z
issue,128071,[user empathy day 2][based] torch.compile issues,https://github.com/pytorch/pytorch/issues/128071,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: compile-time, module: dynamo, oncall: pt2",scan when time allows,2025-08-04T18:32:03Z
issue,159301,#158878 seems to break `test_mempool_empty_cache`,https://github.com/pytorch/pytorch/issues/159301,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: tests",scan when time allows,2025-08-04T17:59:54Z
issue,115526,Can PyTorch support using custom compilers for model's components and the Inductor compiler for the model?,https://github.com/pytorch/pytorch/issues/115526,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",scan when time allows,2025-08-04T16:46:21Z
issue,159564,Re-enable inactive_split_bytes when expandable_segments:True,https://github.com/pytorch/pytorch/issues/159564,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: cudacachingallocator",scan when time allows,2025-08-04T15:51:07Z
issue,153933,PyTorch build fail with GCC 15.1.0 due to third_party/gloo/gloo/types.h:66:11: error: ‘uint8_t’ does not name a type,https://github.com/pytorch/pytorch/issues/153933,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2025-08-04T14:58:50Z
issue,159740,Better error message when requesting pinned memory on cpu-only system,https://github.com/pytorch/pytorch/issues/159740,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: error checking,scan when time allows,2025-08-04T01:40:19Z
issue,142290,RFC: Display Supported Versions in API Documentation#,https://github.com/pytorch/pytorch/issues/142290,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2025-08-02T05:14:49Z
issue,141210,Addmm is not always better than add + mm,https://github.com/pytorch/pytorch/issues/141210,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cublas, module: dependency bug, module: performance",scan when time allows,2025-08-02T04:15:08Z
issue,159559,Support NCCL register userbuffer for P2P communication in NCCL backend,https://github.com/pytorch/pytorch/issues/159559,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nccl, oncall: distributed",scan when time allows,2025-08-02T02:12:31Z
issue,129668,TensorDict - torch.compile issue tracker,https://github.com/pytorch/pytorch/issues/129668,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",scan when time allows,2025-08-01T23:34:43Z
issue,112901,Tracking: Dynamo Tracing Improvements,https://github.com/pytorch/pytorch/issues/112901,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",scan when time allows,2025-08-01T21:38:32Z
issue,158187,"[torch.save] Integrate FastPersist by allowing f (a.k.a., file-like object) to save storage objects",https://github.com/pytorch/pytorch/issues/158187,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: serialization,scan when time allows,2025-08-01T15:23:08Z
issue,105859,Report model flop utilization (mfu) in benchmark,https://github.com/pytorch/pytorch/issues/105859,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: pt2,scan when time allows,2025-08-01T02:44:57Z
issue,106269,Make our source attribution debug prints more useful for Compiler Explorer,https://github.com/pytorch/pytorch/issues/106269,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: pt2,scan when time allows,2025-08-01T02:44:46Z
issue,116575,Reland guarded config,https://github.com/pytorch/pytorch/issues/116575,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",scan when time allows,2025-08-01T02:42:30Z
issue,102853,pdb but for dynamo (and time travel debugging),https://github.com/pytorch/pytorch/issues/102853,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: pt2,scan when time allows,2025-08-01T02:39:21Z
issue,159536,Document NCCL Symmetric memory,https://github.com/pytorch/pytorch/issues/159536,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: nccl",scan when time allows,2025-07-31T23:17:36Z
issue,31837,logsumexp: two little-impact perf suggestions (important because logsumexp is used for optimizing/fusing cross_entropy over large vocabs),https://github.com/pytorch/pytorch/issues/31837,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: performance, module: reductions",scan when time allows,2025-07-31T21:47:45Z
issue,159510,Add more tests to h100 ciflow,https://github.com/pytorch/pytorch/issues/159510,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: tests, oncall: pt2",scan when time allows,2025-07-31T17:17:21Z
issue,153745,[RFC] [Feature] Intra-Device Heterogeneous Memory Allocation Support,https://github.com/pytorch/pytorch/issues/153745,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2025-07-31T16:19:57Z
issue,156694,"Vendored wheels on PyTorch pip repository are outdated (e.g., `cmake`, `certifi`)",https://github.com/pytorch/pytorch/issues/156694,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, oncall: releng",scan when time allows,2025-07-31T07:44:54Z
issue,159354,distributed/rpc/test_share_memory fails with timeouts on 64bit BAR systems,https://github.com/pytorch/pytorch/issues/159354,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2025-07-31T07:06:23Z
issue,157606,Both DTensor TP and SP are missing the last collective in the backward pass,https://github.com/pytorch/pytorch/issues/157606,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2025-07-30T23:10:51Z
issue,159498,[FSDP2] provide profiler metadata for memory bandwidth calculation,https://github.com/pytorch/pytorch/issues/159498,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: fsdp, oncall: distributed",scan when time allows,2025-07-30T18:39:52Z
issue,159353,torch.cond triggers shape mismatch with linear layer when nested in custom model,https://github.com/pytorch/pytorch/issues/159353,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: higher order operators, oncall: pt2",scan when time allows,2025-07-30T18:11:12Z
issue,159384,Graph partition performance,https://github.com/pytorch/pytorch/issues/159384,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda graphs, module: inductor, oncall: pt2",scan when time allows,2025-07-30T17:20:40Z
issue,159124,AOTInductor preallocate output tensors,https://github.com/pytorch/pytorch/issues/159124,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: aotinductor, oncall: export, oncall: pt2",scan when time allows,2025-07-30T15:39:25Z
issue,159333,torch._foreach_copy_() copy all nan as same number,https://github.com/pytorch/pytorch/issues/159333,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: edge cases, module: mta",scan when time allows,2025-07-30T15:30:58Z
issue,106885,"[bug] Dynamo graph break when using pyton module `heapq` (manipulates with `list`s), although succeeds when placing `heapq.py` near the test script",https://github.com/pytorch/pytorch/issues/106885,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2025-07-30T13:52:29Z
issue,159149,"❌ Error generating speech: NVML_SUCCESS == r INTERNAL ASSERT FAILED at ""../c10/cuda/CUDACachingAllocator.cpp"":995, please report a bug to PyTorch.",https://github.com/pytorch/pytorch/issues/159149,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cudacachingallocator,scan when time allows,2025-07-30T13:26:17Z
issue,159363,pad_sequence overflows when padding with larger value than int64.max - 512,https://github.com/pytorch/pytorch/issues/159363,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: edge cases, module: python frontend",scan when time allows,2025-07-30T13:24:01Z
issue,158895,Update docs about running lint,https://github.com/pytorch/pytorch/issues/158895,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: lint",scan when time allows,2025-07-29T20:33:26Z
issue,158876,UNSTABLE pull / linux-jammy-py3_9-clang9-xla / test (xla),https://github.com/pytorch/pytorch/issues/158876,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ci, module: xla",scan when time allows,2025-07-29T20:30:26Z
issue,157379,[CI] Need better way to detect OOMs especially on pet instances,https://github.com/pytorch/pytorch/issues/157379,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ci, module: flaky-tests",scan when time allows,2025-07-29T20:05:38Z
issue,81565,__getitem__ is returned as an OverloadPacket instead of an OpOverload in __torch_dispatch__,https://github.com/pytorch/pytorch/issues/81565,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: __torch_dispatch__,scan when time allows,2025-07-29T14:19:25Z
issue,81717,PyTorch Embedding Op with max_norm is not working as expected,https://github.com/pytorch/pytorch/issues/81717,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: embedding, module: norms and normalization",scan when time allows,2025-07-29T14:19:24Z
issue,91251,Potential bug found with pybind11 dec_ref while gil released,https://github.com/pytorch/pytorch/issues/91251,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: pybind,scan when time allows,2025-07-29T14:19:21Z
issue,91508,Update map_nt to take into account size and strides,https://github.com/pytorch/pytorch/issues/91508,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nestedtensor,scan when time allows,2025-07-29T14:19:20Z
issue,93454,MMDet 3.x cannot run successfully in inductor mode,https://github.com/pytorch/pytorch/issues/93454,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2025-07-29T14:19:16Z
issue,93455,Performance regression on interpolation in Kornia,https://github.com/pytorch/pytorch/issues/93455,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2025-07-29T14:19:15Z
issue,93456,Way to run accuracy minifier on only one particular subgraph,https://github.com/pytorch/pytorch/issues/93456,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2025-07-29T14:19:14Z
issue,93463,Traceable tensor subclasses cannot actually be used with AOTAutograd,https://github.com/pytorch/pytorch/issues/93463,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2025-07-29T14:19:13Z
issue,93464,wav2vec2 model: error trying to do inference,https://github.com/pytorch/pytorch/issues/93464,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2025-07-29T14:19:12Z
issue,93465,Graph breaks with HuggingFace Stable Diffusion,https://github.com/pytorch/pytorch/issues/93465,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2025-07-29T14:19:11Z
issue,93480,Umbrella issue for only populate real_value_cache in export test suite fails,https://github.com/pytorch/pytorch/issues/93480,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",scan when time allows,2025-07-29T14:19:09Z
issue,93486,A Simple Function Causing Graph Break,https://github.com/pytorch/pytorch/issues/93486,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2025-07-29T14:19:08Z
issue,93488,"If minifier test fails, stderr/stdout of subprocess calls is not printed",https://github.com/pytorch/pytorch/issues/93488,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2025-07-29T14:19:07Z
issue,93580,AssertionError: Unknown expression s2,https://github.com/pytorch/pytorch/issues/93580,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2025-07-29T14:19:06Z
issue,93511,support setattr of arbitrary user provided types in tracing,https://github.com/pytorch/pytorch/issues/93511,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",scan when time allows,2025-07-29T14:19:06Z
issue,93592,Turning on minifier causes bug to go away (on DebertaForMaskedLM),https://github.com/pytorch/pytorch/issues/93592,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2025-07-29T14:19:05Z
issue,93596,Minifier doesn't work on DebertaForQuestionAnswering,https://github.com/pytorch/pytorch/issues/93596,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2025-07-29T14:19:04Z
issue,93602,Diffuser pipeline device attribute broken when using optimized model,https://github.com/pytorch/pytorch/issues/93602,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2025-07-29T14:19:03Z
issue,93604,TorchBench - moco - RuntimeError: Tensors must be CUDA and dense,https://github.com/pytorch/pytorch/issues/93604,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2025-07-29T14:19:02Z
issue,93609,Don't store example_value on FX node meta,https://github.com/pytorch/pytorch/issues/93609,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2025-07-29T14:19:01Z
issue,94183,Addition of batch CSR tensors produces incorrect and invalid CSR tensor,https://github.com/pytorch/pytorch/issues/94183,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: sparse",scan when time allows,2025-07-29T14:19:00Z
issue,94186,Addition of hybrid CSR tensors produces incorrect and invalid CSR tensor,https://github.com/pytorch/pytorch/issues/94186,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: sparse",scan when time allows,2025-07-29T14:18:59Z
issue,97268,Desync debugger encounters traceMap error,https://github.com/pytorch/pytorch/issues/97268,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: c10d, oncall: distributed",scan when time allows,2025-07-29T14:18:58Z
issue,99999,Runtime Error,https://github.com/pytorch/pytorch/issues/99999,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2025-07-29T14:18:56Z
issue,101314,Shared library loading logic breaks when CUDA packages are installed in a non-standard location,https://github.com/pytorch/pytorch/issues/101314,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: bazel,scan when time allows,2025-07-29T14:18:54Z
issue,102673,Fix dynamo-related debug Python 3.11 failures,https://github.com/pytorch/pytorch/issues/102673,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python version,scan when time allows,2025-07-29T14:18:53Z
issue,103800,Mis-annotated return for `F._no_grad_embedding_renorm_` (also JIT related),https://github.com/pytorch/pytorch/issues/103800,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: typing, oncall: jit",scan when time allows,2025-07-29T14:18:52Z
issue,107381,sparse_mask method ignores masked-in elements of sparse compressed input tensors,https://github.com/pytorch/pytorch/issues/107381,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: sparse",scan when time allows,2025-07-29T14:18:51Z
issue,107451,Conversion from COO with two sparse dimensions to CSR with dense_dim specified fails,https://github.com/pytorch/pytorch/issues/107451,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2025-07-29T14:18:50Z
issue,107478,Adding batched CSR tensors with different sparsities produces an invalid tensor,https://github.com/pytorch/pytorch/issues/107478,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: sparse",scan when time allows,2025-07-29T14:18:49Z
issue,108984,PPC64le: GCC 11.2.1 Linker Error in bin/torch_shm_manager,https://github.com/pytorch/pytorch/issues/108984,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2025-07-29T14:18:48Z
issue,111962,[pytree] `pytree.tree_map` does not respect type of `torch.Size`,https://github.com/pytorch/pytorch/issues/111962,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: pytree,scan when time allows,2025-07-29T14:18:47Z
issue,114859,PT2 QAT flow fails to get reasonable accuracy with mobilenet_v3_large,https://github.com/pytorch/pytorch/issues/114859,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: pt2 accuracy, oncall: pt2, oncall: quantization",scan when time allows,2025-07-29T14:18:45Z
issue,116013,[Dynamo] bytecode transformed by Dynamo is not serializable by marshal,https://github.com/pytorch/pytorch/issues/116013,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",scan when time allows,2025-07-29T14:18:44Z
issue,118642,Different outputs of `div` and `threshold` with a specific input,https://github.com/pytorch/pytorch/issues/118642,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: decompositions, oncall: pt2",scan when time allows,2025-07-29T14:18:43Z
issue,121220,Misleading `log_abs_det_jacobian` shape for `StackTransform`?,https://github.com/pytorch/pytorch/issues/121220,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distributions,scan when time allows,2025-07-29T14:18:40Z
issue,121439,copy_() produces wrong results for boolean tensors on mps,https://github.com/pytorch/pytorch/issues/121439,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: mps",scan when time allows,2025-07-29T14:18:39Z
issue,121725,torch.logsumexp throws RuntimeError instead of reducing all dimensions when dim=None,https://github.com/pytorch/pytorch/issues/121725,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: named tensor, module: python frontend",scan when time allows,2025-07-29T14:18:37Z
issue,122529,Distributed checkpoint state_dict load may report nccl error and hide the real root-cause exception,https://github.com/pytorch/pytorch/issues/122529,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: oncall: distributed, oncall: distributed checkpointing",scan when time allows,2025-07-29T14:18:35Z
issue,133010,torch.multiprocessing.start_processes is blocking with large input arguments,https://github.com/pytorch/pytorch/issues/133010,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: multiprocessing,scan when time allows,2025-07-29T14:18:31Z
issue,144530,[Pipelining] PP+DDP does not work for Zero Bubble,https://github.com/pytorch/pytorch/issues/144530,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: pipelining, oncall: distributed",scan when time allows,2025-07-29T14:18:26Z
issue,146945,New CachingAutotuner pickling logic may be brittle to triton upgrades,https://github.com/pytorch/pytorch/issues/146945,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2025-07-29T14:18:23Z
issue,147814,unbind_copy opinformation cause exception while running test_dtensor_ops.py,https://github.com/pytorch/pytorch/issues/147814,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dtensor, oncall: distributed",scan when time allows,2025-07-29T14:18:19Z
issue,148106,NotImplementedError: FlexAttentionHigherOrderVariable() has no type,https://github.com/pytorch/pytorch/issues/148106,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamo, module: flex attention, module: higher order operators, module: pt2-dispatcher",scan when time allows,2025-07-29T14:18:18Z
issue,152190,HUD Dashboard sort by perf speedup doesn't do anything,https://github.com/pytorch/pytorch/issues/152190,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: devx,scan when time allows,2025-07-29T14:18:14Z
issue,157054,NVFp4 Cublas Error,https://github.com/pytorch/pytorch/issues/157054,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cublas, module: cuda, module: floatx (formerly float8)",scan when time allows,2025-07-29T14:18:10Z
issue,159138,Some bugs that related to `ReflectionPadNd` on input dim and padding,https://github.com/pytorch/pytorch/issues/159138,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2025-07-29T14:18:09Z
issue,159164,Improvements to CUDACachingAllocator When expandable_segments Is Enabled,https://github.com/pytorch/pytorch/issues/159164,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cudacachingallocator,scan when time allows,2025-07-29T03:01:27Z
issue,159234,Lingering reference to setup.py in CONTRIBUTING.md,https://github.com/pytorch/pytorch/issues/159234,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2025-07-28T22:06:56Z
issue,153534,`pca_lowrank` on CUDA is significantly slower than on CPU,https://github.com/pytorch/pytorch/issues/153534,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: linear algebra",scan when time allows,2025-07-28T15:56:39Z
issue,159122,CI lint unexpected failure,https://github.com/pytorch/pytorch/issues/159122,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: lint, module: typing",scan when time allows,2025-07-28T14:15:17Z
issue,73656,torchrun: Hostname/endpoint mismatch not handled,https://github.com/pytorch/pytorch/issues/73656,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: c10d,scan when time allows,2025-07-28T10:39:47Z
issue,150186,`MaskedTensor` 10x slower than `Tensor` compared with `nn.functional.softmax`,https://github.com/pytorch/pytorch/issues/150186,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: masked operators, module: performance",scan when time allows,2025-07-28T10:27:22Z
issue,38671,Add CUDA callback to Python API,https://github.com/pytorch/pytorch/issues/38671,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: rpc",scan when time allows,2025-07-28T10:06:55Z
issue,159230,[Precompile] Investigate recompile in NanoGPT training,https://github.com/pytorch/pytorch/issues/159230,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2025-07-27T20:33:03Z
issue,159229,[Precompile] Assertion error when loading package in Nanogpt training with caching precompile fails after second load,https://github.com/pytorch/pytorch/issues/159229,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2025-07-27T20:24:23Z
issue,159228,[Precompile] mark_static_address behavior,https://github.com/pytorch/pytorch/issues/159228,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2025-07-27T20:10:32Z
issue,61582,Many reduction operators do not support reducing over multiple dimensions,https://github.com/pytorch/pytorch/issues/61582,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: numpy, module: python array api, module: reductions",scan when time allows,2025-07-27T16:26:16Z
issue,159156,Fully remove the protoc dependency in core,https://github.com/pytorch/pytorch/issues/159156,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: onnx, module: third_party",scan when time allows,2025-07-25T17:14:54Z
issue,158738,searchsorted handling of nan,https://github.com/pytorch/pytorch/issues/158738,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python frontend,scan when time allows,2025-07-25T09:56:38Z
issue,158873,Erroneous Recompile Warnings with Multiple Models,https://github.com/pytorch/pytorch/issues/158873,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: compile ux, module: dynamo, oncall: pt2",scan when time allows,2025-07-24T20:37:01Z
issue,158969,Scaled dot product attention accepts head_num for attn_mask,https://github.com/pytorch/pytorch/issues/158969,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sdpa,scan when time allows,2025-07-24T19:33:14Z
issue,158199,GPT-2 compiled model perf/accuracy,https://github.com/pytorch/pytorch/issues/158199,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: mps, module: performance, oncall: pt2",scan when time allows,2025-07-24T16:26:34Z
issue,157919,tighten pt2 FX graph invariants,https://github.com/pytorch/pytorch/issues/157919,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: higher order operators, module: pt2-dispatcher, oncall: pt2",scan when time allows,2025-07-24T14:07:37Z
issue,158936,TorchAO FP8 training + HF transformer model + torch.compile + DDP does not work,https://github.com/pytorch/pytorch/issues/158936,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ddp, module: floatx (formerly float8), oncall: distributed, oncall: pt2",scan when time allows,2025-07-24T01:04:25Z
issue,155325,Excessively restrictive dependencies,https://github.com/pytorch/pytorch/issues/155325,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, oncall: releng",scan when time allows,2025-07-23T22:23:36Z
issue,155066,Add extensions `flash_attention` and `vllm` as test of new PyTorch releases for known issues of compat of their binaries and of possibility of compiling these from source,https://github.com/pytorch/pytorch/issues/155066,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: releng,scan when time allows,2025-07-23T22:23:32Z
issue,158871,torch.classes module doesn't set all module attributes,https://github.com/pytorch/pytorch/issues/158871,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python frontend,scan when time allows,2025-07-23T20:08:12Z
issue,51511,Python-3.9: Importing numba and torch triggers internal error,https://github.com/pytorch/pytorch/issues/51511,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2025-07-23T20:07:46Z
issue,158819,[FSDP2] return original unsharded nn.Module from FSDPModule,https://github.com/pytorch/pytorch/issues/158819,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: fsdp, oncall: distributed",scan when time allows,2025-07-23T15:20:17Z
issue,158401,torch.compile hits internal tuple index out of range inside subclass_meta.unwrapped_idx,https://github.com/pytorch/pytorch/issues/158401,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: pt2-dispatcher, oncall: pt2",scan when time allows,2025-07-23T13:36:34Z
issue,97432,suspicious memory leak when increase DataLoader's prefetch_factor and enable pin_memory,https://github.com/pytorch/pytorch/issues/97432,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dataloader, module: memory usage",scan when time allows,2025-07-23T11:19:31Z
issue,90820,Support `divmod` for tensors,https://github.com/pytorch/pytorch/issues/90820,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: numpy,scan when time allows,2025-07-23T09:52:08Z
issue,158832,Lazy Layer Norm,https://github.com/pytorch/pytorch/issues/158832,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2025-07-22T22:36:20Z
issue,158566,torch.compile fails with UncapturedHigherOrderOpError when using torch.ops.higher_order.scan,https://github.com/pytorch/pytorch/issues/158566,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: higher order operators, oncall: pt2",scan when time allows,2025-07-22T18:40:30Z
issue,153234,`torch.quantile` perform differently on cpu and cuda,https://github.com/pytorch/pytorch/issues/153234,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: edge cases, module: error checking, module: reductions",scan when time allows,2025-07-22T16:20:19Z
issue,158643,torch.compile() produces inconsistent outputs with models containing dropout and in-place ReLU operations,https://github.com/pytorch/pytorch/issues/158643,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: pt2,scan when time allows,2025-07-22T15:43:31Z
issue,149393,Torch RPC examples from docs say usage is deprecated.,https://github.com/pytorch/pytorch/issues/149393,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: rpc, oncall: distributed",scan when time allows,2025-07-22T15:14:07Z
issue,158785,[lint] Set_linter suggests wrong changes,https://github.com/pytorch/pytorch/issues/158785,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: lint,scan when time allows,2025-07-22T14:46:07Z
issue,158240,[Windows][inductor] Windows compilers building was slower 4x than on Linux.,https://github.com/pytorch/pytorch/issues/158240,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, module: windows, oncall: pt2",scan when time allows,2025-07-22T14:34:18Z
issue,158664,DISABLED test_sum_all_cpu_float64 (__main__.TestReductionsCPU),https://github.com/pytorch/pytorch/issues/158664,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: oncall: cpu inductor, oncall: pt2",scan when time allows,2025-07-22T14:29:57Z
issue,72821,"`{Batch,Instance}Norm{1,2,3}d` works when `num_features != C`!",https://github.com/pytorch/pytorch/issues/72821,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2025-07-22T09:05:00Z
issue,158326,Support for etcd v3 api as rendezvous backend,https://github.com/pytorch/pytorch/issues/158326,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: elastic, oncall: distributed",scan when time allows,2025-07-21T20:44:41Z
issue,146330,DeepSeek: MLA attention,https://github.com/pytorch/pytorch/issues/146330,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sdpa,scan when time allows,2025-07-20T20:22:49Z
issue,158641,support checkPoolLiveAllocations for cudaMallocAsync,https://github.com/pytorch/pytorch/issues/158641,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2025-07-20T15:37:53Z
issue,92927,"Inplace fused (leaky)relu+(leaky)dropout for memory savings (I think, can be made fully allocation-less if never fully allocating random mask in FlashAttention style and recover the mask from the output)",https://github.com/pytorch/pytorch/issues/92927,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: aotdispatch, module: pt2-dispatcher, oncall: pt2",scan when time allows,2025-07-19T20:51:32Z
issue,61819,"[discussion] Support other index dtypes for indexing, take, put, scatter, scatter_reduce and other indexing functions in addition to int64: int8, uint8, int16, int32, uint16 etc (without casting copies/reallocations)",https://github.com/pytorch/pytorch/issues/61819,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: advanced indexing, module: scatter & gather ops, module: type promotion",scan when time allows,2025-07-19T15:58:51Z
issue,157976,support DTensor's local_map in compile,https://github.com/pytorch/pytorch/issues/157976,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dtensor, module: dynamo, oncall: pt2",scan when time allows,2025-07-19T00:33:37Z
issue,158610,vmap will always disable gradient calculation,https://github.com/pytorch/pytorch/issues/158610,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: functorch, module: vmap",scan when time allows,2025-07-18T16:15:36Z
issue,103329,"RuntimeError: torch.vmap a function that includes in-place arithmetic operations on a zero-initialized tensor, an error ""vmap: inplace arithmetic(self, *extra_args) is not possible"" is raised.",https://github.com/pytorch/pytorch/issues/103329,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: functorch,scan when time allows,2025-07-18T04:12:49Z
issue,158364,[feature request] [memory_viz] Flexible time/range memory event filtering controls to skip loading all memory events from pickle in the browser's memory,https://github.com/pytorch/pytorch/issues/158364,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: profiler,scan when time allows,2025-07-17T21:14:42Z
issue,138452,Cholesky factorization of sparse tensors,https://github.com/pytorch/pytorch/issues/138452,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: linear algebra, module: sparse",scan when time allows,2025-07-17T19:57:33Z
issue,152220,[C10D] Allow NCCL single P2P ops to use parent/collective communicator,https://github.com/pytorch/pytorch/issues/152220,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nccl, oncall: distributed",scan when time allows,2025-07-17T19:34:50Z
issue,152756,Cuda-12.9 removed libnvToolsExt.so.* and is now purely header nvtx3,https://github.com/pytorch/pytorch/issues/152756,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2025-07-17T18:23:32Z
issue,31252,[feature request] Better handling for CUDA Out of Memory,https://github.com/pytorch/pytorch/issues/31252,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2025-07-17T16:57:47Z
issue,157465,Nightly NCCL builds are missing optional features from NCCL,https://github.com/pytorch/pytorch/issues/157465,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: nccl",scan when time allows,2025-07-17T16:54:18Z
issue,93716,End-to-End AMP training with GradScaler,https://github.com/pytorch/pytorch/issues/93716,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: compile-time, oncall: pt2",scan when time allows,2025-07-17T14:20:20Z
issue,158412,Inconsistent torch.absolute results on complex128 between CPU and CUDA,https://github.com/pytorch/pytorch/issues/158412,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: edge cases, module: nans and infs",scan when time allows,2025-07-17T12:39:18Z
issue,152032,RFC: The State of Custom CUDA extensions in PyTorch,https://github.com/pytorch/pytorch/issues/152032,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp-extensions, module: cuda",scan when time allows,2025-07-17T05:40:10Z
issue,158419,torch.copysign for float16 produces inconsistent results between CPU and CUDA,https://github.com/pytorch/pytorch/issues/158419,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: edge cases,scan when time allows,2025-07-17T04:28:27Z
issue,106664,[docs] URL and link format proposal to make function page URLs more concise,https://github.com/pytorch/pytorch/issues/106664,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2025-07-16T20:28:54Z
issue,113232,[Torch Inductor] Torch Inductor Better Support for GNN workload and Inductor Sparse Compiler,https://github.com/pytorch/pytorch/issues/113232,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamic shapes, module: inductor, oncall: pt2",scan when time allows,2025-07-16T20:17:03Z
issue,153463,Feedback about Class Tensor,https://github.com/pytorch/pytorch/issues/153463,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: docs",scan when time allows,2025-07-16T17:04:44Z
issue,102517,[feature request] PyTorch support for sub-interpreters with PEP 684 accepted and release in Python 3.12,https://github.com/pytorch/pytorch/issues/102517,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python frontend,scan when time allows,2025-07-16T16:07:38Z
issue,124714,`torch.distributed` hangs when using `torch.distributed.barrier` before any other communication primitives.,https://github.com/pytorch/pytorch/issues/124714,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2025-07-16T14:37:38Z
issue,37162,torch.distributions bug in RelaxedOneHotCategorical.log_prob,https://github.com/pytorch/pytorch/issues/37162,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distributions,scan when time allows,2025-07-16T09:26:36Z
issue,157624,file_name is not correctly read in here,https://github.com/pytorch/pytorch/issues/157624,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: serialization, module: windows",scan when time allows,2025-07-16T09:08:47Z
issue,148807,Native Infinite Sampler for Datasets,https://github.com/pytorch/pytorch/issues/148807,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: data, module: dataloader",scan when time allows,2025-07-16T00:28:09Z
issue,158169,Invalid `torch.linalg.lstsq` check during vmapped implementation,https://github.com/pytorch/pytorch/issues/158169,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: functorch, module: linear algebra, module: vmap",scan when time allows,2025-07-15T09:39:49Z
issue,158244,[QST] [DTensor] MoE + EP support,https://github.com/pytorch/pytorch/issues/158244,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2025-07-15T02:26:58Z
issue,158125,index.Tensor doesn't properly account for boolean masks acting on multiple dimensions.,https://github.com/pytorch/pytorch/issues/158125,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python frontend,scan when time allows,2025-07-14T23:29:09Z
issue,111519,"[pt2+profiler] attach aot_id to CompiledFunction, _compiled_fn number without aot_autograd",https://github.com/pytorch/pytorch/issues/111519,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: pt2,scan when time allows,2025-07-14T22:42:46Z
issue,112883,Torchbench inference failures,https://github.com/pytorch/pytorch/issues/112883,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: pt2,scan when time allows,2025-07-14T22:42:13Z
issue,155616,Additional Documentation About Size Checking with Custom Operators,https://github.com/pytorch/pytorch/issues/155616,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: custom-operators, module: pt2-dispatcher, oncall: pt2",scan when time allows,2025-07-14T21:05:59Z
issue,158080,loaded storage won't be cached during deserialization if location is meta,https://github.com/pytorch/pytorch/issues/158080,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: performance, module: serialization",scan when time allows,2025-07-14T14:47:22Z
issue,158172,`torch.fmin` has inconsistent overflow behavior on CPU and GPU,https://github.com/pytorch/pytorch/issues/158172,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: edge cases",scan when time allows,2025-07-14T02:55:01Z
issue,108565,`bytes(...)` support of torch tensor does not match numpy + it would be nice to support tensor.tobytes() as alias,https://github.com/pytorch/pytorch/issues/108565,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: numpy,scan when time allows,2025-07-13T22:14:48Z
issue,1410,BatchNorm should use Bessel's correction consistently,https://github.com/pytorch/pytorch/issues/1410,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: norms and normalization",scan when time allows,2025-07-13T21:31:56Z
issue,118715,Add capturable Adagrad implementation,https://github.com/pytorch/pytorch/issues/118715,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda graphs, module: optimizer, oncall: pt2",scan when time allows,2025-07-12T20:26:32Z
issue,157391,[FSDP2] document the contract for modifying DTensor model.parameters(),https://github.com/pytorch/pytorch/issues/157391,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2025-07-12T00:52:24Z
issue,157764,Drop SSE4 support in oneDNN,https://github.com/pytorch/pytorch/issues/157764,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bc-breaking, module: intel, module: mkldnn",scan when time allows,2025-07-12T00:16:11Z
issue,157495,[DTensor] Improve `tensor_metadata` and `redistribute_cost` coverage for op strategies.,https://github.com/pytorch/pytorch/issues/157495,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dtensor, oncall: distributed",scan when time allows,2025-07-11T22:31:37Z
issue,103990,[Inductor] Freezing Add support for Caching Parameter Conversions,https://github.com/pytorch/pytorch/issues/103990,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",scan when time allows,2025-07-11T18:17:07Z
issue,120667,Triton kernel unexpectedly gets 1.35x slower by more specializaiton,https://github.com/pytorch/pytorch/issues/120667,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: performance, oncall: pt2",scan when time allows,2025-07-11T18:10:35Z
issue,121371,Match HuggingFace T5 SDPA pattern in Inductor,https://github.com/pytorch/pytorch/issues/121371,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, module: performance, oncall: pt2",scan when time allows,2025-07-11T18:04:49Z
issue,121162,opcheck should support TorchBind custom classes,https://github.com/pytorch/pytorch/issues/121162,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: tests, module: torchbind, oncall: pt2",scan when time allows,2025-07-11T18:04:26Z
issue,120116,Use Incremental Fake Tensor Updater more uniformly across torch.compile compilation,https://github.com/pytorch/pytorch/issues/120116,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",scan when time allows,2025-07-11T18:03:39Z
issue,120079,Use sys.settrace or torch function mode to compute how much of a model was not covered by Dynamo,https://github.com/pytorch/pytorch/issues/120079,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: logging, oncall: pt2",scan when time allows,2025-07-11T18:03:17Z
issue,120057,Torch compile fx graph is not removing constant propagation,https://github.com/pytorch/pytorch/issues/120057,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",scan when time allows,2025-07-11T18:02:32Z
issue,118774,Tune whether to use mm or bmm for matmul in inductor max-autotune,https://github.com/pytorch/pytorch/issues/118774,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cublas, module: cuda, module: inductor, oncall: pt2",scan when time allows,2025-07-11T17:59:43Z
issue,128012,compile PixArt-sigma error,https://github.com/pytorch/pytorch/issues/128012,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",scan when time allows,2025-07-11T17:57:42Z
issue,41213,libTorch cpp docs missing for Tensor::item(),https://github.com/pytorch/pytorch/issues/41213,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: docs",scan when time allows,2025-07-11T17:55:35Z
issue,124866,support side effects in HOPs?,https://github.com/pytorch/pytorch/issues/124866,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: compiled autograd, module: higher order operators, module: pt2-dispatcher, oncall: pt2",scan when time allows,2025-07-11T16:45:50Z
issue,151303,[Feature Request] Experimental support to Moore Threads GPU MUSA,https://github.com/pytorch/pytorch/issues/151303,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: backend, module: privateuse1",scan when time allows,2025-07-11T16:01:47Z
issue,29734,Tensor.nbytes() returns itemsize * numel for sparse tensors,https://github.com/pytorch/pytorch/issues/29734,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: sparse",scan when time allows,2025-07-11T12:49:38Z
issue,121631,[inductor] Incorrect handle of `autocast` results in type mismatch,https://github.com/pytorch/pytorch/issues/121631,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",scan when time allows,2025-07-11T00:58:00Z
issue,103173,[RFC] Emit better Telemetry in PyTorch,https://github.com/pytorch/pytorch/issues/103173,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: logging, oncall: pt2",scan when time allows,2025-07-10T23:58:53Z
issue,126644,[torch.compile]: Enhanced Error Reporting and Performance Canary Mode,https://github.com/pytorch/pytorch/issues/126644,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: pt2,scan when time allows,2025-07-10T23:53:05Z
issue,118175,Different behaviors in `torch.nn.functional.hinge_embedding_loss` between eagermode and torch.compile,https://github.com/pytorch/pytorch/issues/118175,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",scan when time allows,2025-07-10T23:30:06Z
issue,125672,[inductor][cpu]mobilenet_v2_quantized_qat float32 single thread static/dynamic shape CPP/default wrapper performance regression in 2024-04-28 nightly release,https://github.com/pytorch/pytorch/issues/125672,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: cpu inductor, oncall: pt2",scan when time allows,2025-07-10T22:48:24Z
issue,123548,An error occurs when ‘max_split_size_mb ’and ‘expandable_segments ’ are enabled at the same time.,https://github.com/pytorch/pytorch/issues/123548,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: cudacachingallocator, module: docs, module: memory usage",scan when time allows,2025-07-10T20:28:25Z
issue,158041,pow_test fails on Aarch64,https://github.com/pytorch/pytorch/issues/158041,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: arm, module: correctness (silent), module: macos, module: nans and infs",scan when time allows,2025-07-10T17:57:59Z
issue,157334,Inefficient 2D convolution compared to JAX,https://github.com/pytorch/pytorch/issues/157334,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: nn, module: performance",scan when time allows,2025-07-10T16:31:33Z
issue,125217,MaxPool2D memory leakage on device MPS,https://github.com/pytorch/pytorch/issues/125217,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: memory usage, module: mps",scan when time allows,2025-07-10T16:28:15Z
issue,133250,[torch.compile] tighten FX graph restrictions post-functionalization,https://github.com/pytorch/pytorch/issues/133250,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, module: pt2-dispatcher, oncall: pt2",scan when time allows,2025-07-10T13:20:16Z
issue,55261,Batched multi_dot / chain_matmul + let it accept a tensor instead of tuple,https://github.com/pytorch/pytorch/issues/55261,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: linear algebra,scan when time allows,2025-07-10T07:52:02Z
issue,148289,Triton Kernel Rejects NamedTupleVariable Arguments,https://github.com/pytorch/pytorch/issues/148289,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamo, module: fx, module: user triton, oncall: pt2",scan when time allows,2025-07-10T00:17:56Z
issue,157502,Failure with cub::TransformInputIterator in 12.9 periodic CI test,https://github.com/pytorch/pytorch/issues/157502,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2025-07-09T22:02:30Z
issue,156258,[ONNX] Create a tutorial for exporting hf transformers model,https://github.com/pytorch/pytorch/issues/156258,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: onnx, oncall: pt2",scan when time allows,2025-07-09T18:05:12Z
issue,153472,Multihead Attention does not work with jagged tensors due to __torch_function__,https://github.com/pytorch/pytorch/issues/153472,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nestedtensor,scan when time allows,2025-07-09T17:31:22Z
issue,150328,☂️ Update submodule dependencies to supported version of Cmake,https://github.com/pytorch/pytorch/issues/150328,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2025-07-09T17:03:25Z
issue,150118,PyTorch 2.6 License Issues,https://github.com/pytorch/pytorch/issues/150118,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: third_party, oncall: releng",scan when time allows,2025-07-09T16:53:30Z
issue,150399,torch wheels are unusable if CUDA RPMs are installed on the system (was Import error in nvidia/cuda:12.6.3-cudnn-devel-rockylinux9),https://github.com/pytorch/pytorch/issues/150399,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: cuda, module: third_party",scan when time allows,2025-07-09T16:51:23Z
issue,147011,Deprecation of NVTX 2 (`nvToolsExt`): Recommended to move to NVTX 3,https://github.com/pytorch/pytorch/issues/147011,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, oncall: profiler",scan when time allows,2025-07-09T16:43:04Z
issue,125198,[CUDA][Compex] `test_reference_numerics_large_jiterator_unary_cuda_complex64` broken after updating to `numpy >= 1.25.0`,https://github.com/pytorch/pytorch/issues/125198,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: complex, module: cuda, module: jiterator, module: numpy",scan when time allows,2025-07-09T15:43:49Z
issue,157918,[Precompile] Umbrella task,https://github.com/pytorch/pytorch/issues/157918,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: pt2,scan when time allows,2025-07-09T14:34:30Z
issue,155378,RuntimeError: Could not find libnvrtc.so. Please make sure CUDA is installed.,https://github.com/pytorch/pytorch/issues/155378,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp-extensions,scan when time allows,2025-07-09T12:26:47Z
issue,107112,from_blob python api,https://github.com/pytorch/pytorch/issues/107112,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python frontend,scan when time allows,2025-07-09T12:19:15Z
issue,42654,RuntimeError: each element in list of batch should be of equal size,https://github.com/pytorch/pytorch/issues/42654,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2025-07-09T10:10:35Z
issue,156956,Allow clamp to work on absolute values and preserve sign,https://github.com/pytorch/pytorch/issues/156956,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python frontend,scan when time allows,2025-07-09T09:35:11Z
issue,150734,Training/Fine-tuning fails with PyTorch 2.8 + 4x 5090 GPUs using DDP/FSDP/DeepSpeed,https://github.com/pytorch/pytorch/issues/150734,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ddp, module: fsdp, oncall: distributed",scan when time allows,2025-07-09T06:29:05Z
issue,157591,PT2E Quantization Migration Tracker,https://github.com/pytorch/pytorch/issues/157591,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2025-07-08T22:31:37Z
issue,152212,[Release improvements] Have cherry-pick bot always add the current release to the PR,https://github.com/pytorch/pytorch/issues/152212,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ci, oncall: releng",scan when time allows,2025-07-08T19:41:48Z
issue,148762,magma builds should be part of the docker image builds,https://github.com/pytorch/pytorch/issues/148762,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ci, oncall: releng",scan when time allows,2025-07-08T19:28:56Z
issue,157808,sparse_csr bfloat16 matrix multiplication backward is 10x slower than float16,https://github.com/pytorch/pytorch/issues/157808,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bfloat16, module: cuda, module: performance, module: sparse",scan when time allows,2025-07-08T16:05:56Z
issue,156607,[OSS tooling] pytorchbot fail to revert a PR,https://github.com/pytorch/pytorch/issues/156607,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ci, oncall: releng",scan when time allows,2025-07-08T15:31:52Z
issue,68905,Shared `~/.cache/torch_extensions` needs to be pytorch version aware.,https://github.com/pytorch/pytorch/issues/68905,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp-extensions,scan when time allows,2025-07-08T09:42:29Z
issue,155825,Running dispatch modes on compile-disabled regions of a compiled model,https://github.com/pytorch/pytorch/issues/155825,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamo, module: pt2-dispatcher, oncall: pt2",scan when time allows,2025-07-08T02:19:38Z
issue,157545,test_dtensor.py::test_dtensor_save_load_import conflicts with autoloader importing torch._dynamo,https://github.com/pytorch/pytorch/issues/157545,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dtensor, oncall: distributed",scan when time allows,2025-07-07T22:27:14Z
issue,156810,Add the XPU item to pytorch.org/get-started,https://github.com/pytorch/pytorch/issues/156810,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: docs, oncall: releng",scan when time allows,2025-07-07T22:24:09Z
issue,153337,`torch.combinations` exhibits excessive memory usage and hangs for moderate `n` and `r` due to `n^r`,https://github.com/pytorch/pytorch/issues/153337,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: edge cases,scan when time allows,2025-07-07T20:22:32Z
issue,157662,OffsetBasedRNGTracker's run_state_sync causes deadlock due to inconsistent broadcast order across ranks,https://github.com/pytorch/pytorch/issues/157662,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: random, oncall: distributed",scan when time allows,2025-07-07T19:56:43Z
issue,157721,[Precompile] Caching precompile and presence of packages in general should be safe re: guard serialization,https://github.com/pytorch/pytorch/issues/157721,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2025-07-07T18:34:41Z
issue,1529,[feature request] Caching allocator diagnostics and memory allocation tracing/visualization,https://github.com/pytorch/pytorch/issues/1529,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: memory usage,scan when time allows,2025-07-07T14:47:21Z
issue,156616,Add `is_outputs_batched` param to `autograd.grad`,https://github.com/pytorch/pytorch/issues/156616,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: functorch",scan when time allows,2025-07-06T12:05:05Z
issue,88802,ImportError: libcupti.so.11.2: cannot open shared object file: No such file or directory,https://github.com/pytorch/pytorch/issues/88802,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cuda",scan when time allows,2025-07-06T09:12:41Z
issue,145608,ROCm+gcc 15 asserts,https://github.com/pytorch/pytorch/issues/145608,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rocm,scan when time allows,2025-07-05T21:53:07Z
issue,107041,"I want to calculate the matrix multiplication of two Boolean matrices, but torch.mm will report an error. Is there any more efficient alternative?",https://github.com/pytorch/pytorch/issues/107041,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: boolean tensor,scan when time allows,2025-07-05T17:27:43Z
issue,69197,RendezvousConnectionError when use C10d on multi nodes,https://github.com/pytorch/pytorch/issues/69197,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: r2p,scan when time allows,2025-07-05T17:01:16Z
issue,151591,Device Error on vmap,https://github.com/pytorch/pytorch/issues/151591,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: functorch, module: vmap",scan when time allows,2025-07-04T04:41:49Z
issue,153531,Documentation: explaining the STFT formula,https://github.com/pytorch/pytorch/issues/153531,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: fft",scan when time allows,2025-07-03T22:43:29Z
issue,157548,Nightly cu128 aarch64 wheels haven't been built for weeks,https://github.com/pytorch/pytorch/issues/157548,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: cuda, module: docs",scan when time allows,2025-07-03T19:59:33Z
issue,40375,Add batched torch.combinations,https://github.com/pytorch/pytorch/issues/40375,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: batching,scan when time allows,2025-07-03T14:25:46Z
issue,126753,[cudagraph] simplify usage of how cudagraph dumps debug file,https://github.com/pytorch/pytorch/issues/126753,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda graphs,scan when time allows,2025-07-03T02:18:47Z
issue,110858,Broadcasting matmul is much slower than corresponding einsum,https://github.com/pytorch/pytorch/issues/110858,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: performance",scan when time allows,2025-07-02T23:47:37Z
issue,124487,PyTorch Memory Management in GPU-to-CPU Transfers issue,https://github.com/pytorch/pytorch/issues/124487,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: memory usage,scan when time allows,2025-07-02T18:15:29Z
issue,157431,``torch.quantile`` edge case,https://github.com/pytorch/pytorch/issues/157431,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: edge cases,scan when time allows,2025-07-02T16:21:07Z
issue,156085,Idea: Add SBOM Generation (and optional vuln scan) for better supply chain insight,https://github.com/pytorch/pytorch/issues/156085,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: releng,scan when time allows,2025-07-02T15:03:03Z
issue,156680,torch compile does not support SyncBatchNorm with fullgraph=True,https://github.com/pytorch/pytorch/issues/156680,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ddp, module: dynamo, oncall: distributed, oncall: pt2",scan when time allows,2025-07-02T05:11:48Z
issue,149660,cd: Migrate binary builds off of Jinja,https://github.com/pytorch/pytorch/issues/149660,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: releng,scan when time allows,2025-07-01T20:27:08Z
issue,157075,Improve debug message for metadata guard failure,https://github.com/pytorch/pytorch/issues/157075,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",scan when time allows,2025-07-01T19:18:17Z
issue,116414,Convolution NN for complex numbers and more special functions,https://github.com/pytorch/pytorch/issues/116414,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: complex,scan when time allows,2025-07-01T17:58:35Z
issue,101385,torch.Tensor.is_sparse returns false for non-COO sparse tensors,https://github.com/pytorch/pytorch/issues/101385,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2025-07-01T16:57:40Z
issue,157220,[ONNX] torch.nn.functional.interpolate \w antialias=True isn't op.Resize compatible,https://github.com/pytorch/pytorch/issues/157220,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2025-06-30T22:16:24Z
issue,157321,Symmetric memory test failed with TORCH_SYMMMEM=NVSHMEM,https://github.com/pytorch/pytorch/issues/157321,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2025-06-30T22:14:46Z
issue,157272,`torch.reciprocal` and `torch.divide` for Complex `inf` Incorrectly Returns `NaN` Only for Tensors with >= 4 Elements on CPU,https://github.com/pytorch/pytorch/issues/157272,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: complex, module: cpu, module: edge cases, module: nans and infs",scan when time allows,2025-06-30T21:13:55Z
issue,157266,"Better typechecking of `int` only-operators `|`, `^`, `&`, `<<`, `>>`, `~` and `@`",https://github.com/pytorch/pytorch/issues/157266,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: typing,scan when time allows,2025-06-30T20:04:48Z
issue,156725,CMake improperly configures pybind11. 3 different versions of pybind11 in use at the sametime.,https://github.com/pytorch/pytorch/issues/156725,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: pybind",scan when time allows,2025-06-30T17:07:13Z
issue,33890,Libtorch segfault when used with libqpOASES,https://github.com/pytorch/pytorch/issues/33890,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2025-06-30T08:16:37Z
issue,95882,Proposal: accept ellipsis as index in `torch.index_put()` to enable broadcasting,https://github.com/pytorch/pytorch/issues/95882,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python frontend,scan when time allows,2025-06-30T05:31:17Z
issue,157101,Documentation update: Contribution.md,https://github.com/pytorch/pytorch/issues/157101,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2025-06-30T01:57:34Z
issue,26334,"Supporting ""cdf"" for Student-T distribution",https://github.com/pytorch/pytorch/issues/26334,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distributions,scan when time allows,2025-06-28T23:03:48Z
issue,101107,Make compiled models serializable,https://github.com/pytorch/pytorch/issues/101107,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",scan when time allows,2025-06-28T04:13:42Z
issue,133221,Tensor parallel for convolutions and groupnorm,https://github.com/pytorch/pytorch/issues/133221,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dtensor,scan when time allows,2025-06-27T10:12:41Z
issue,140015,Find ninja from local environment in `torch.utils.cpp_extension.load`,https://github.com/pytorch/pytorch/issues/140015,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp-extensions,scan when time allows,2025-06-27T09:07:58Z
issue,90775,"Extend ""torch.utils.cpp_extension.load"" for both lib64 and **lib**",https://github.com/pytorch/pytorch/issues/90775,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp-extensions,scan when time allows,2025-06-27T04:34:02Z
issue,152964,[FSDP2] need dummy forward/backward to stay SPMD,https://github.com/pytorch/pytorch/issues/152964,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2025-06-26T23:27:27Z
issue,156811,Allow to set a seed for random tensors in torch.onnx.export,https://github.com/pytorch/pytorch/issues/156811,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2025-06-26T18:14:47Z
issue,94718,Option to bypass NOLA check in torch.istft,https://github.com/pytorch/pytorch/issues/94718,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fft,scan when time allows,2025-06-26T16:58:19Z
issue,156733,FX graph printing do newlines for parameters by default,https://github.com/pytorch/pytorch/issues/156733,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2025-06-26T15:45:45Z
issue,156820,put compile-based cudagraph re-recordings in tlparse,https://github.com/pytorch/pytorch/issues/156820,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda graphs, oncall: pt2",scan when time allows,2025-06-26T00:32:43Z
issue,56697,Avoid no-op suggest_memory_format call in SparseCsrTensorImpl::resize_as_sparse_csr_tensor_,https://github.com/pytorch/pytorch/issues/56697,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2025-06-25T20:51:20Z
issue,152261,Context Parallel -- unsharded output doesn't match output without CP.,https://github.com/pytorch/pytorch/issues/152261,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2025-06-25T19:32:01Z
issue,156072,[typing][docs] `torch.amin` and `torch.amax` do not document `dim=None`,https://github.com/pytorch/pytorch/issues/156072,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2025-06-25T09:31:57Z
issue,156238,Upgrade torch._scaled_grouped_mm to SM100+,https://github.com/pytorch/pytorch/issues/156238,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2025-06-25T09:31:21Z
issue,156411,[compile] torch._dynamo.exc.TorchRuntimeError: Failed running call_function aten.lift_fresh_copy.default,https://github.com/pytorch/pytorch/issues/156411,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: oncall: export, oncall: pt2",scan when time allows,2025-06-24T18:17:21Z
issue,149315,How to Retain Computational Graph in torch.func.jvp() for Parameter Gradients?,https://github.com/pytorch/pytorch/issues/149315,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: functorch",scan when time allows,2025-06-24T14:30:39Z
issue,149155,torch.multiprocessing.Queue Zeroes Out Tensors on Retrieval,https://github.com/pytorch/pytorch/issues/149155,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: multiprocessing, module: windows",scan when time allows,2025-06-24T09:51:21Z
issue,121968,[RFC] Use CUDA graphs by default on torch.compile,https://github.com/pytorch/pytorch/issues/121968,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",scan when time allows,2025-06-23T21:59:38Z
issue,156520,[user empathy] compile for `transformers` model,https://github.com/pytorch/pytorch/issues/156520,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: compile-time, module: cuda graphs, module: dynamo, oncall: pt2",scan when time allows,2025-06-23T18:52:15Z
issue,156622,Add a test to torch.library.opcheck for if a C++ meta kernel specializes on a symint,https://github.com/pytorch/pytorch/issues/156622,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: custom-operators, module: dynamic shapes, module: pt2-dispatcher, oncall: pt2",scan when time allows,2025-06-23T18:51:52Z
issue,89255,torch.lobpcg should support black-box linear operators like SciPy,https://github.com/pytorch/pytorch/issues/89255,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: linear algebra,scan when time allows,2025-06-23T17:23:37Z
issue,153241,Online softmax is disabled on the fly,https://github.com/pytorch/pytorch/issues/153241,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",scan when time allows,2025-06-22T10:22:26Z
issue,111374,Tensor.lerp inconsistent when using -Infinity between MPS and CPU,https://github.com/pytorch/pytorch/issues/111374,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: mps,scan when time allows,2025-06-21T13:05:36Z
issue,88327,MSE documentation is weak,https://github.com/pytorch/pytorch/issues/88327,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: nn",scan when time allows,2025-06-20T23:41:11Z
issue,155171,[AC] torch.utils.checkpoint.CheckpointError from HF qwen2,https://github.com/pytorch/pytorch/issues/155171,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: activation checkpointing,scan when time allows,2025-06-20T21:26:42Z
issue,156444,Tensor.is_pinned() raises error after renaming privateuseone backend.,https://github.com/pytorch/pytorch/issues/156444,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: privateuse1,scan when time allows,2025-06-20T21:23:14Z
issue,154810,Multi-dimensional tensors in datasets might get incorrectly flattened when fetching data from dataloader which is specified 'batch_sampler' when created,https://github.com/pytorch/pytorch/issues/154810,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2025-06-20T19:38:55Z
issue,148335,[CI] [anaconda] Docker files have conda environment installed,https://github.com/pytorch/pytorch/issues/148335,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2025-06-20T17:21:04Z
issue,148339,[Docs] [anaconda] Review and update,https://github.com/pytorch/pytorch/issues/148339,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2025-06-20T17:20:02Z
issue,148338,[CI] [anaconda] CI Build and Test scripts Windows,https://github.com/pytorch/pytorch/issues/148338,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2025-06-20T17:19:40Z
issue,148336,[CI] [anaconda] CI Build and Test scripts Linux,https://github.com/pytorch/pytorch/issues/148336,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2025-06-20T17:19:22Z
issue,147720,[ONNX Convert] Error when input to nn.AdaptiveAvgPool2d size is variable,https://github.com/pytorch/pytorch/issues/147720,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2025-06-20T14:32:01Z
issue,151064,[ONNX] Use dlpack to transfer tensors when onnxruntime implements proper support,https://github.com/pytorch/pytorch/issues/151064,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2025-06-20T14:27:17Z
issue,113817,[ONNX] ONNX export of simple quantized model fails,https://github.com/pytorch/pytorch/issues/113817,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2025-06-20T00:15:19Z
issue,150277,"Sparse tensor indexing not implemented, but partially supported by using index_select",https://github.com/pytorch/pytorch/issues/150277,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2025-06-19T16:38:05Z
issue,152226,NotImplementedError: Could not run 'aten::index.Tensor' with arguments from the 'SparseCUDA' backend.,https://github.com/pytorch/pytorch/issues/152226,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2025-06-19T16:30:16Z
issue,153329,Segmentation fault when converting sparse COO tensor with complex values to dense,https://github.com/pytorch/pytorch/issues/153329,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: edge cases, module: sparse",scan when time allows,2025-06-19T15:58:03Z
issue,150972,Escape hatch: way to dynamically add or remove tags from custom operators,https://github.com/pytorch/pytorch/issues/150972,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: custom-operators, module: pt2-dispatcher, oncall: pt2",scan when time allows,2025-06-18T18:11:05Z
issue,106801,"Timer benchmark stores only one time value, and therefore has broken mean/median/etc metrics",https://github.com/pytorch/pytorch/issues/106801,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: benchmark,scan when time allows,2025-06-18T17:07:06Z
issue,154602,Cudnn attention is very slow when sequence length changed in every step,https://github.com/pytorch/pytorch/issues/154602,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: cudnn, module: sdpa",scan when time allows,2025-06-18T14:54:59Z
issue,138329,DTensor RNG state for non CUDA backends,https://github.com/pytorch/pytorch/issues/138329,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2025-06-18T03:46:30Z
issue,155552,Reproducibility of results without AVX512 by setting ATEN_CPU_CAPABILITY=avx2,https://github.com/pytorch/pytorch/issues/155552,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: determinism, module: intel, module: mkl, module: mkldnn",scan when time allows,2025-06-18T02:03:55Z
issue,123834,"RuntimeError: NVML_SUCCESS == r INTERNAL ASSERT FAILED at ""/opt/pytorch/pytorch/c10/cuda/CUDACachingAllocator.cpp"":830, please report a bug to PyTorch.",https://github.com/pytorch/pytorch/issues/123834,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: cudacachingallocator",scan when time allows,2025-06-17T22:07:43Z
issue,156177,nn.Module._load_from_state_dict is always called with strict=True,https://github.com/pytorch/pytorch/issues/156177,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2025-06-17T21:46:30Z
issue,155203,support for cuDNN 9.8+,https://github.com/pytorch/pytorch/issues/155203,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: cudnn",scan when time allows,2025-06-17T20:47:16Z
issue,107214,The difference between input grad computed by channels last backward and the input grad computed by channels first backward of Hardswish on MPS is too large,https://github.com/pytorch/pytorch/issues/107214,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: memory format, module: mps, module: nn",scan when time allows,2025-06-17T15:59:42Z
issue,156024,Ошибка установки torch для CUDA 12.1 на GTX 1660 Ti,https://github.com/pytorch/pytorch/issues/156024,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: windows",scan when time allows,2025-06-17T15:30:03Z
issue,156152,Inconsistent `torch.rsqrt` results on complex128 between CPU and CUDA,https://github.com/pytorch/pytorch/issues/156152,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: complex, module: edge cases, module: nans and infs",scan when time allows,2025-06-17T15:28:30Z
issue,156073,[codespell] fix typos in the codebase,https://github.com/pytorch/pytorch/issues/156073,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: lint,scan when time allows,2025-06-17T14:58:13Z
issue,150544,[ONNX] dynamic_axes does not rename dynamic dimension in torch.onnx.export,https://github.com/pytorch/pytorch/issues/150544,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2025-06-17T14:31:33Z
issue,10454,[feature request] Rank-Revealing QR - Adding dgeqp3 support to torch.qr,https://github.com/pytorch/pytorch/issues/10454,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: linear algebra,scan when time allows,2025-06-17T08:46:14Z
issue,55289,Feature Request: Add a rounding mode to round,https://github.com/pytorch/pytorch/issues/55289,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: numpy,scan when time allows,2025-06-17T07:20:40Z
issue,61509,[feature request] Exact euclidean distance transform,https://github.com/pytorch/pytorch/issues/61509,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distance functions,scan when time allows,2025-06-17T02:00:48Z
issue,146566,[DCP] Allow for rank-specific tensors with duplicate keys,https://github.com/pytorch/pytorch/issues/146566,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed checkpointing,scan when time allows,2025-06-17T01:33:50Z
issue,64491,TypeError when using torch.cuda.list_gpu_processes() on Windows with the WDDM driver,https://github.com/pytorch/pytorch/issues/64491,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2025-06-16T13:34:45Z
issue,156020,torch.fft.ifft for complex64 produces inconsistent results between CPU and CUDA,https://github.com/pytorch/pytorch/issues/156020,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: edge cases, module: fft, module: nans and infs",scan when time allows,2025-06-16T02:42:23Z
issue,88227,Enable `torch.topk` to support `stable` flag,https://github.com/pytorch/pytorch/issues/88227,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sorting and selection,scan when time allows,2025-06-15T14:16:35Z
issue,146221,"Incompatible Torch and Torchvision while building from source for 2.6.0 and CUDA 12.6, RuntimeError: operator torchvision::nms does not exist",https://github.com/pytorch/pytorch/issues/146221,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: dependency bug",scan when time allows,2025-06-15T13:17:20Z
issue,117287,Error computing the norm of MaskedTensor,https://github.com/pytorch/pytorch/issues/117287,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: masked operators,scan when time allows,2025-06-14T22:11:42Z
issue,143474,[ONNX] Failed to export PyTorch-2-Export-Quantized model to onnx,https://github.com/pytorch/pytorch/issues/143474,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2025-06-13T22:54:27Z
issue,155838,intmm is being compiled inconsistently with errors.,https://github.com/pytorch/pytorch/issues/155838,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2, oncall: quantization",scan when time allows,2025-06-13T20:35:30Z
issue,51284,result_type doesn't take dtypes and doesn't match numpy,https://github.com/pytorch/pytorch/issues/51284,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: numpy,scan when time allows,2025-06-13T00:11:23Z
issue,82091,"RuntimeError: [1] is setting up NCCL communicator and retreiving ncclUniqueId from [0] via c10d key-value store by key '0', but store->get('0') got error: Timeout waiting for key: default_pg/0/0 after 1800000 ms",https://github.com/pytorch/pytorch/issues/82091,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: c10d, module: nccl, oncall: distributed",scan when time allows,2025-06-12T18:41:11Z
issue,155218,3D Conv Slow with Large Tensors (> 2**31) Elements,https://github.com/pytorch/pytorch/issues/155218,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: cuda, module: performance",scan when time allows,2025-06-12T10:21:04Z
issue,155001,The profiler does not seem to be able to record cuda runtime nodes,https://github.com/pytorch/pytorch/issues/155001,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: profiler,scan when time allows,2025-06-12T07:34:57Z
issue,81985,"nn.InstanceNorm and nn.GroupNorm are affected by padding, so they need to masking",https://github.com/pytorch/pytorch/issues/81985,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: masked operators, module: nestedtensor, module: norms and normalization, module: padding",scan when time allows,2025-06-12T04:22:59Z
issue,155680,[FSDP] optimizer_state_dict fails if FSDP model is not on global rank 0,https://github.com/pytorch/pytorch/issues/155680,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2025-06-12T03:53:41Z
issue,155667,[libtorch] Debug and Release should be shipped together on Windows,https://github.com/pytorch/pytorch/issues/155667,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: windows,scan when time allows,2025-06-11T18:38:23Z
issue,144344,custom_op's backward changes can't invalidate `torch.compile` cache for backward,https://github.com/pytorch/pytorch/issues/144344,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: custom-operators, module: pt2-dispatcher, oncall: pt2",scan when time allows,2025-06-11T17:20:15Z
issue,155386,OpOverloads should have annotations,https://github.com/pytorch/pytorch/issues/155386,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: custom-operators, module: dispatch, module: pt2-dispatcher, oncall: pt2",scan when time allows,2025-06-11T16:32:41Z
issue,131774,Different Cholesky results between Windows & Linux,https://github.com/pytorch/pytorch/issues/131774,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: linear algebra, module: windows",scan when time allows,2025-06-11T16:21:13Z
issue,152720,When scoped_libary is destroyed the fake impls are not cleared,https://github.com/pytorch/pytorch/issues/152720,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: custom-operators, module: pt2-dispatcher, oncall: pt2",scan when time allows,2025-06-11T16:20:40Z
issue,155703,[pt2] [precompile] Support cudagraphs in BundledAOTAutogradCacheArtifact,https://github.com/pytorch/pytorch/issues/155703,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2025-06-11T14:38:11Z
issue,102400,Unexpected Behavior when using torch.isclose(),https://github.com/pytorch/pytorch/issues/102400,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nans and infs, module: testing",scan when time allows,2025-06-10T14:58:36Z
issue,155519,[Tracking] Triton 3.2 deprecation,https://github.com/pytorch/pytorch/issues/155519,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: user triton,scan when time allows,2025-06-10T13:31:36Z
issue,152294,Unexpected overflow behavior when using `torch.addcmul`,https://github.com/pytorch/pytorch/issues/152294,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpu,scan when time allows,2025-06-10T08:40:31Z
issue,155490,PP activation offloading,https://github.com/pytorch/pytorch/issues/155490,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: pipelining, oncall: distributed",scan when time allows,2025-06-10T07:14:49Z
issue,153967,Memory Corruption in `torch.batch_norm_update_stats`,https://github.com/pytorch/pytorch/issues/153967,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2025-06-10T04:00:13Z
issue,155105,"[dynamo && compiled autograd] compiled autograd recompiles when using custom autograd functions and compiling the forward with backend ""eager""",https://github.com/pytorch/pytorch/issues/155105,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: compiled autograd, module: dynamo, module: higher order operators, module: pt2-dispatcher",scan when time allows,2025-06-09T23:49:42Z
issue,155230,[scan] scan is broken in nightly,https://github.com/pytorch/pytorch/issues/155230,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: higher order operators,scan when time allows,2025-06-09T23:39:42Z
issue,155065,Flex Attention and Nested Tensor: very high VRAM usage,https://github.com/pytorch/pytorch/issues/155065,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: pt2,scan when time allows,2025-06-09T20:07:50Z
issue,155193,Should be ReLU6(Module),https://github.com/pytorch/pytorch/issues/155193,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2025-06-09T20:06:54Z
issue,152870,Docs Update `wrap_triton`,https://github.com/pytorch/pytorch/issues/152870,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: higher order operators, module: pt2-dispatcher, module: user triton, oncall: pt2",scan when time allows,2025-06-09T17:43:28Z
issue,155333,[A100][cusparselt] Non-determinstic correctness issue with some algorithms,https://github.com/pytorch/pytorch/issues/155333,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2025-06-09T17:27:44Z
issue,154428,Inconsistent sin(x) output between CPU and CUDA for very large arguments,https://github.com/pytorch/pytorch/issues/154428,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: edge cases, module: python frontend",scan when time allows,2025-06-09T16:58:50Z
issue,120584,`torch.set_default_tensor_type() is deprecated as of PyTorch 2.1` appearing in logs even when not using this function,https://github.com/pytorch/pytorch/issues/120584,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: deprecation, module: docs, module: error checking",scan when time allows,2025-06-09T08:59:50Z
issue,154332,Support jvp for flex attention,https://github.com/pytorch/pytorch/issues/154332,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: flex attention, module: forward ad, module: functorch",scan when time allows,2025-06-08T00:39:49Z
issue,120986,Intel MKL DFTI ERROR,https://github.com/pytorch/pytorch/issues/120986,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: edge cases, module: empty tensor, module: fft, module: intel",scan when time allows,2025-06-08T00:02:46Z
issue,155350,CUDA_HOME doesn't seem to work with setup script,https://github.com/pytorch/pytorch/issues/155350,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cuda",scan when time allows,2025-06-07T14:32:22Z
issue,75173,LowRankMultivariateNormal doesn't work with 0 diagonal,https://github.com/pytorch/pytorch/issues/75173,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distributions,scan when time allows,2025-06-07T12:44:45Z
issue,155157,Multi-worker dataloader for `IterableDataset` holds open process on macOS Python 3.12,https://github.com/pytorch/pytorch/issues/155157,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2025-06-06T22:04:34Z
issue,133676,"Can pytorch add sparse linear solvers like scipy.sparse.linalg.gmres, scipy.sparse.linalg.bicg etc.",https://github.com/pytorch/pytorch/issues/133676,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: linear algebra, module: sparse",scan when time allows,2025-06-06T13:46:37Z
issue,151695,FP8 Support for FlexAttention,https://github.com/pytorch/pytorch/issues/151695,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: flex attention, module: higher order operators, module: pt2-dispatcher, oncall: pt2",scan when time allows,2025-06-06T13:19:39Z
issue,129656,RuntimeError: Shared memory manager connection has timed out,https://github.com/pytorch/pytorch/issues/129656,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: multiprocessing,scan when time allows,2025-06-06T02:45:08Z
issue,152922,Enable 12.8.1,https://github.com/pytorch/pytorch/issues/152922,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2025-06-06T00:00:56Z
issue,155266,torch.compile with InvokeAI: 'UserDefinedObjectVariable' object has no attribute 'proxy' / compute_ancestors KeyError op4,https://github.com/pytorch/pytorch/issues/155266,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamo, module: inductor, oncall: pt2",scan when time allows,2025-06-05T22:14:24Z
issue,135126,"""RuntimeError: CUDA error: operation not supported"" fixed by downgrading toolkit version",https://github.com/pytorch/pytorch/issues/135126,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cuda, oncall: pt2",scan when time allows,2025-06-05T16:13:26Z
issue,144068,redundant recompilation caused by duplicated Sym(),https://github.com/pytorch/pytorch/issues/144068,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamic shapes, module: dynamo, oncall: pt2",scan when time allows,2025-06-05T08:04:30Z
issue,154847,don't require recompiles when switching between torch.Tensor vs AsyncCollectiveTensor graph inputs,https://github.com/pytorch/pytorch/issues/154847,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: c10d, oncall: distributed, oncall: pt2",scan when time allows,2025-06-05T05:37:28Z
issue,141204,Stack trace from pytest is very far away and far too find on some tests,https://github.com/pytorch/pytorch/issues/141204,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ci, module: log classifier, oncall: pt2",scan when time allows,2025-06-05T03:15:51Z
issue,154984,[FSDP2] Slower Convergence with fully_shard() Compared to DDP during Qwen2-VL Fine-Tuning,https://github.com/pytorch/pytorch/issues/154984,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: fsdp, oncall: distributed",scan when time allows,2025-06-05T02:19:58Z
issue,154824,torch inductor cudagraph tree could incorrectly release some input nodes during replay,https://github.com/pytorch/pytorch/issues/154824,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda graphs, oncall: pt2",scan when time allows,2025-06-04T22:47:29Z
issue,154351,bfloat16 Conv2d slower than float16 on 4090,https://github.com/pytorch/pytorch/issues/154351,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bfloat16, module: cuda, module: half, module: performance",scan when time allows,2025-06-04T20:20:38Z
issue,119548,Inconsistent behavior when indexing a Tensor with a list of lists,https://github.com/pytorch/pytorch/issues/119548,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: advanced indexing,scan when time allows,2025-06-04T15:56:12Z
issue,154214,AssertionError: Scalars are not equal! ```test_require_stride_expanded_dynamic_shapes_cuda```,https://github.com/pytorch/pytorch/issues/154214,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: pt2,scan when time allows,2025-06-04T15:25:56Z
issue,122454,Triangular matrix storage + matmul,https://github.com/pytorch/pytorch/issues/122454,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: linear algebra,scan when time allows,2025-06-04T03:10:21Z
issue,147643,Representation string of a meta tensor is not a valid `tensor` call,https://github.com/pytorch/pytorch/issues/147643,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: meta tensors, module: python frontend",scan when time allows,2025-06-04T02:03:42Z
issue,54816,Per channel weight observer for ConvTranspose,https://github.com/pytorch/pytorch/issues/54816,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2025-06-03T22:46:37Z
issue,139108,LibTorch build error on Windows for CUDA version (debug/release),https://github.com/pytorch/pytorch/issues/139108,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cpp, module: cuda, module: windows",scan when time allows,2025-06-03T20:59:29Z
issue,144767,Unable to build with ATEN_THREADING=TBB option,https://github.com/pytorch/pytorch/issues/144767,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: docs, module: tbb",scan when time allows,2025-06-03T20:38:45Z
issue,154560,Enable GB200 for dynamo/torchbench tests,https://github.com/pytorch/pytorch/issues/154560,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ci, module: cuda, oncall: pt2",scan when time allows,2025-06-03T20:24:00Z
issue,153273,Auto format lint not making suggestions,https://github.com/pytorch/pytorch/issues/153273,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: devx,scan when time allows,2025-06-03T20:12:53Z
issue,152608,[triton pin update] Run Inductor CI on pin updates for Triton and the PyTorch nightly branch,https://github.com/pytorch/pytorch/issues/152608,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: user triton, oncall: releng",scan when time allows,2025-06-03T20:08:38Z
issue,154128,matmul uses excessive memory in batch cases with more than 3 dimensions,https://github.com/pytorch/pytorch/issues/154128,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: memory usage,scan when time allows,2025-06-03T16:55:05Z
issue,154960,[FSDP2] all_gather_copy_in for cpu offload,https://github.com/pytorch/pytorch/issues/154960,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fsdp,scan when time allows,2025-06-03T15:59:54Z
issue,154855,"test_tensorboard.py::TestTensorBoardSummary::test_histogram_auto Fails with s390x-periodic / linux-manylinux-2_28-py3-cpu-s390x / test (default, 7, 10, linux.s390x)",https://github.com/pytorch/pytorch/issues/154855,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: tensorboard, oncall: visualization",scan when time allows,2025-06-03T13:55:05Z
issue,154791,[discussion] Specialized frontend method for computing Gram / covariance matrix,https://github.com/pytorch/pytorch/issues/154791,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: linear algebra, module: python frontend",scan when time allows,2025-06-03T13:52:37Z
issue,115122,[dynamo] Graph breaks from copy.deepcopy,https://github.com/pytorch/pytorch/issues/115122,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamo, module: graph breaks, oncall: pt2",scan when time allows,2025-06-02T19:23:18Z
issue,84520,MPS backend appears to be limited to 32 bits,https://github.com/pytorch/pytorch/issues/84520,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: mps,scan when time allows,2025-06-02T17:18:45Z
issue,150612,RAM leak during data loading with multiprocessing and Conv3d on CPU in Dataset __getitem__,https://github.com/pytorch/pytorch/issues/150612,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: data, module: dataloader, module: memory usage",scan when time allows,2025-06-02T08:46:02Z
issue,154836,[FSDP2] fix unit test test_all_gather_extension_outer_size_stride,https://github.com/pytorch/pytorch/issues/154836,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: fsdp, oncall: distributed",scan when time allows,2025-06-02T04:08:04Z
issue,148670,MPS Backend Error: ComplexDouble (complex128) Conversion Fails When Diffusers Transformer Creates 64‐bit Complex Tensors,https://github.com/pytorch/pytorch/issues/148670,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: complex, module: mps",scan when time allows,2025-06-02T03:31:20Z
issue,154489,[wishlist item] Assume data-dependent info based on tensor construction,https://github.com/pytorch/pytorch/issues/154489,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamic shapes, oncall: pt2",scan when time allows,2025-06-02T01:34:15Z
issue,154463,Mega Cache/ Torchcompile cache Not working,https://github.com/pytorch/pytorch/issues/154463,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: pt2,scan when time allows,2025-06-02T00:35:34Z
issue,111317,Torch 2.1 compile + FSDP (mixed precision) + LlamaForCausalLM: `RuntimeError: attempting to assign a gradient with dtype 'c10::BFloat16' to a tensor with dtype 'float'.`,https://github.com/pytorch/pytorch/issues/111317,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: fsdp, oncall: distributed, oncall: pt2",scan when time allows,2025-06-01T09:11:52Z
issue,154456,torch.compile tracing doesn't seem to be using the cache,https://github.com/pytorch/pytorch/issues/154456,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: compile-time, module: dynamo, oncall: pt2",scan when time allows,2025-06-01T07:58:57Z
issue,154474,Multiple torch.fft APIs produce inconsistent results for infinite inputs between CPU and GPU,https://github.com/pytorch/pytorch/issues/154474,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: edge cases, module: fft, module: nans and infs",scan when time allows,2025-05-31T00:27:59Z
issue,154083,torch.compile joint trace materializes constant tensors,https://github.com/pytorch/pytorch/issues/154083,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: aotdispatch, module: pt2-dispatcher, oncall: pt2",scan when time allows,2025-05-30T22:55:50Z
issue,73501,`index_copy` has different index behavior with `index_fill`,https://github.com/pytorch/pytorch/issues/73501,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: advanced indexing, module: edge cases",scan when time allows,2025-05-30T22:08:33Z
issue,150017,`__setitem__` with bool mask and dtype mismatch fails,https://github.com/pytorch/pytorch/issues/150017,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: advanced indexing, module: edge cases, module: type promotion",scan when time allows,2025-05-30T21:54:32Z
issue,154557,grouped_mm optional zero initialization of the output,https://github.com/pytorch/pytorch/issues/154557,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python frontend,scan when time allows,2025-05-30T16:23:22Z
issue,153861,[BUG] `functionalize` silent data corruption with pre-strided tensors,https://github.com/pytorch/pytorch/issues/153861,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: functorch,scan when time allows,2025-05-30T15:05:45Z
issue,69519,Feature Request: CUDA torch.histogram (and histogramdd),https://github.com/pytorch/pytorch/issues/69519,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, module: cuda, module: sorting and selection",scan when time allows,2025-05-30T08:44:07Z
issue,153334,Selective Activation Checkpointing on custom autograd.Function,https://github.com/pytorch/pytorch/issues/153334,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: activation checkpointing,scan when time allows,2025-05-29T21:40:48Z
issue,152566,[Benchmark] High compilation time variance on benchmark dashboards,https://github.com/pytorch/pytorch/issues/152566,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ci, module: infra",scan when time allows,2025-05-29T20:16:30Z
issue,154603,"RuntimeError: devptr INTERNAL ASSERT FAILED at ""/build/source/c10/cuda/CUDACachingAllocator.cpp"":3696, please report a bug to PyTorch. entry in cache has missing shared_ptr",https://github.com/pytorch/pytorch/issues/154603,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cudacachingallocator,scan when time allows,2025-05-29T18:18:55Z
issue,153957,MPS backend gradient correctness issues with large shapes,https://github.com/pytorch/pytorch/issues/153957,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: mps",scan when time allows,2025-05-29T17:30:42Z
issue,50192,GRU and LSTM fail for seq_len = 0,https://github.com/pytorch/pytorch/issues/50192,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: rnn",scan when time allows,2025-05-29T14:56:44Z
issue,107503,[nightly][jit] bad constant exponent (e+38.f) in default_program fused_mul_div_add,https://github.com/pytorch/pytorch/issues/107503,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2025-05-29T01:32:09Z
issue,154419,Crash in `torch.sparse.softmax`,https://github.com/pytorch/pytorch/issues/154419,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2025-05-28T22:01:03Z
issue,154496,torch.fft.irfft(n) doesn’t handle non-Hermitian inputs in a consistent way,https://github.com/pytorch/pytorch/issues/154496,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: complex, module: fft",scan when time allows,2025-05-28T21:00:44Z
issue,141133,CuDNN SDPA Issue Tracker,https://github.com/pytorch/pytorch/issues/141133,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: cudnn, module: sdpa",scan when time allows,2025-05-28T16:52:06Z
issue,154501,`torch.add` and `torch.sub` return incorrect results when computing complex numbers containing `inf` on both CPU and GPU,https://github.com/pytorch/pytorch/issues/154501,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: complex, module: nans and infs",scan when time allows,2025-05-28T13:31:40Z
issue,152295,"`vmap` not working on `torch.arange`, `torch.scalar_tensor`, and `torch.ones`",https://github.com/pytorch/pytorch/issues/152295,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: functorch, module: vmap",scan when time allows,2025-05-28T02:36:42Z
issue,154241,`RNNBase` modules break parameter sharing due to `flatten_parameters()`,https://github.com/pytorch/pytorch/issues/154241,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: rnn",scan when time allows,2025-05-27T20:51:13Z
issue,154248,[Testing][CUDA] test_cublas_addmm_reduced_precision_size_10000_backend_cublaslt_cuda_float16 failure with linux-focal-cuda12.6-py3-gcc11-slow-gradcheck,https://github.com/pytorch/pytorch/issues/154248,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cublas, module: cuda, module: half",scan when time allows,2025-05-27T16:35:37Z
issue,136750,Poor-quality random numbers generated by torch.poisson on gpus,https://github.com/pytorch/pytorch/issues/136750,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distributions,scan when time allows,2025-05-27T16:27:12Z
issue,151884,"The ""eager"" and ""aot_eager"" backends have different behavior for the expected gradient tensor of the torch.expend_as operator",https://github.com/pytorch/pytorch/issues/151884,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: aotdispatch, module: pt2-dispatcher, oncall: pt2",scan when time allows,2025-05-27T16:14:04Z
issue,117449,Deprecate and remove usage of from __future__ import annotations in codebase,https://github.com/pytorch/pytorch/issues/117449,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: typing,scan when time allows,2025-05-27T10:56:28Z
issue,109770,Slow performance when running torch.jit traced model with Flash Attention using libtorch on Windows,https://github.com/pytorch/pytorch/issues/109770,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: sdpa, module: windows",scan when time allows,2025-05-27T10:24:53Z
issue,154322,float16 → float32 conversion yields unexpected zero matrix for matrices > 43000 × 43000 in MPS,https://github.com/pytorch/pytorch/issues/154322,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: 64-bit, module: correctness (silent), module: mps",scan when time allows,2025-05-25T13:05:47Z
issue,144615,`collect_env.py` fails with `'NoneType' object has no attribute 'splitlines'` if pytorch is installed without pip,https://github.com/pytorch/pytorch/issues/144615,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: collect_env.py, module: devx",scan when time allows,2025-05-25T12:43:20Z
issue,127337,Add option for custom ops to automatically get a FakeTensor kernel (during static shapes),https://github.com/pytorch/pytorch/issues/127337,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: custom-operators, module: pt2-dispatcher, oncall: pt2",scan when time allows,2025-05-25T03:31:16Z
issue,154187,pybind11 loading for c10::Scalar NYI,https://github.com/pytorch/pytorch/issues/154187,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp-extensions,scan when time allows,2025-05-24T02:47:13Z
issue,130114,[feature request] [discussion] Baseline ONNX interpreter / executor in python / PyTorch,https://github.com/pytorch/pytorch/issues/130114,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2025-05-24T01:53:15Z
issue,154272,"[FSDP2] for mixed precision, input casting can get blocked when cuda streams are full",https://github.com/pytorch/pytorch/issues/154272,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: fsdp, oncall: distributed",scan when time allows,2025-05-24T01:03:40Z
issue,64208,"[feature request] ""Batched"" index_select (i.e. simplified torch.gather with not specifying full index)",https://github.com/pytorch/pytorch/issues/64208,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: scatter & gather ops,scan when time allows,2025-05-23T21:55:23Z
issue,120115,`detectron2_maskrcnn` OOMs on eager with A100 40G.,https://github.com/pytorch/pytorch/issues/120115,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2025-05-23T20:42:49Z
issue,128971,Trying to use forward AD with _scaled_dot_product_flash_attention that does not support it because it has not been implemented yet.,https://github.com/pytorch/pytorch/issues/128971,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2025-05-23T17:19:35Z
issue,154028,Add clamped FP8 (E4M3) cast for overflow-safe inference,https://github.com/pytorch/pytorch/issues/154028,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: floatx (formerly float8),scan when time allows,2025-05-23T16:12:33Z
issue,154096,CMake Error (CMAKE_MINIMUM_REQUIRED) when build 2.7.0,https://github.com/pytorch/pytorch/issues/154096,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: third_party",scan when time allows,2025-05-23T05:43:17Z
issue,153103,[Inductor] Investigate computing global amaxes via atomics instead of 2 stage reduction approach,https://github.com/pytorch/pytorch/issues/153103,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",scan when time allows,2025-05-23T04:08:29Z
issue,28777,cudnn.determinstic=True causes dilated convolution to be >10x slower,https://github.com/pytorch/pytorch/issues/28777,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cudnn,scan when time allows,2025-05-23T04:05:57Z
issue,149196,(Will PR) Multiprocessing with CUDA_VISIBLE_DEVICES seems to give the wrong device,https://github.com/pytorch/pytorch/issues/149196,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: multiprocessing",scan when time allows,2025-05-23T03:12:57Z
issue,154017,[RFC] Generalized Per-Operator Device Capability Registry for PyTorch Operator Testing,https://github.com/pytorch/pytorch/issues/154017,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: cuda, module: privateuse1, module: rocm",scan when time allows,2025-05-22T17:35:09Z
issue,106602,onednn ops supported in pytorch,https://github.com/pytorch/pytorch/issues/106602,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: intel,scan when time allows,2025-05-22T17:10:24Z
issue,127915,Document the torch.fx.traceback.preserve_node_meta function,https://github.com/pytorch/pytorch/issues/127915,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2025-05-22T17:10:20Z
issue,127886,Document the torch.fx.experimental.proxy_tensor.make_fx function,https://github.com/pytorch/pytorch/issues/127886,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2025-05-22T17:10:20Z
issue,127917,Document the torch.cuda.profiler.start function,https://github.com/pytorch/pytorch/issues/127917,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2025-05-22T17:10:19Z
issue,112587,"Fix docstring errors in __init__.py, _tensor_docs.py, _meta_registrations.py, _tensor.py",https://github.com/pytorch/pytorch/issues/112587,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2025-05-22T17:10:16Z
issue,112585,"Fix docstring errors in _VF.py, _appdirs.py, hub.py, _classes.py, _storage_docs.py, _linalg_utils.py, torch_version.py, quasirandom.py, random.py, __future__.py, _lowrank.py, _vmap_internals.py, _sources.py, __config__.py, _lobpcg.py, _namedtensor_internals.py",https://github.com/pytorch/pytorch/issues/112585,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2025-05-22T17:10:16Z
issue,112588,"Fix docstring errors in _torch_docs.py, serialization.py, overrides.py, _utils.py",https://github.com/pytorch/pytorch/issues/112588,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2025-05-22T17:10:15Z
issue,112600,Fix docstring errors in loss.py,https://github.com/pytorch/pytorch/issues/112600,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2025-05-22T17:10:14Z
issue,112641,"Fix docstring errors in post_localSGD_optimizer.py, functional_sgd.py, _functional_collectives.py, optimizer.py, utils.py, api.py, server_process_global_profiler.py, functions.py, backend_registry.py, __init__.py, options.py, internal.py, functional_adam.py",https://github.com/pytorch/pytorch/issues/112641,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2025-05-22T17:10:13Z
issue,112638,"Fix docstring errors in spectral_ops_fuzz_test.py, simple_timeit.py, timer_interface.py, op_benchmark.py, _stubs.py, fuzzer.py, compare.py, compile.py, interp.py, hipify_python.py, common.py, end_to_end.py, timer.py, __init__.py, sparse_fuzzer.py, blas_compare_setup.py",https://github.com/pytorch/pytorch/issues/112638,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2025-05-22T17:10:13Z
issue,112644,"Fix docstring errors in default_hooks.py, post_localSGD_hook.py, debugging_hooks.py, utils.py, hierarchical_model_averager.py, optimizer_overlap_hooks.py, mixed_precision_hooks.py, quantization_hooks.py, ddp_zero_hook.py, __init__.py, powerSGD_hook.py, averagers.py",https://github.com/pytorch/pytorch/issues/112644,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2025-05-22T17:10:12Z
issue,112646,"Fix docstring errors in stream.py, pipe.py, blockpartition.py, microbatch.py, namespace.py, profile.py, tracker.py, portal.py, layout.py, __init__.py",https://github.com/pytorch/pytorch/issues/112646,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2025-05-22T17:10:11Z
issue,113186,"Fix docstring errors in shard.py, op_registry_utils.py, local_timer.py, file_based_local_timer.py, api.py, distributed.py, cycling_iterator.py, __init__.py, _utils.py, reshard.py, common_op_utils.py, elastic_distributed_sampler.py, utils.py, log_level.py, store.py, logging.py, sharder.py, metadata.py",https://github.com/pytorch/pytorch/issues/113186,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2025-05-22T17:10:10Z
issue,113122,Request to create a tutorial for loading a model dumped by `torch.export`,https://github.com/pytorch/pytorch/issues/113122,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: export,scan when time allows,2025-05-22T17:10:10Z
issue,113187,"Fix docstring errors in embedding.py, _limiter_utils.py, _dynamo_utils.py, embedding_bag.py, tensor_ops.py, api.py, _internals.py, _common.py, init.py, _exec_order_utils.py, _traversal_utils.py, chunk_sharding_spec.py, _trace_utils.py",https://github.com/pytorch/pytorch/issues/113187,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2025-05-22T17:10:09Z
issue,113188,"Fix docstring errors in _common_utils.py, _optim_utils.py, _wrap_utils.py, _unshard_param_utils.py, _fsdp_extensions.py, api.py, _debug_utils.py, _utils.py, wrap.py, sharded_grad_scaler.py",https://github.com/pytorch/pytorch/issues/113188,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2025-05-22T17:10:08Z
issue,113192,"Fix docstring errors in checkpoint_example.py, basic_strategy.py, op_schema.py, contract.py, redistribute.py, __init__.py, api.py, device_mesh.py, _utils.py, sharding_prop.py, random.py, checkpoint_activation.py, placement_types.py, fully_shard.py, parallel_mode.py, replicate.py",https://github.com/pytorch/pytorch/issues/113192,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2025-05-22T17:10:07Z
issue,113190,"Fix docstring errors in _state_dict_utils.py, _runtime_utils.py, _shard_utils.py",https://github.com/pytorch/pytorch/issues/113190,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2025-05-22T17:10:07Z
issue,154086,[cuBLAS] Strange behavior with `out_dtype` argument in `torch.addmm`,https://github.com/pytorch/pytorch/issues/154086,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bfloat16, module: cublas, module: cuda, module: half",scan when time allows,2025-05-22T16:24:35Z
issue,102559,Mergebot should merge non-stacked PR,https://github.com/pytorch/pytorch/issues/102559,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: devx,scan when time allows,2025-05-22T13:31:11Z
issue,153471,RFC for vector length agnostic SVE Vectorized class,https://github.com/pytorch/pytorch/issues/153471,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: arm, module: vectorization",scan when time allows,2025-05-22T10:57:12Z
issue,148451,BC-linter should ignore testing/linter/adapters/,https://github.com/pytorch/pytorch/issues/148451,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: lint,scan when time allows,2025-05-22T10:22:55Z
issue,113761,[DTensor][TP] 1-way TP raises index out of range,https://github.com/pytorch/pytorch/issues/113761,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dtensor,scan when time allows,2025-05-22T10:07:12Z
issue,152150,"[Scaled MM] Update to support on B200 TN, NT, NN, TT Layouts are supported",https://github.com/pytorch/pytorch/issues/152150,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: floatx (formerly float8), module: performance",scan when time allows,2025-05-21T18:48:56Z
issue,147428,Add a config to allow print and generate all recompile reasons and not stop at first.,https://github.com/pytorch/pytorch/issues/147428,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: compile ux, module: dynamo, oncall: pt2",scan when time allows,2025-05-21T18:20:31Z
issue,122550,Adam optimizer：RuntimeError: t == DeviceType::CUDA INTERNAL ASSERT FAILED,https://github.com/pytorch/pytorch/issues/122550,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2025-05-21T17:24:49Z
issue,153638,Can't run lintrunner locally on Windows machine,https://github.com/pytorch/pytorch/issues/153638,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: lint, module: windows",scan when time allows,2025-05-21T04:33:55Z
issue,153649,Use opmath_t and not double compute in fused optimizers,https://github.com/pytorch/pytorch/issues/153649,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: optimizer, module: performance",scan when time allows,2025-05-20T20:57:09Z
issue,153842,Duplicated `milestones` keys in `torch.optim.lr_scheduler.MultiStepLr` after loading a checkpoint,https://github.com/pytorch/pytorch/issues/153842,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: lrscheduler, oncall: distributed checkpointing",scan when time allows,2025-05-20T18:50:28Z
issue,153832,export torch.cond fails at capturing identity,https://github.com/pytorch/pytorch/issues/153832,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: oncall: export, oncall: pt2",scan when time allows,2025-05-20T15:55:21Z
issue,153930,[FX] Support `torch.cond` in `torch.fx.symbolic_trace`,https://github.com/pytorch/pytorch/issues/153930,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2025-05-20T15:43:32Z
issue,153826,Unable to reproduce dynamo bechmark script results,https://github.com/pytorch/pytorch/issues/153826,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: benchmark, module: rocm, module: tests",scan when time allows,2025-05-20T15:28:53Z
issue,153938,Subtests not properly accounted for in test reports (e.g. on CI),https://github.com/pytorch/pytorch/issues/153938,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ci, module: testing",scan when time allows,2025-05-20T14:42:07Z
issue,153532,[RFC] [Inductor] Custom pass registration interface,https://github.com/pytorch/pytorch/issues/153532,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",scan when time allows,2025-05-20T13:24:38Z
issue,118173,ModuleNotFoundError: pynvml does not seem to be installed or it can't be imported.,https://github.com/pytorch/pytorch/issues/118173,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: cuda, module: windows",scan when time allows,2025-05-20T11:50:35Z
issue,133984,cudagraph capture p2p ops fail,https://github.com/pytorch/pytorch/issues/133984,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda graphs, oncall: distributed",scan when time allows,2025-05-20T09:12:13Z
issue,111187,"torch.distributed.DistBackendError: NCCL error in: ../torch/csrc/distributed/c10d/ProcessGroupNCCL.cpp:1331, unhandled cuda error (run with NCCL_DEBUG=INFO for details)",https://github.com/pytorch/pytorch/issues/111187,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: wsl, oncall: distributed",scan when time allows,2025-05-20T06:22:54Z
issue,8374,[Feature request] Add torch.multiprocessing.Pipe,https://github.com/pytorch/pytorch/issues/8374,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, module: multiprocessing",scan when time allows,2025-05-20T03:54:36Z
issue,148065,Support alpha=inf consistently for torch.celu,https://github.com/pytorch/pytorch/issues/148065,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: python frontend",scan when time allows,2025-05-20T02:43:42Z
issue,76071,HIPFFT_EXEC_FAILED when using AMD GPU run FFT,https://github.com/pytorch/pytorch/issues/76071,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: fft",scan when time allows,2025-05-19T12:28:48Z
issue,153542,torch.set_ on a view does not sever view relation,https://github.com/pytorch/pytorch/issues/153542,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: viewing and reshaping,scan when time allows,2025-05-19T08:53:10Z
issue,153594,Significant DataLoader worker startup delay and sequential-like initialization with num_workers > 0 on Windows (using 'spawn'),https://github.com/pytorch/pytorch/issues/153594,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dataloader, module: performance, module: windows",scan when time allows,2025-05-19T04:41:19Z
issue,105325,Padded tensor subclass,https://github.com/pytorch/pytorch/issues/105325,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nestedtensor,scan when time allows,2025-05-18T20:48:28Z
issue,150861,Allow non-Tensor subclass in `torch.Tensor._make_wrapper_subclass`,https://github.com/pytorch/pytorch/issues/150861,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2025-05-18T10:52:58Z
issue,105644,Tensor subclass is not preserved during backward with gradient checkpointing,https://github.com/pytorch/pytorch/issues/105644,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: __torch_function__, module: activation checkpointing",scan when time allows,2025-05-17T20:07:02Z
issue,153606,native/indexingUtils.h `AdvancedIndex` has internal linkage,https://github.com/pytorch/pytorch/issues/153606,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: static linking",scan when time allows,2025-05-16T17:01:37Z
issue,73016,Unable to build and use libtorch function via pybind11: undefined symbol error upon import,https://github.com/pytorch/pytorch/issues/73016,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: mkldnn",scan when time allows,2025-05-16T10:10:19Z
issue,110056,torch.onnx.export causes floating point exception with core dump for empty slice assignment,https://github.com/pytorch/pytorch/issues/110056,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2025-05-16T06:30:02Z
issue,150476,Better test coverage on _inductor/scheduler.py,https://github.com/pytorch/pytorch/issues/150476,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",scan when time allows,2025-05-16T04:26:50Z
issue,153673,[FSDP2] revisit NCCL group coalescing vs copy-in/copy-out,https://github.com/pytorch/pytorch/issues/153673,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2025-05-16T00:48:28Z
issue,152963,DTensor support for dynamic shapes is soft,https://github.com/pytorch/pytorch/issues/152963,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: oncall: distributed, oncall: pt2",scan when time allows,2025-05-15T22:54:18Z
issue,146956,[c10d] Consolidate watchdog threads,https://github.com/pytorch/pytorch/issues/146956,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: c10d, oncall: distributed",scan when time allows,2025-05-15T22:42:51Z
issue,151037,[release] CPU perf benchmark latency increase for 2.6->2.7 on c5.24xlarge and A100 instances,https://github.com/pytorch/pytorch/issues/151037,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: intel, module: performance, oncall: releng",scan when time allows,2025-05-15T16:08:53Z
issue,143176,Compiled Autograd + Activation Checkpointing/Offloading,https://github.com/pytorch/pytorch/issues/143176,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: activation checkpointing, module: compiled autograd, oncall: pt2",scan when time allows,2025-05-14T18:35:13Z
issue,90633,torch.min document not up to date,https://github.com/pytorch/pytorch/issues/90633,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2025-05-14T18:15:30Z
issue,151912,bfloat16 numerical errors for SDPA math backend,https://github.com/pytorch/pytorch/issues/151912,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sdpa,scan when time allows,2025-05-14T15:50:05Z
issue,152829,Don't hardcoded support for DTensor to_local/from_local/redistribute into dynamo,https://github.com/pytorch/pytorch/issues/152829,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: distributed, oncall: pt2",scan when time allows,2025-05-13T22:55:47Z
issue,153302,[FSDP2] avoid GPU OOM for reshard_after_forward=int with shared post_forward_mesh,https://github.com/pytorch/pytorch/issues/153302,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2025-05-13T18:20:29Z
issue,152337,Unusually slow draft_export time,https://github.com/pytorch/pytorch/issues/152337,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: oncall: export, oncall: pt2",scan when time allows,2025-05-13T18:09:22Z
issue,153242,[RFC] Add a SlimTensor representation to help AOTInductor generate standalone binaries,https://github.com/pytorch/pytorch/issues/153242,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: pt2,scan when time allows,2025-05-13T16:12:07Z
issue,153139,Floating Point exception in Convolution with disabled SMT,https://github.com/pytorch/pytorch/issues/153139,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: dependency bug, module: nn, module: nnpack",scan when time allows,2025-05-13T09:09:02Z
issue,153360,torch.aminmax reducing over all dims should not accept keepdim,https://github.com/pytorch/pytorch/issues/153360,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: reductions,scan when time allows,2025-05-13T02:04:04Z
issue,153361,MANIFEST.in is not being used,https://github.com/pytorch/pytorch/issues/153361,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2025-05-13T01:00:35Z
issue,136946,DTensor does not support `nn.init.eye_`,https://github.com/pytorch/pytorch/issues/136946,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dtensor,scan when time allows,2025-05-12T20:26:39Z
issue,121160,RFCS review request: fast Viterbi Decoding,https://github.com/pytorch/pytorch/issues/121160,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2025-05-12T14:30:58Z
issue,91252,Open file leak when dataloader is using persistent_workers and pin_memory AND you create multiple dataloaders.,https://github.com/pytorch/pytorch/issues/91252,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: data, module: dataloader",scan when time allows,2025-05-12T12:54:55Z
issue,98115,Request to cherrypick a fix into v1.13.1 (v1.8 has a CVE),https://github.com/pytorch/pytorch/issues/98115,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: binaries,scan when time allows,2025-05-12T08:58:33Z
issue,72643,upstream `apex.normalization.FusedRMSNorm`,https://github.com/pytorch/pytorch/issues/72643,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: norms and normalization",scan when time allows,2025-05-12T06:34:14Z
issue,134576,Would it be possible to have a CMAKE option USE_SYSTEM_FMT,https://github.com/pytorch/pytorch/issues/134576,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2025-05-11T18:15:02Z
issue,73466,[feature request] torch.hub.load_state_dict_from_url to be replaced by a new good general download-a-file function and to also support local paths and google drive links / private github release links,https://github.com/pytorch/pytorch/issues/73466,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: hub,scan when time allows,2025-05-11T10:23:08Z
issue,147179,[FSDP2] OOM when use integer `reshard_after_forward` that smaller than DP size,https://github.com/pytorch/pytorch/issues/147179,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: fsdp, oncall: distributed",scan when time allows,2025-05-10T01:27:05Z
issue,150864,[dynamo] `torch.compile` graph breaks on `setattr` of type objects,https://github.com/pytorch/pytorch/issues/150864,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",scan when time allows,2025-05-09T20:32:59Z
issue,153041,gen_alias_from_base ruins the result of view after inductor generated a copy for the results of the view operations.,https://github.com/pytorch/pytorch/issues/153041,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2025-05-09T20:06:19Z
issue,152746,"[FSDP2] fully_shard(mesh=(shard, shard)) for intra and inter node all-gathers",https://github.com/pytorch/pytorch/issues/152746,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2025-05-09T19:26:43Z
issue,153210,Inconsistent Error Handling in `torch.addcmul` with `int16` Tensors,https://github.com/pytorch/pytorch/issues/153210,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: edge cases, module: error checking, module: int overflow",scan when time allows,2025-05-09T15:28:21Z
issue,107999,`upsample_bilinear2d_backward_out_cuda` is nondeterministic,https://github.com/pytorch/pytorch/issues/107999,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: determinism,scan when time allows,2025-05-09T06:58:02Z
issue,148086,FSDP2 without sharding works slower than DDP,https://github.com/pytorch/pytorch/issues/148086,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2025-05-08T23:49:13Z
issue,76082,Eliminate uses of deprecated `FindCUDA.cmake`,https://github.com/pytorch/pytorch/issues/76082,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cuda",scan when time allows,2025-05-08T14:14:19Z
issue,123447,dist.barrier() hangs after calling async_save,https://github.com/pytorch/pytorch/issues/123447,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2025-05-08T07:24:07Z
issue,148306,torch.vmap incompatibility with DLPack functions,https://github.com/pytorch/pytorch/issues/148306,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dlpack, module: functorch, module: vmap",scan when time allows,2025-05-08T07:05:18Z
issue,153025,Multiple CUDA graphs utilizing multiple CUDA GPUs encounter illegal memory access during replay,https://github.com/pytorch/pytorch/issues/153025,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda graphs,scan when time allows,2025-05-07T19:09:39Z
issue,152962,TestNestedTensorOpInfoCUDA.test_compile_backward_matmul_cuda_float32 Test Failure,https://github.com/pytorch/pytorch/issues/152962,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nestedtensor, module: tests",scan when time allows,2025-05-07T16:58:53Z
issue,150836,`Aborted (core dumped)` in `torch.cuda.nccl.reduce`,https://github.com/pytorch/pytorch/issues/150836,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2025-05-07T10:10:32Z
issue,151322,[CI] No workflows scheduled on PRs,https://github.com/pytorch/pytorch/issues/151322,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ci, module: flaky-tests, module: third_party",scan when time allows,2025-05-06T22:28:04Z
issue,152994,[dynamo] Actually support functools.lru_cache,https://github.com/pytorch/pytorch/issues/152994,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",scan when time allows,2025-05-06T21:20:16Z
issue,152884,UNSTABLE Lint / Link checks / Lint URLs / linux-job,https://github.com/pytorch/pytorch/issues/152884,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2025-05-06T17:14:25Z
issue,152426,"[Manylinux 2.28] Migrate Docker container to use gcc 13, CUDA 12.6 and gcc14 CUDA 12.8",https://github.com/pytorch/pytorch/issues/152426,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: binaries,scan when time allows,2025-05-06T17:01:03Z
issue,152869,DISABLED testAssertNotRegex (__main__.CPythonTest_Assertions),https://github.com/pytorch/pytorch/issues/152869,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamo, module: flaky-tests, oncall: pt2",scan when time allows,2025-05-06T12:55:34Z
issue,152752,Checkpoint sequential doesn't raise clear error when segments is greater than number of functions,https://github.com/pytorch/pytorch/issues/152752,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: activation checkpointing,scan when time allows,2025-05-06T12:26:50Z
issue,58846,Simplification of pruned models,https://github.com/pytorch/pytorch/issues/58846,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2025-05-06T07:51:58Z
issue,152736,[MPS] TensorIterator and accuracy,https://github.com/pytorch/pytorch/issues/152736,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: mps,scan when time allows,2025-05-05T18:13:21Z
issue,152527,can't reconstruct the communication group using PyTorch.,https://github.com/pytorch/pytorch/issues/152527,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: c10d, oncall: distributed",scan when time allows,2025-05-05T08:24:12Z
issue,152645,ProcessGroupGloo.allgather_into_tensor_coalesced crashes with CUDA tensors,https://github.com/pytorch/pytorch/issues/152645,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2025-05-05T07:41:30Z
issue,66200,An exported onnx model can't reduce on dim with value of 0 if 'keepdims' is false,https://github.com/pytorch/pytorch/issues/66200,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2025-05-04T08:50:41Z
issue,151182,A problem discovered when computing complex matrices in deep neural networks,https://github.com/pytorch/pytorch/issues/151182,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: complex, module: nans and infs",scan when time allows,2025-05-03T09:03:18Z
issue,121857,AdamW(fused=True) slower than unfused AdamW,https://github.com/pytorch/pytorch/issues/121857,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: optimizer, module: performance",scan when time allows,2025-05-02T21:41:53Z
issue,117540,Unwanted Warning in lr_scheduler.step(),https://github.com/pytorch/pytorch/issues/117540,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: deprecation, module: lrscheduler",scan when time allows,2025-05-02T21:38:46Z
issue,151936,fully_shard() for huggingface 72B model: pytorch caches too much GPU memory,https://github.com/pytorch/pytorch/issues/151936,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dtensor, oncall: distributed",scan when time allows,2025-05-02T19:28:29Z
issue,114357,Pytorch DDP across nodes: self._store = TCPStore( # type: ignore[call-arg] RuntimeError: Stop_waiting response is expected,https://github.com/pytorch/pytorch/issues/114357,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2025-05-02T18:43:21Z
issue,152699,[CI] [anaconda] Triton windows build,https://github.com/pytorch/pytorch/issues/152699,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2025-05-02T17:22:35Z
issue,152607,Loops impacting output when utilizing hooks,https://github.com/pytorch/pytorch/issues/152607,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2025-05-02T13:34:44Z
issue,152647,Check for if two tensors are overall similar instead of bitwise similar?,https://github.com/pytorch/pytorch/issues/152647,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: testing,scan when time allows,2025-05-02T13:30:47Z
issue,9674,The state of sparse Tensors,https://github.com/pytorch/pytorch/issues/9674,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2025-05-01T23:53:10Z
issue,147570,`view()` + modify-in-place fails silently with DTensor,https://github.com/pytorch/pytorch/issues/147570,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dtensor, oncall: distributed",scan when time allows,2025-05-01T20:29:20Z
issue,152097,"When using torch to convert to oxxn model, testing the inference results with actual images shows tensor mismatch",https://github.com/pytorch/pytorch/issues/152097,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2025-05-01T18:34:36Z
issue,152550,MPS varying seq len SDPA memory leak,https://github.com/pytorch/pytorch/issues/152550,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: memory usage, module: mps, module: sdpa",scan when time allows,2025-05-01T17:50:30Z
issue,152437,`nn.CrossEntropyLoss` accepts negative target probabilities,https://github.com/pytorch/pytorch/issues/152437,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: error checking, module: nn, module: performance",scan when time allows,2025-05-01T14:53:50Z
issue,127928,Ability to do aot/inductor compilation from a jit model (or torch.exported model),https://github.com/pytorch/pytorch/issues/127928,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: aotinductor, oncall: export, oncall: pt2",scan when time allows,2025-05-01T12:05:20Z
issue,126518,missing docs for torch.Tag,https://github.com/pytorch/pytorch/issues/126518,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: library",scan when time allows,2025-04-30T20:47:05Z
issue,129657,Implementation of a numerically stable log(1 - softmax) function in PyTorch,https://github.com/pytorch/pytorch/issues/129657,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: numerical-stability, module: python frontend",scan when time allows,2025-04-30T15:17:04Z
issue,67970,Error when using torch.fx on bert,https://github.com/pytorch/pytorch/issues/67970,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2025-04-30T01:38:26Z
issue,152489,UNSTABLE Lint / Lint URLs / linux-job,https://github.com/pytorch/pytorch/issues/152489,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2025-04-30T00:22:28Z
issue,152127,[C10D] Make collectives backwards throw an error,https://github.com/pytorch/pytorch/issues/152127,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2025-04-30T00:03:20Z
issue,151579,[BE] Consolidate PR labeling logic,https://github.com/pytorch/pytorch/issues/151579,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, module: devx",scan when time allows,2025-04-29T21:02:57Z
issue,152257,[FR] Support BSHM-layout scaled_dot_product_attention without transpose.,https://github.com/pytorch/pytorch/issues/152257,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sdpa,scan when time allows,2025-04-29T19:56:32Z
issue,152381,Outdated contents in dynamo overview,https://github.com/pytorch/pytorch/issues/152381,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2025-04-29T00:22:57Z
issue,151776,[dynamo] Some inefficiencies around handling __torch_function__,https://github.com/pytorch/pytorch/issues/151776,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",scan when time allows,2025-04-28T23:45:10Z
issue,152269,Arbitrary Code Execution Risk in `torch.distributed.utils.overload` When Misused in Type Annotations,https://github.com/pytorch/pytorch/issues/152269,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2025-04-28T22:24:39Z
issue,117350,CUDA deps cannot be preloaded under Bazel,https://github.com/pytorch/pytorch/issues/117350,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bazel, module: build",scan when time allows,2025-04-28T22:13:31Z
issue,107573,caching keys+values in TransformerDecoderLayer for faster inference,https://github.com/pytorch/pytorch/issues/107573,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2025-04-26T17:51:07Z
issue,106624,"[feature request] [ux] Frontend methods for fused elementwise affine transform: mul+add+dtype convert + support `integer_tensor.mul_(float_constant)` and `float_tensor.mul(some_constant, out = integer_tensor)` maybe via new args `rounding_mode=...` and `dtype=...` + maybe support OpenCV-style saturated dtype conversions (e.g. `clamp_` before conversion)",https://github.com/pytorch/pytorch/issues/106624,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: python frontend, module: type promotion",scan when time allows,2025-04-26T10:18:31Z
issue,57947,`torch.lerp` to support argument type promotion / broadcasting - including of `input` / `end` arguments,https://github.com/pytorch/pytorch/issues/57947,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: interpolation, module: type promotion",scan when time allows,2025-04-26T10:04:34Z
issue,146913,Add a `TORCH_LOGS_RANK=0` env var that integrates with `TORCH_LOGS`,https://github.com/pytorch/pytorch/issues/146913,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: logging, oncall: pt2",scan when time allows,2025-04-26T02:52:37Z
issue,150574,[mem profiler] mem fragmentation and pynvml view,https://github.com/pytorch/pytorch/issues/150574,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: cudacachingallocator, module: memory usage",scan when time allows,2025-04-25T23:43:51Z
issue,152152,padding_mode `reflect` works different from others in Conv,https://github.com/pytorch/pytorch/issues/152152,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: padding",scan when time allows,2025-04-25T19:59:15Z
issue,152188,The input for layers other than the first layer should be the hidden state from the previous layer.,https://github.com/pytorch/pytorch/issues/152188,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: rnn",scan when time allows,2025-04-25T19:57:21Z
issue,151592,Xcode 16+: duplicate LC_RPATH '@loader_path',https://github.com/pytorch/pytorch/issues/151592,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2025-04-25T10:11:40Z
issue,141094,plus prefix-less wheels not being published on https://download.pytorch.org/whl/torch/ for linux x86_64,https://github.com/pytorch/pytorch/issues/141094,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: binaries,scan when time allows,2025-04-24T22:00:57Z
issue,151960,[rfc][c10d] First Class Object-Oriented ProcessGroup and Store APIs,https://github.com/pytorch/pytorch/issues/151960,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: c10d, oncall: distributed",scan when time allows,2025-04-24T18:53:47Z
issue,143759,compiled autograd tests should use expecttest,https://github.com/pytorch/pytorch/issues/143759,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: compiled autograd, module: tests, oncall: pt2",scan when time allows,2025-04-24T18:18:49Z
issue,145693,AMD MI300A Unified Memory Support,https://github.com/pytorch/pytorch/issues/145693,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rocm,scan when time allows,2025-04-24T17:56:44Z
issue,151978,torch.gather returns incorrect output on cuda after unsqueezing and expanding with double precision,https://github.com/pytorch/pytorch/issues/151978,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: scatter & gather ops,scan when time allows,2025-04-24T16:50:32Z
issue,151858,construcing DTensor on a 2D device mesh SIGTERMs,https://github.com/pytorch/pytorch/issues/151858,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dtensor, oncall: distributed",scan when time allows,2025-04-24T15:50:12Z
issue,108108,[BC BREAKING] Change default behavior of scaled_dot_product_attention's causal masking alignment,https://github.com/pytorch/pytorch/issues/108108,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bc-breaking, module: nn, module: sdpa",scan when time allows,2025-04-24T09:41:13Z
issue,151878,Potential error in calling scaled_dot_product_attention in pytorch/torch/testing/_internal/distributed/_tensor /common_dtensor.py ?,https://github.com/pytorch/pytorch/issues/151878,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: context parallel, module: dtensor, oncall: distributed",scan when time allows,2025-04-24T06:37:51Z
issue,145263,[Pipelining] Problem using `torch.distributed.pipelining` on `Gemma2ForCausalLM`,https://github.com/pytorch/pytorch/issues/145263,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2025-04-23T21:02:31Z
issue,151759,"""_get_pg_default_device"" deprecated warning in ""Getting Started with Distributed Checkpoint (DCP)""",https://github.com/pytorch/pytorch/issues/151759,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2025-04-23T20:54:49Z
issue,148257,[FSDP2] HSDP with globally sharded fp32 weights and optimizer states,https://github.com/pytorch/pytorch/issues/148257,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: fsdp, oncall: distributed",scan when time allows,2025-04-23T20:54:49Z
issue,148269,Should DTensor support `Shard()` placement without dim requirement?,https://github.com/pytorch/pytorch/issues/148269,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dtensor, oncall: distributed",scan when time allows,2025-04-23T20:54:48Z
issue,148162,Specifying device_id in init_process_group causes tensor parallel + pipeline parallel to fail,https://github.com/pytorch/pytorch/issues/148162,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2025-04-23T20:54:47Z
issue,148150,[pytorch elastic] [RHEL] multiple processes call to dist.destroy_process_group() cause an RuntimeError: CUDA error: CUDA-capable device(s) is/are busy or unavailable,https://github.com/pytorch/pytorch/issues/148150,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2025-04-23T20:54:47Z
issue,148048,[RFC] Customization of ElasticAgent for fault-tolerance and node-replacement in big ddp job,https://github.com/pytorch/pytorch/issues/148048,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2025-04-23T20:54:46Z
issue,147980,Torch distribute module when local dtensor is already on the correct device,https://github.com/pytorch/pytorch/issues/147980,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dtensor, oncall: distributed",scan when time allows,2025-04-23T20:54:45Z
issue,147793,context_parallel fails with plain sdpa kernel SDPBackend.MATH,https://github.com/pytorch/pytorch/issues/147793,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: context parallel, module: sdpa, oncall: distributed",scan when time allows,2025-04-23T20:54:44Z
issue,147732,[Distribute] len(input_specs) == len(input_args_strategy) AssertionError,https://github.com/pytorch/pytorch/issues/147732,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2025-04-23T20:54:43Z
issue,147677,distributed_c10d.broadcast causing unexpected CUDA oom error,https://github.com/pytorch/pytorch/issues/147677,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: c10d, oncall: distributed",scan when time allows,2025-04-23T20:54:43Z
issue,147631,torch.distributed.init_process_group with unspecified backend out of date,https://github.com/pytorch/pytorch/issues/147631,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: c10d, oncall: distributed",scan when time allows,2025-04-23T20:54:42Z
issue,147563,torch.distributed.elastic.multiprocessing.start_process description does not reflect API,https://github.com/pytorch/pytorch/issues/147563,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: elastic, oncall: distributed",scan when time allows,2025-04-23T20:54:42Z
issue,147502,Slow first time torch.distributed op with ray,https://github.com/pytorch/pytorch/issues/147502,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: c10d, oncall: distributed",scan when time allows,2025-04-23T20:54:41Z
issue,147348,[pipelining] AttributeError: 'InterpreterModule' object has no attribute,https://github.com/pytorch/pytorch/issues/147348,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: pipelining, oncall: distributed",scan when time allows,2025-04-23T20:54:41Z
issue,147263,How to trigger several independent communications simultaneously?,https://github.com/pytorch/pytorch/issues/147263,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2025-04-23T20:54:39Z
issue,147141,NCCL Update 2.25.1 with CUDA 12.4 build is failing in CI,https://github.com/pytorch/pytorch/issues/147141,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nccl, oncall: distributed",scan when time allows,2025-04-23T20:54:37Z
issue,149177,[Dist] Async op isend and irecv bug,https://github.com/pytorch/pytorch/issues/149177,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: c10d, oncall: distributed",scan when time allows,2025-04-23T20:42:57Z
issue,149100,Memory leak when using get_model_state_dict with FSDP-sharded models,https://github.com/pytorch/pytorch/issues/149100,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: fsdp, oncall: distributed",scan when time allows,2025-04-23T20:42:57Z
issue,148831,FSDP2 and autocast compatibility issue,https://github.com/pytorch/pytorch/issues/148831,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: fsdp, oncall: distributed",scan when time allows,2025-04-23T20:42:57Z
issue,148756,FSPD ValueError: expected to be in states [<TrainingState.FORWARD_BACKWARD: 2>] but current state is TrainingState.IDLE,https://github.com/pytorch/pytorch/issues/148756,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: fsdp, oncall: distributed",scan when time allows,2025-04-23T20:42:56Z
issue,149049,"`pytroch.distribute` should support ""meta"" device tensors",https://github.com/pytorch/pytorch/issues/149049,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2025-04-23T20:42:55Z
issue,148837,NVLS support in Pytorch,https://github.com/pytorch/pytorch/issues/148837,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2025-04-23T20:42:55Z
issue,148504,[FSDP2] improve error msg for duplicate wraps,https://github.com/pytorch/pytorch/issues/148504,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: fsdp, oncall: distributed",scan when time allows,2025-04-23T20:42:53Z
issue,148689,NUMA Binding Integration with torchrun,https://github.com/pytorch/pytorch/issues/148689,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2025-04-23T20:42:52Z
issue,148440,torch.distributed hangs between 2 Mac Devices,https://github.com/pytorch/pytorch/issues/148440,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: macos, oncall: distributed",scan when time allows,2025-04-23T20:42:51Z
issue,148302,broadcast_object_list not release GPU,https://github.com/pytorch/pytorch/issues/148302,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2025-04-23T20:42:48Z
issue,151457,"When distributed.destroy_process_group() is executed, new memory usage will be generated on device 0, which may cause OOM under extreme conditions and thus abnormal exit.",https://github.com/pytorch/pytorch/issues/151457,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2025-04-23T20:37:15Z
issue,151240,[funcol] wait() breaks the chain for backwards,https://github.com/pytorch/pytorch/issues/151240,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2025-04-23T20:37:15Z
issue,150585,[distributed] Crash when trying to use default PG after creating new PG,https://github.com/pytorch/pytorch/issues/150585,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2025-04-23T20:37:13Z
issue,150096,FSDP OOM during `sync_params_and_buffers`,https://github.com/pytorch/pytorch/issues/150096,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: fsdp, oncall: distributed",scan when time allows,2025-04-23T20:37:12Z
issue,150336,[parallelize_module] uneven sharding + use_local_output breaks,https://github.com/pytorch/pytorch/issues/150336,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dtensor, oncall: distributed",scan when time allows,2025-04-23T20:37:11Z
issue,150140,"FSDP2 issue with mp_policy, checkpoint() and float input",https://github.com/pytorch/pytorch/issues/150140,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: fsdp, oncall: distributed",scan when time allows,2025-04-23T20:37:10Z
issue,150049,2.7 test docker image has NCCL with older CUDA,https://github.com/pytorch/pytorch/issues/150049,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docker, oncall: distributed, oncall: releng",scan when time allows,2025-04-23T20:37:09Z
issue,149984,"ncclSystemError: System call (e.g. socket, malloc) or external library call failed or device error",https://github.com/pytorch/pytorch/issues/149984,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nccl, oncall: distributed",scan when time allows,2025-04-23T20:37:08Z
issue,149934,inconsistant behavior to pass tensor/tensor_list from torch.distributed module to cpp API,https://github.com/pytorch/pytorch/issues/149934,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2025-04-23T20:37:07Z
issue,149762,terminate called after throwing an instance of 'c10::Error',https://github.com/pytorch/pytorch/issues/149762,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ddp, oncall: distributed",scan when time allows,2025-04-23T20:37:06Z
issue,149516,```StateDictOptions``` in combination with ```cpu_offload=True``` and ```strict=False``` not working,https://github.com/pytorch/pytorch/issues/149516,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2025-04-23T20:37:05Z
issue,151556,"Compatibility Issues Between DCP, LoRA, and Hugging Face Models",https://github.com/pytorch/pytorch/issues/151556,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2025-04-23T20:35:49Z
issue,150943,"[RFC][TCPStore] advanced store operations (queues, pub/sub)",https://github.com/pytorch/pytorch/issues/150943,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: c10d, oncall: distributed",scan when time allows,2025-04-23T20:35:12Z
issue,151159,Failed to destroy or init process group after calling _abort_process_group,https://github.com/pytorch/pytorch/issues/151159,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2025-04-23T20:34:11Z
issue,151269,[PP] Hang when num_microbatches < stages for Interleaved1F1B,https://github.com/pytorch/pytorch/issues/151269,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: pipelining, oncall: distributed",scan when time allows,2025-04-23T20:27:04Z
issue,151222,[FSDP] Detail information about parameters when raising errors,https://github.com/pytorch/pytorch/issues/151222,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2025-04-23T20:09:31Z
issue,151965,"[rfc][c10d] RDMA APIs (read/write, rkey)",https://github.com/pytorch/pytorch/issues/151965,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: c10d, oncall: distributed",scan when time allows,2025-04-23T18:46:24Z
issue,151717,[NJT] Dropout with p=0 still drops values,https://github.com/pytorch/pytorch/issues/151717,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nestedtensor,scan when time allows,2025-04-23T14:27:03Z
issue,4474,[docs] Return value type hint to be specified in docs for torch.is_grad_enabled (i.e. -> bool),https://github.com/pytorch/pytorch/issues/4474,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2025-04-23T09:52:15Z
issue,21176,Adding a method called `T` in native_functions causes undefined behavior on Windows,https://github.com/pytorch/pytorch/issues/21176,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: windows,scan when time allows,2025-04-23T04:47:10Z
issue,151793,Create decomp for searchsorted,https://github.com/pytorch/pytorch/issues/151793,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2025-04-22T23:20:45Z
issue,147513,[RFC] Request for Feedback and Review on PRs Adding RISC-V and RVV Support,https://github.com/pytorch/pytorch/issues/147513,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cpu, module: risc-v",scan when time allows,2025-04-22T22:06:27Z
issue,145949,[CUDA][Blackwell] Blackwell Tracking Issue,https://github.com/pytorch/pytorch/issues/145949,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cuda",scan when time allows,2025-04-22T22:05:11Z
issue,141616,Clarify that `compile(module)` only affects the `forward` method.,https://github.com/pytorch/pytorch/issues/141616,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, oncall: pt2",scan when time allows,2025-04-22T15:32:06Z
issue,151113,Failure in test_vmap_autograd_grad_nn_functional_conv2d_cpu_float32,https://github.com/pytorch/pytorch/issues/151113,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: functorch, module: tests, module: vmap",scan when time allows,2025-04-22T12:07:44Z
issue,151561,2.6: vmap doesn't vectorize all args,https://github.com/pytorch/pytorch/issues/151561,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: functorch, module: vmap",scan when time allows,2025-04-22T03:15:11Z
issue,151555,[Inductor] Convolution auto-generation through Triton backend,https://github.com/pytorch/pytorch/issues/151555,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",scan when time allows,2025-04-21T22:28:48Z
issue,144206,Error when compiling document,https://github.com/pytorch/pytorch/issues/144206,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: docs",scan when time allows,2025-04-21T21:54:41Z
issue,150364,AOTI doesn't error if dynamic shape gets specialized during lowering,https://github.com/pytorch/pytorch/issues/150364,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: aotinductor, oncall: export, oncall: pt2",scan when time allows,2025-04-21T18:41:13Z
issue,124025,Different results for single sample vs. batched,https://github.com/pytorch/pytorch/issues/124025,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: collect_env.py, module: docs, module: numerical-stability",scan when time allows,2025-04-21T17:20:14Z
issue,137888,Inconsistent digamma result with other frameworks (1e-6 difference),https://github.com/pytorch/pytorch/issues/137888,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: special,scan when time allows,2025-04-21T08:01:38Z
issue,129140,[RFC][C10D] Avoid creating new nccl communicator for each P2P pair,https://github.com/pytorch/pytorch/issues/129140,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2025-04-21T02:58:38Z
issue,137080,Discrepancy between torch.nn.functional.linear and manual computation with bias under torch.autocast,https://github.com/pytorch/pytorch/issues/137080,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision), module: cuda, module: nn, module: numerical-stability",scan when time allows,2025-04-20T09:19:33Z
issue,37386,Add BufferDict container,https://github.com/pytorch/pytorch/issues/37386,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2025-04-19T08:50:14Z
issue,148564,[Feature Request] Dynamic shapes API requires spec for all arguments.,https://github.com/pytorch/pytorch/issues/148564,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamic shapes, oncall: pt2",scan when time allows,2025-04-18T22:59:04Z
issue,71300,"Kernel fusion for Gather, Apply, Scatter (GAS) model",https://github.com/pytorch/pytorch/issues/71300,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: scatter & gather ops, module: sparse",scan when time allows,2025-04-17T16:08:04Z
issue,150252,RuntimeError: Subtracting Reconstructed Jagged Nested Tensor Fails with Shape Mismatch,https://github.com/pytorch/pytorch/issues/150252,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: nestedtensor",scan when time allows,2025-04-17T14:25:19Z
issue,151463,Inconsistent Output from nn.Conv2d with padding_mode='circular' When Using MKLDNN on a AVX2_Processor,https://github.com/pytorch/pytorch/issues/151463,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: mkldnn",scan when time allows,2025-04-17T13:50:57Z
issue,151559,SDPA doesn't document requirement for head dimension in masks,https://github.com/pytorch/pytorch/issues/151559,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sdpa,scan when time allows,2025-04-17T13:47:59Z
issue,149837,"Intermittent ""AssertionError: can only test a child process"" Warning with PyTorch DataLoader on Colab",https://github.com/pytorch/pytorch/issues/149837,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dataloader, module: multiprocessing",scan when time allows,2025-04-17T09:34:28Z
issue,132395,`torch.multinomial` generates incorrect distribution,https://github.com/pytorch/pytorch/issues/132395,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: distributions, module: intel, module: numerical-stability",scan when time allows,2025-04-17T01:58:40Z
issue,151402,"RuntimeError: curPos <= (kUpperBound - kAppendInterval) INTERNAL ASSERT FAILED at ""../torch/csrc/jit/ir/ir.cpp"":698, please report a bug to PyTorch.",https://github.com/pytorch/pytorch/issues/151402,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2025-04-16T14:53:44Z
issue,151286,vmap- out of memory,https://github.com/pytorch/pytorch/issues/151286,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: vmap,scan when time allows,2025-04-16T14:49:03Z
issue,149110,[XPU] Enable Windows CI/CD test for XPU,https://github.com/pytorch/pytorch/issues/149110,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ci, module: xpu",scan when time allows,2025-04-15T20:59:27Z
issue,150571,Improve speed of pytorch docs build,https://github.com/pytorch/pytorch/issues/150571,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: docs",scan when time allows,2025-04-15T20:38:21Z
issue,149872,Do all lazy imports for torch.compile in one place?,https://github.com/pytorch/pytorch/issues/149872,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",scan when time allows,2025-04-15T17:25:26Z
issue,150961,[ued] `torch.compile` cause more than 2x slow down with HF diffuser GGUF Auraflow,https://github.com/pytorch/pytorch/issues/150961,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: pt2,scan when time allows,2025-04-14T22:51:33Z
issue,151015,update user defined triton kernel table to include strict vs non-strict difference,https://github.com/pytorch/pytorch/issues/151015,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamo, module: user triton, oncall: pt2",scan when time allows,2025-04-14T22:46:59Z
issue,150963,[ued] `torch.compile` yields lower latency when compiling transformer blocks only for ComfyUI GGUF Flux,https://github.com/pytorch/pytorch/issues/150963,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: pt2,scan when time allows,2025-04-14T22:46:04Z
issue,151251,Actually support LOAD_BUILD_CLASS,https://github.com/pytorch/pytorch/issues/151251,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamo, module: graph breaks, oncall: pt2",scan when time allows,2025-04-14T18:45:17Z
issue,146889,How to customize a torch.Tensor() method to access the underlying data structure of a PyTorch tensor.,https://github.com/pytorch/pytorch/issues/146889,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2025-04-14T17:40:25Z
issue,150928,ShardTensor gather will encounter an error when a local tensor on certain ranks has zero elements,https://github.com/pytorch/pytorch/issues/150928,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2025-04-14T17:26:47Z
issue,32137,torch.nn.functional.normalize epsilon too small for half precision,https://github.com/pytorch/pytorch/issues/32137,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: half, module: nn, module: norms and normalization, module: numerical-stability",scan when time allows,2025-04-14T16:57:30Z
issue,39178,Add support for rsample to MixtureSameFamily Distribution,https://github.com/pytorch/pytorch/issues/39178,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distributions,scan when time allows,2025-04-14T15:25:06Z
issue,149194,LSTM slow on PackedSequence,https://github.com/pytorch/pytorch/issues/149194,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rnn,scan when time allows,2025-04-14T10:42:35Z
issue,28440,Python/C++ API Parity: torch.optim optimizers,https://github.com/pytorch/pytorch/issues/28440,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp,scan when time allows,2025-04-14T10:33:37Z
issue,41667,[Feature Request] named tensor support for `tensordot`(tensor contract),https://github.com/pytorch/pytorch/issues/41667,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: named tensor,scan when time allows,2025-04-14T03:10:05Z
issue,149453,Support launch_enter and launch_exit hooks,https://github.com/pytorch/pytorch/issues/149453,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: pt2,scan when time allows,2025-04-11T22:50:58Z
issue,150124,torch.unique undocumented behaviour,https://github.com/pytorch/pytorch/issues/150124,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python frontend,scan when time allows,2025-04-11T18:37:00Z
issue,148232,Add Structured Knowledge Accumulation (SKA) Layer to PyTorch,https://github.com/pytorch/pytorch/issues/148232,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2025-04-11T15:05:06Z
issue,83510,Cdist backward dependent on compute_mode,https://github.com/pytorch/pytorch/issues/83510,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2025-04-10T18:23:40Z
issue,149440,[pt2] Support statically launching triton compiled cuda kernels,https://github.com/pytorch/pytorch/issues/149440,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, module: user triton, oncall: pt2",scan when time allows,2025-04-10T16:59:38Z
issue,68272,Query CUDA version LibTorch has been compiled with,https://github.com/pytorch/pytorch/issues/68272,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2025-04-10T12:27:06Z
issue,17835,PyPy support,https://github.com/pytorch/pytorch/issues/17835,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: binaries,scan when time allows,2025-04-10T03:36:52Z
issue,148743,"torch.unique(t, dim=1, return_counts=True) does not work",https://github.com/pytorch/pytorch/issues/148743,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python frontend,scan when time allows,2025-04-09T21:44:38Z
issue,148536,Computing only the n first rows of a distance matrix with pdist,https://github.com/pytorch/pytorch/issues/148536,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python frontend,scan when time allows,2025-04-09T20:46:09Z
issue,150922,StrictMinMaxConstraint issue in pytorch 2.4.0,https://github.com/pytorch/pytorch/issues/150922,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2025-04-09T20:07:11Z
issue,149304,Machine with isolated cores is not respected by torch,https://github.com/pytorch/pytorch/issues/149304,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: multiprocessing",scan when time allows,2025-04-09T19:26:49Z
issue,143041,Use labels to create dedicated pools of non-scalable runners for trunk jobs,https://github.com/pytorch/pytorch/issues/143041,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ci, module: infra",scan when time allows,2025-04-09T17:02:26Z
issue,132856,Enhance the compiler module to use cudagraph with PrivateUse1,https://github.com/pytorch/pytorch/issues/132856,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda graphs, module: privateuse1, oncall: pt2",scan when time allows,2025-04-09T16:57:19Z
issue,74236,"""histogram_cpu"" not implemented for 'Int'",https://github.com/pytorch/pytorch/issues/74236,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sorting and selection,scan when time allows,2025-04-09T16:47:30Z
issue,33811,"Models saved in C++ LibTorch with torch::save, cannot be loaded in python using torch.load",https://github.com/pytorch/pytorch/issues/33811,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: serialization",scan when time allows,2025-04-09T07:11:42Z
issue,150873,Add Created On | Last Updated On to the docs,https://github.com/pytorch/pytorch/issues/150873,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: docs",scan when time allows,2025-04-08T21:24:37Z
issue,150871,register_replacement should also respect arg_kwarg_vals,https://github.com/pytorch/pytorch/issues/150871,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",scan when time allows,2025-04-08T19:19:54Z
issue,150663,Proposing torch.empty_cache (device_type) as generalization of torch.cuda.empty_cache() .,https://github.com/pytorch/pytorch/issues/150663,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: accelerator,scan when time allows,2025-04-08T17:10:48Z
issue,150700,CuDNN + H100 Gives Weird Gradients,https://github.com/pytorch/pytorch/issues/150700,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: cudnn",scan when time allows,2025-04-08T16:44:04Z
issue,145978,What is the recommended way to use Distributed Checkpointing Save/Load with HSDP?,https://github.com/pytorch/pytorch/issues/145978,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: oncall: distributed, oncall: distributed checkpointing",scan when time allows,2025-04-08T15:58:03Z
issue,146307,Intermittent false positives in lintrunner's mypy,https://github.com/pytorch/pytorch/issues/146307,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: flaky-tests, module: lint",scan when time allows,2025-04-08T09:50:50Z
issue,150718,Codegen or Lint for python-api.md,https://github.com/pytorch/pytorch/issues/150718,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: docs",scan when time allows,2025-04-07T15:38:37Z
issue,36748,return_index option for torch.unique,https://github.com/pytorch/pytorch/issues/36748,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: numpy,scan when time allows,2025-04-07T12:54:34Z
issue,138696,[CI/CD] Deprecating PyTorch’s official Anaconda channel,https://github.com/pytorch/pytorch/issues/138696,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2025-04-05T18:44:01Z
issue,37406,3D grouped & depthwise convolution very slow on backward pass,https://github.com/pytorch/pytorch/issues/37406,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: cudnn",scan when time allows,2025-04-05T11:48:50Z
issue,119076,torch.bmm fails when multiplying a batch of sparse matrices by a dense tensor,https://github.com/pytorch/pytorch/issues/119076,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2025-04-04T23:12:21Z
issue,119482,`pylint`: `torch.nn.functional.linear` is not callable,https://github.com/pytorch/pytorch/issues/119482,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: typing",scan when time allows,2025-04-04T11:55:37Z
issue,149962,[ued][wan] Dynamic shape issue with tolist,https://github.com/pytorch/pytorch/issues/149962,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",scan when time allows,2025-04-03T18:07:19Z
issue,150614,conv2d fp8 support,https://github.com/pytorch/pytorch/issues/150614,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: floatx (formerly float8)",scan when time allows,2025-04-03T15:30:22Z
issue,128149,torch.compile with Custom tensor subclass doesn't inline the tensor subclass methods,https://github.com/pytorch/pytorch/issues/128149,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dynamo,scan when time allows,2025-04-03T00:10:25Z
issue,141299,Pytorch notifications,https://github.com/pytorch/pytorch/issues/141299,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: devx,scan when time allows,2025-04-02T23:42:57Z
issue,150548,Segmentation fault when using torch.tensor from a non python created thread,https://github.com/pytorch/pytorch/issues/150548,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: pybind",scan when time allows,2025-04-02T22:46:33Z
issue,150531,Intermittent SSL certificate expiry warnings for `download.pytorch.org` (load balancer?),https://github.com/pytorch/pytorch/issues/150531,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2025-04-02T22:27:28Z
issue,150368,Support `copy` kwarg in `torch.reshape()` following Python array API standard,https://github.com/pytorch/pytorch/issues/150368,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python array api,scan when time allows,2025-04-02T18:50:16Z
issue,150402,Feature Request: Add minLSTM and minGRU Modules,https://github.com/pytorch/pytorch/issues/150402,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: python frontend, module: rnn",scan when time allows,2025-04-02T18:48:44Z
issue,150284,torch.fill bug,https://github.com/pytorch/pytorch/issues/150284,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python frontend,scan when time allows,2025-04-02T18:44:56Z
issue,150257,Type hint bug report relating to Sequential,https://github.com/pytorch/pytorch/issues/150257,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: typing,scan when time allows,2025-04-02T18:44:49Z
issue,150285,CuSparse doesn't work on sparse tensor,https://github.com/pytorch/pytorch/issues/150285,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2025-04-02T18:44:30Z
issue,150078,[RFC] Remove periodic/unstable jobs that has been continuously broken for more than 30 days,https://github.com/pytorch/pytorch/issues/150078,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2025-04-02T15:47:17Z
issue,146707,AutoAC can cause deadlocks with tensor parallelism and data-dependent flop formulas,https://github.com/pytorch/pytorch/issues/146707,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: activation checkpointing, module: pt2-dispatcher, oncall: pt2",scan when time allows,2025-04-01T23:23:02Z
issue,149879,Inductor Pattern Matcher's register_replacement function only works with functional `search_fn`s,https://github.com/pytorch/pytorch/issues/149879,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",scan when time allows,2025-04-01T19:22:24Z
issue,61474,Implement missing torch.nan* operators,https://github.com/pytorch/pytorch/issues/61474,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nans and infs, module: numpy, module: reductions",scan when time allows,2025-04-01T17:46:04Z
issue,82430,Minor inconsistency in description of `attn_output_weights` in MultiheadAttention docs,https://github.com/pytorch/pytorch/issues/82430,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: nn",scan when time allows,2025-04-01T12:26:37Z
issue,68352,`Delta` not defined in the docs of `HingeEmbeddingLoss`,https://github.com/pytorch/pytorch/issues/68352,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: loss, module: nn",scan when time allows,2025-04-01T03:02:55Z
issue,146898,[RFC] Test Cases Enabling for Accelerators,https://github.com/pytorch/pytorch/issues/146898,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tests,scan when time allows,2025-04-01T03:00:16Z
issue,146693,Partitioner moves useless and memory-heavy op from bwd to fwd,https://github.com/pytorch/pytorch/issues/146693,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: aotdispatch, module: pt2-dispatcher, oncall: distributed, oncall: pt2",scan when time allows,2025-04-01T00:59:50Z
issue,149269,Add Optimizer: Sharpness Aware Minimization,https://github.com/pytorch/pytorch/issues/149269,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2025-03-31T20:07:53Z
issue,150242,Feature Request: Enhance Nested Tensor Operations for Direct RoPE Application,https://github.com/pytorch/pytorch/issues/150242,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nestedtensor,scan when time allows,2025-03-31T15:25:53Z
issue,71367,"F.cross_entropy do not have a deterministic implementation, adding deterministic support for this operation.",https://github.com/pytorch/pytorch/issues/71367,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: loss,scan when time allows,2025-03-30T09:01:31Z
issue,149957,[ued] Prohibitive warm start latency,https://github.com/pytorch/pytorch/issues/149957,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: compile-time, module: inductor, oncall: pt2",scan when time allows,2025-03-29T15:41:23Z
issue,149908,[ued][ChatTTS][guards] Too many recompilations,https://github.com/pytorch/pytorch/issues/149908,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",scan when time allows,2025-03-29T15:38:33Z
issue,149906,[ued][chatTTS][dynamo] graph break on should_compile_partial_graph=False,https://github.com/pytorch/pytorch/issues/149906,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",scan when time allows,2025-03-29T15:38:32Z
issue,149970,[ued][deepssek-vl2] Graph break on copy.deepcopy,https://github.com/pytorch/pytorch/issues/149970,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",scan when time allows,2025-03-29T15:35:24Z
issue,149956,[ued][flux][dynamo] Wrong error message with dynamo.disable,https://github.com/pytorch/pytorch/issues/149956,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",scan when time allows,2025-03-29T15:35:22Z
issue,149944,include cudagraph skip reasons in tlparse output,https://github.com/pytorch/pytorch/issues/149944,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda graphs, module: dynamo, oncall: pt2",scan when time allows,2025-03-29T15:35:22Z
issue,147062,[Feature Request] Release original parameters by layer when turning on `freezing_discard_parameters`,https://github.com/pytorch/pytorch/issues/147062,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",scan when time allows,2025-03-29T15:23:32Z
issue,146930,pytest test/dynamo fails,https://github.com/pytorch/pytorch/issues/146930,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",scan when time allows,2025-03-29T15:22:14Z
issue,79272,"PyTorch leaks a macro definition called ""CHECK"" in the C++ version",https://github.com/pytorch/pytorch/issues/79272,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp,scan when time allows,2025-03-29T15:16:24Z
issue,132695,Backward on nested jagged tensor vector multiplication throws error on 0 dim (but not last dim),https://github.com/pytorch/pytorch/issues/132695,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nestedtensor,scan when time allows,2025-03-28T18:50:14Z
issue,149609,Batching Implementation for aten_nested_tensor_from_mask_left_aligned,https://github.com/pytorch/pytorch/issues/149609,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: functorch, module: nestedtensor, module: vmap",scan when time allows,2025-03-28T18:43:13Z
issue,138901,Runtime error: retains_grad_hooks not implemented for compiled autograd,https://github.com/pytorch/pytorch/issues/138901,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: compiled autograd, oncall: pt2",scan when time allows,2025-03-28T08:41:57Z
issue,149743,dynamic annotations that still allow duck sizing,https://github.com/pytorch/pytorch/issues/149743,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamic shapes, oncall: pt2",scan when time allows,2025-03-28T06:49:02Z
issue,6564,[Feature Request] Optimization with constraint (L-BFGS-B),https://github.com/pytorch/pytorch/issues/6564,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2025-03-27T16:42:06Z
issue,146201,L-BFGS-B support,https://github.com/pytorch/pytorch/issues/146201,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2025-03-27T16:40:07Z
issue,104152,Error 101: invalid device ordinal (Triggered internally at ../c10/cuda/CUDAFunctions.cpp:109.),https://github.com/pytorch/pytorch/issues/104152,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: third_party",scan when time allows,2025-03-27T16:25:21Z
issue,106529,Pytorch + ROCm+ Windows,https://github.com/pytorch/pytorch/issues/106529,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rocm,scan when time allows,2025-03-27T15:20:28Z
issue,75773,Handle noncontiguous inputs in distributed backend layer,https://github.com/pytorch/pytorch/issues/75773,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: c10d, oncall: distributed",scan when time allows,2025-03-26T22:56:16Z
issue,149169,[RelEng] wheel testing for new arch versions,https://github.com/pytorch/pytorch/issues/149169,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ci, oncall: releng",scan when time allows,2025-03-26T22:26:19Z
issue,148984,[Sync file] the new file is not sync properly between pytorch/pytorch and pytorch/benchmark,https://github.com/pytorch/pytorch/issues/148984,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: releng,scan when time allows,2025-03-26T22:26:11Z
issue,68978,SequentialLR cannot be used with ReduceLROnPlateau due to .step() not allowing for optional arguments,https://github.com/pytorch/pytorch/issues/68978,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: lrscheduler, module: optimizer",scan when time allows,2025-03-26T19:38:07Z
issue,138226,No guards for frozen dataclasses and readonly configs,https://github.com/pytorch/pytorch/issues/138226,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamo, module: guards, oncall: pt2",scan when time allows,2025-03-26T17:45:15Z
issue,146647,placeholder for shell dtype improvements in pytorch/pytorch,https://github.com/pytorch/pytorch/issues/146647,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: internals,scan when time allows,2025-03-26T17:08:14Z
issue,149734,"[RFC] Multi-backend, multi-device test class instantiation for Inductor",https://github.com/pytorch/pytorch/issues/149734,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, module: testing, oncall: pt2",scan when time allows,2025-03-26T15:35:35Z
issue,139290,Using xpu module in the cuda version Pytorch,https://github.com/pytorch/pytorch/issues/139290,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cuda, module: xpu",scan when time allows,2025-03-26T07:52:03Z
issue,136381,[Meta] torch.export readiness for Recommendation models,https://github.com/pytorch/pytorch/issues/136381,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: oncall: export, oncall: pt2",scan when time allows,2025-03-26T00:43:22Z
issue,133458,[CPU] jx_nest_base AMP both inductor and eager performance regression in 2024-08-10 nightly release,https://github.com/pytorch/pytorch/issues/133458,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: oncall: cpu inductor, oncall: pt2",scan when time allows,2025-03-25T23:22:22Z
issue,148682,can we make inductor create faster fusions for tiled reductions across dim0 and dim1?,https://github.com/pytorch/pytorch/issues/148682,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",scan when time allows,2025-03-25T21:27:39Z
issue,137217,Automagically lift assertions / shape checks into torch._check,https://github.com/pytorch/pytorch/issues/137217,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamic shapes, oncall: pt2",scan when time allows,2025-03-25T18:36:00Z
issue,136119,[torch.export] Dim discrete/finite set,https://github.com/pytorch/pytorch/issues/136119,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamic shapes, oncall: export, oncall: pt2",scan when time allows,2025-03-25T17:48:58Z
issue,139268,Torch Inductor should have a way for new backend to provide build options,https://github.com/pytorch/pytorch/issues/139268,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",scan when time allows,2025-03-25T17:24:46Z
issue,139401,[export] Failure to export Mixtral-8x7B-Instruct-v0.1,https://github.com/pytorch/pytorch/issues/139401,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: oncall: export, oncall: pt2",scan when time allows,2025-03-25T17:15:32Z
issue,94083,Tracker for `scatter_reduce` additional reduction options requests,https://github.com/pytorch/pytorch/issues/94083,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: scatter & gather ops,scan when time allows,2025-03-25T12:34:02Z
issue,98861,Sparse Tensor not working for `torch.cat`,https://github.com/pytorch/pytorch/issues/98861,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2025-03-25T08:58:44Z
issue,149845,CUDA error: no kernel image is available for execution on the device RTX5090D,https://github.com/pytorch/pytorch/issues/149845,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cuda",scan when time allows,2025-03-25T02:58:38Z
issue,148045,Run Performance Regression Tests on new Version of Triton,https://github.com/pytorch/pytorch/issues/148045,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: performance,scan when time allows,2025-03-24T23:55:15Z
issue,149815,comparison operators only accept scalars as the 2nd argument but not as a 1st argument,https://github.com/pytorch/pytorch/issues/149815,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python frontend,scan when time allows,2025-03-24T23:30:06Z
issue,23859,CUDA: THTensor code complains about devices not matching when creating tensor from blob,https://github.com/pytorch/pytorch/issues/23859,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp,scan when time allows,2025-03-24T20:31:01Z
issue,71978,from_blob / make_tensor support MemoryFormat,https://github.com/pytorch/pytorch/issues/71978,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: memory format,scan when time allows,2025-03-24T20:29:28Z
issue,118421,Increase default bucket_cap_mb value to from 25MB to a more optimal value,https://github.com/pytorch/pytorch/issues/118421,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2025-03-24T19:37:20Z
issue,149804,torch.linalg.norm RuntimeError with torch.func.grad( vmap( hessian(.) ) ),https://github.com/pytorch/pytorch/issues/149804,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: functorch, module: linear algebra, module: vmap",scan when time allows,2025-03-24T18:52:44Z
issue,149803,checking out NCCL when it is not used,https://github.com/pytorch/pytorch/issues/149803,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2025-03-24T18:52:20Z
issue,149798,Build Extension Failed with setuptools==77.0.3,https://github.com/pytorch/pytorch/issues/149798,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp-extensions,scan when time allows,2025-03-24T18:48:11Z
issue,149841,`weight` parameter on functional not in nn.Module,https://github.com/pytorch/pytorch/issues/149841,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2025-03-24T18:43:46Z
issue,149788,`torch.vmap` does not work with tensor subclasses as expected,https://github.com/pytorch/pytorch/issues/149788,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: functorch,scan when time allows,2025-03-24T18:34:12Z
issue,149802,(With PR) `torch.cuda.MemPool()` internal assertion failure when changing devices,https://github.com/pytorch/pytorch/issues/149802,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2025-03-24T18:28:30Z
issue,149796,Cannot compile SGlang with Torch 2.7 or Torch 2.8 and CUDA 12.8 (sm_120).,https://github.com/pytorch/pytorch/issues/149796,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2025-03-24T14:27:59Z
issue,149603,"[inductor] API for user-controlled fusion ""escape-hatch"" mechanism in inductor",https://github.com/pytorch/pytorch/issues/149603,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",scan when time allows,2025-03-24T10:32:26Z
issue,135704,[feature request] Varlen indexing function for lookup and concat of varlen BPE tokens from a tensor vocab (i.e. `detokenize(...)` and arrays of strings). NestedTensor could become a representation for arrays-of-strings,https://github.com/pytorch/pytorch/issues/135704,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nestedtensor,scan when time allows,2025-03-23T12:29:20Z
issue,148858,RuntimeError: OffsetBasedRNGTracker instantiation requires the presence of CUDA/CUDA-like device,https://github.com/pytorch/pytorch/issues/148858,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2025-03-21T23:52:27Z
issue,149724,"`torch.linalg.ldl_factor_ex`, `torch.linalg.ldl_factor`, and `torch.linalg.lstsq` Raise INTERNAL ASSERT FAILED",https://github.com/pytorch/pytorch/issues/149724,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: error checking, module: linear algebra",scan when time allows,2025-03-21T15:32:19Z
issue,149718,Constraints for distributions with mixed support,https://github.com/pytorch/pytorch/issues/149718,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distributions,scan when time allows,2025-03-21T15:07:28Z
issue,149720,Implement `permute` for masked tensor,https://github.com/pytorch/pytorch/issues/149720,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nestedtensor,scan when time allows,2025-03-21T15:06:52Z
issue,21462,"Slow convolution with large kernels, should be using FFT",https://github.com/pytorch/pytorch/issues/21462,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: cudnn, module: fft, module: performance",scan when time allows,2025-03-21T00:25:30Z
issue,147475,[Export AOTI] dynamic_shapes export and compile degraded output,https://github.com/pytorch/pytorch/issues/147475,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: aotinductor, oncall: export, oncall: pt2",scan when time allows,2025-03-20T22:03:17Z
issue,149452,"Support save_cubin (and therefore, support cpp_wrapper use cases)",https://github.com/pytorch/pytorch/issues/149452,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: pt2,scan when time allows,2025-03-20T19:38:52Z
issue,149455,Support num_ctas > 1?,https://github.com/pytorch/pytorch/issues/149455,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: pt2,scan when time allows,2025-03-20T19:29:05Z
issue,149454,Support user defined triton kernels,https://github.com/pytorch/pytorch/issues/149454,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: pt2,scan when time allows,2025-03-20T19:28:56Z
issue,149620,`vmap` not working on `torch.range`,https://github.com/pytorch/pytorch/issues/149620,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: functorch, module: vmap",scan when time allows,2025-03-20T19:26:55Z
issue,149545,"`torch.load(..., map_location=""meta"")` hangs indefinitely",https://github.com/pytorch/pytorch/issues/149545,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: serialization,scan when time allows,2025-03-20T19:18:59Z
issue,148144,Torch export does not preserve original edges between nodes,https://github.com/pytorch/pytorch/issues/148144,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: export, oncall: pt2",scan when time allows,2025-03-20T16:43:10Z
issue,115305,torchrun to support NUMA affiinity,https://github.com/pytorch/pytorch/issues/115305,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: distributed_tool, oncall: distributed",scan when time allows,2025-03-20T15:39:32Z
issue,149184,"[macOS] instantiating optimizer after torch.set_default_device(""mps"") throws ""RuntimeError: Placeholder storage has not been allocated on MPS device!""",https://github.com/pytorch/pytorch/issues/149184,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: mps,scan when time allows,2025-03-19T20:30:19Z
issue,72759,Add softplus inverse,https://github.com/pytorch/pytorch/issues/72759,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: numerical-stability,scan when time allows,2025-03-19T09:48:35Z
issue,31772,Support Python builtins on iterators in JIT,https://github.com/pytorch/pytorch/issues/31772,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2025-03-18T19:25:12Z
issue,128150,More economical use of FX nodes when generating symbolic shapes related manipulations,https://github.com/pytorch/pytorch/issues/128150,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamic shapes, oncall: pt2",scan when time allows,2025-03-18T18:07:33Z
issue,36437,Issue when linking C++ code with libtorch_cpu: cuda not detected,https://github.com/pytorch/pytorch/issues/36437,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cpp",scan when time allows,2025-03-18T16:03:19Z
issue,148828,MiniCPM-o compilation issue,https://github.com/pytorch/pytorch/issues/148828,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",scan when time allows,2025-03-18T13:54:12Z
issue,124711,Output Discrepancy between PyTorch Model and Converted ONNX Model,https://github.com/pytorch/pytorch/issues/124711,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2025-03-18T08:18:19Z
issue,149313,"Training error: due to torch.load(weights, map_location='cpu') - Weights only load failed.",https://github.com/pytorch/pytorch/issues/149313,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: serialization,scan when time allows,2025-03-18T06:57:29Z
issue,149319,Need some sort of pattern matcher recipes somewhere,https://github.com/pytorch/pytorch/issues/149319,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",scan when time allows,2025-03-17T23:59:33Z
issue,149094,How to skip backward specific steps in torch.compile,https://github.com/pytorch/pytorch/issues/149094,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: pt2,scan when time allows,2025-03-17T23:55:31Z
issue,148920,Rewrite cpp extension tests to not be crazy,https://github.com/pytorch/pytorch/issues/148920,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp-extensions, module: tests",scan when time allows,2025-03-17T20:45:20Z
issue,149318,BypassFxGraphCache should have better error message for HOPs (also we should support more),https://github.com/pytorch/pytorch/issues/149318,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: pt2,scan when time allows,2025-03-17T20:33:05Z
issue,149324,Unguarded Usage of Facebook Internal Code?,https://github.com/pytorch/pytorch/issues/149324,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: third_party, oncall: pt2",scan when time allows,2025-03-17T20:29:04Z
issue,122485,DTensor doesn't support expand_as dispath mode,https://github.com/pytorch/pytorch/issues/122485,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dtensor,scan when time allows,2025-03-17T18:49:43Z
issue,149261,MPS Error: NDArray > 2^32 bytes in scaled_dot_product_attention,https://github.com/pytorch/pytorch/issues/149261,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: 64-bit, module: mps, module: sdpa",scan when time allows,2025-03-17T16:05:32Z
issue,149284,No examples in documentation for masked_fill and masked_fill_,https://github.com/pytorch/pytorch/issues/149284,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2025-03-17T15:33:18Z
issue,149209,Wrong argument in `CompileCounterWithBackend` when running `torch.utils.benchmark.utils.compile.bench_all`,https://github.com/pytorch/pytorch/issues/149209,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: benchmark,scan when time allows,2025-03-17T15:30:34Z
issue,149250,Implement batching for torch.isin operator,https://github.com/pytorch/pytorch/issues/149250,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: functorch, module: vmap",scan when time allows,2025-03-17T15:26:55Z
issue,149227,Unexpected behaviour in using torch.nn.utils.rnn.pack_padded_sequence API,https://github.com/pytorch/pytorch/issues/149227,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rnn,scan when time allows,2025-03-17T15:25:26Z
issue,149224,Unexpected results in using torch.save API,https://github.com/pytorch/pytorch/issues/149224,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: serialization,scan when time allows,2025-03-17T15:20:20Z
issue,78065,Gamma and Related Functions,https://github.com/pytorch/pytorch/issues/78065,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: special,scan when time allows,2025-03-15T20:01:14Z
issue,149013,Unexpected out-of-boundary behavior in `grid_sample`,https://github.com/pytorch/pytorch/issues/149013,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: edge cases, module: nn, module: numpy",scan when time allows,2025-03-15T09:53:35Z
issue,141014,ROCm: libamdhip64.so is linked twice when building a torch C++ program,https://github.com/pytorch/pytorch/issues/141014,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rocm,scan when time allows,2025-03-14T19:36:47Z
issue,149040,"Only 2D, 3D, 4D, 5D padding with non-constant padding are supported for now",https://github.com/pytorch/pytorch/issues/149040,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2025-03-14T19:01:28Z
issue,148976,`dist.barrier()` fails with TORCH_DISTRIBUTED_DEBUG=DETAIL and after dist.send/dist.recv calls,https://github.com/pytorch/pytorch/issues/148976,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2025-03-14T01:25:03Z
issue,92909,[RFC] XLA Lazy Backend Support In DistributedTensor API,https://github.com/pytorch/pytorch/issues/92909,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: xla,scan when time allows,2025-03-13T18:22:18Z
issue,99304,Need TransformerEncoder to output attention map,https://github.com/pytorch/pytorch/issues/99304,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2025-03-13T16:29:12Z
issue,149120,Add `keepdim` parameter for `torch.nn.functional.cosine_similarity`,https://github.com/pytorch/pytorch/issues/149120,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2025-03-13T15:24:10Z
issue,149090,Better warning once for `cuDNN/MIOpen` not enabled,https://github.com/pytorch/pytorch/issues/149090,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cudnn, module: tests",scan when time allows,2025-03-13T15:20:21Z
issue,148954,Aborted (core dumped) double free or corruption (out),https://github.com/pytorch/pytorch/issues/148954,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: linear algebra,scan when time allows,2025-03-13T15:13:06Z
issue,148885,test_memory_profiler_viz failed on cudamallocasync,https://github.com/pytorch/pytorch/issues/148885,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: testing",scan when time allows,2025-03-13T15:05:07Z
issue,125358,"[DeviceMesh] Add support for `group: Tuple[ProcessGroup, ...]` in `from_group()`",https://github.com/pytorch/pytorch/issues/125358,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: devicemesh,scan when time allows,2025-03-13T08:46:41Z
issue,149096,How to determine which part of torch.compile undergoes recompiling after caching,https://github.com/pytorch/pytorch/issues/149096,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: pt2,scan when time allows,2025-03-13T06:40:24Z
issue,148960,Does torch compile affect results ?,https://github.com/pytorch/pytorch/issues/148960,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",scan when time allows,2025-03-13T05:44:37Z
issue,37664,Remove everything from a GPU (including drivers),https://github.com/pytorch/pytorch/issues/37664,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2025-03-12T13:10:14Z
issue,147661,Add shape fn for einsum,https://github.com/pytorch/pytorch/issues/147661,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: lazy, module: linear algebra, module: xla",scan when time allows,2025-03-11T23:16:50Z
issue,107130,[docs] Document dtype conversions dtype.to_complex() dtype.to_real(),https://github.com/pytorch/pytorch/issues/107130,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2025-03-11T19:50:37Z
issue,130396,"Add some way to automatically opt-out of CI features, like TD.",https://github.com/pytorch/pytorch/issues/130396,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2025-03-11T19:09:02Z
issue,67978,c10::CUDAError,https://github.com/pytorch/pytorch/issues/67978,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: nccl, oncall: distributed",scan when time allows,2025-03-11T17:16:01Z
issue,68923,[cpp extension] making it possible to parallelize the building process,https://github.com/pytorch/pytorch/issues/68923,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: ci, module: cpp-extensions",scan when time allows,2025-03-11T07:13:52Z
issue,147628,"Optimize memory usage of torch.sort significantly, with dynamic dtype indices",https://github.com/pytorch/pytorch/issues/147628,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python frontend,scan when time allows,2025-03-11T02:49:29Z
issue,148927,default cudagraphable policy for custom op,https://github.com/pytorch/pytorch/issues/148927,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda graphs, oncall: pt2",scan when time allows,2025-03-10T23:57:20Z
issue,145158,Pytorch matmul for nested 4D tensors in jagged layout doesn't work,https://github.com/pytorch/pytorch/issues/145158,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nestedtensor,scan when time allows,2025-03-10T23:02:22Z
issue,148775,Calling `torch.linalg.lstsq` with a wrongly-shaped `out=output` argument does not properly reshape `output`,https://github.com/pytorch/pytorch/issues/148775,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: linear algebra,scan when time allows,2025-03-10T20:02:02Z
issue,81104,Resize/reshape of sparse compressed tensors - design,https://github.com/pytorch/pytorch/issues/81104,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2025-03-10T19:17:19Z
issue,148247,`torch.multinomial` outputs inconsistency on ARM and x86,https://github.com/pytorch/pytorch/issues/148247,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: distributions, module: random",scan when time allows,2025-03-10T02:38:31Z
issue,148546,Pytorch nightly broken Flash Attention 3 compile with sycl commit - TypeError: _write_ninja_file() got an unexpected keyword argument 'sycl_cflags',https://github.com/pytorch/pytorch/issues/148546,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2025-03-09T22:37:30Z
issue,137268,DDP deadlock ProcessGroupNCCL's watchdog got stuck,https://github.com/pytorch/pytorch/issues/137268,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: c10d, module: ddp, oncall: distributed",scan when time allows,2025-03-09T01:20:02Z
issue,72581,Zero-copy on shared memory of NVIDIA Jetson devices,https://github.com/pytorch/pytorch/issues/72581,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: jetson,scan when time allows,2025-03-07T22:03:39Z
issue,97539,InfoNCE loss for contrastive learning,https://github.com/pytorch/pytorch/issues/97539,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: loss,scan when time allows,2025-03-07T14:38:15Z
issue,148568,Suggested fixes sometimes not enough in export,https://github.com/pytorch/pytorch/issues/148568,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: oncall: export, oncall: pt2",scan when time allows,2025-03-07T01:43:49Z
issue,145511,Activation Checkpointing composability with split backward computation,https://github.com/pytorch/pytorch/issues/145511,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: activation checkpointing, module: pipelining, oncall: distributed",scan when time allows,2025-03-06T22:17:21Z
issue,148518,Preview (Nightly) version cuda12.8 cannot find torchaudio file,https://github.com/pytorch/pytorch/issues/148518,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cuda",scan when time allows,2025-03-06T01:51:38Z
issue,148548,Issue with Sparse Tensor Matrix Multiplication and Broadcasting,https://github.com/pytorch/pytorch/issues/148548,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2025-03-06T01:50:06Z
issue,148379,Improve nested jagged tensor select performance on batch dim,https://github.com/pytorch/pytorch/issues/148379,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nestedtensor, module: performance",scan when time allows,2025-03-06T00:47:29Z
issue,148044,Sweep through Potentially BC Breaking Commits in Triton,https://github.com/pytorch/pytorch/issues/148044,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2025-03-05T23:40:01Z
issue,148078,Video-Llama (version 1) runs much slower using Float16 than Float32 on Kunpeng CPU,https://github.com/pytorch/pytorch/issues/148078,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: arm, module: cpu",scan when time allows,2025-03-05T08:01:06Z
issue,148193,[Request Help] “torch._dynamo.exc.UserError: Dynamic control flow is not supported at the moment.” “torch._dynamo.exc.UncapturedHigherOrderOpError: Cond doesn't work unless it is captured completely with torch.compile.”,https://github.com/pytorch/pytorch/issues/148193,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: oncall: export, oncall: pt2",scan when time allows,2025-03-05T03:17:22Z
issue,148073,torch.compile with the inductor backend slows down (exponentially?) for certain graphs,https://github.com/pytorch/pytorch/issues/148073,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",scan when time allows,2025-03-04T19:36:04Z
issue,126795,Using DTensor with device meshes that use different devices for input and output,https://github.com/pytorch/pytorch/issues/126795,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dtensor, oncall: distributed",scan when time allows,2025-03-04T18:49:42Z
issue,148116,[inductor][triton] Explicit kernel-arg mismatch checks,https://github.com/pytorch/pytorch/issues/148116,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",scan when time allows,2025-03-04T17:33:35Z
issue,148146,Unify OpOverload._get_dispatch and HigherOrderOperator.dispatch,https://github.com/pytorch/pytorch/issues/148146,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: aotdispatch, oncall: pt2",scan when time allows,2025-03-04T16:03:01Z
issue,148196,"[inductor][triton] Decide how to deprecate ""old triton versions""",https://github.com/pytorch/pytorch/issues/148196,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",scan when time allows,2025-03-04T15:39:46Z
issue,148103,Initial investigation for removing MOD_SKIPLIST,https://github.com/pytorch/pytorch/issues/148103,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",scan when time allows,2025-03-04T15:05:25Z
issue,148052,[inductor][user triton] support on-device TMA / tensor descriptor API,https://github.com/pytorch/pytorch/issues/148052,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, module: user triton, oncall: pt2",scan when time allows,2025-03-04T02:41:35Z
issue,147999,Adjust test_mm_triton_kernel_benchmark for unpadded tensors,https://github.com/pytorch/pytorch/issues/147999,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",scan when time allows,2025-03-04T02:39:26Z
issue,138813,Strange recompilations on torch 2.5 + FSDP + UNet,https://github.com/pytorch/pytorch/issues/138813,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamic shapes, module: dynamo, module: fsdp, oncall: distributed",scan when time allows,2025-03-04T02:09:59Z
issue,136605,automatic_dynamic_shapes for mark_unbacked,https://github.com/pytorch/pytorch/issues/136605,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dynamic shapes,scan when time allows,2025-03-04T02:08:56Z
issue,143774,CUDA error when compiling loss function,https://github.com/pytorch/pytorch/issues/143774,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: activation checkpointing, oncall: pt2",scan when time allows,2025-03-03T19:29:30Z
issue,144470,Partitioner's auto-AC misbehaves with mixed dtypes,https://github.com/pytorch/pytorch/issues/144470,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: activation checkpointing, module: autograd, oncall: pt2",scan when time allows,2025-03-03T19:29:22Z
issue,119070,[DCP] Review and update examples and docstring to reflect DCP save/load API updates,https://github.com/pytorch/pytorch/issues/119070,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed checkpointing,scan when time allows,2025-03-03T19:06:02Z
issue,126047,`torch_save_to_dcp`'s produced dcp checkpoint doesn't fit the load interface for FSDP,https://github.com/pytorch/pytorch/issues/126047,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed checkpointing,scan when time allows,2025-03-03T19:05:47Z
issue,137327,[torch2.4][Distributed Checkpoint] new flatten logic is error-prone,https://github.com/pytorch/pytorch/issues/137327,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed checkpointing,scan when time allows,2025-03-03T19:05:06Z
issue,148098,onnx dynamo export does not support aten bucketize,https://github.com/pytorch/pytorch/issues/148098,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2025-03-03T17:39:09Z
issue,148113,"[inductor][triton] introduce better ""APIs"" in triton that can clean up our triton/inductor integration",https://github.com/pytorch/pytorch/issues/148113,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2025-03-03T17:06:32Z
issue,147874,Does CUDACachingAllocator.cpp still require deferred event creation?,https://github.com/pytorch/pytorch/issues/147874,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: cudacachingallocator",scan when time allows,2025-03-03T17:02:58Z
issue,147884,`FxGraphDrawer` fails on `einsum` nodes,https://github.com/pytorch/pytorch/issues/147884,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2025-03-03T17:00:15Z
issue,129942,Torch Compile mode giving Graph Break for Higher Order Op for ProcessGroup param,https://github.com/pytorch/pytorch/issues/129942,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: distributed, oncall: pt2",scan when time allows,2025-03-03T14:12:33Z
issue,148219,MPS vs Metal vs CPU performance comparison,https://github.com/pytorch/pytorch/issues/148219,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: mps, module: performance",scan when time allows,2025-03-01T00:49:52Z
issue,125958,Support AOT Autograd level Caching,https://github.com/pytorch/pytorch/issues/125958,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: compile-time, oncall: pt2",scan when time allows,2025-02-28T18:32:19Z
issue,102374,Hooks not working in version 2.0.1+cu118,https://github.com/pytorch/pytorch/issues/102374,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2025-02-28T08:28:51Z
issue,134191,"AdamW is CPU-bottlenecked with FSDP2, with slow foreach kernels",https://github.com/pytorch/pytorch/issues/134191,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: fsdp, module: optimizer",scan when time allows,2025-02-28T02:25:02Z
issue,143670,Potential rooms for fewer recompilations by introducing higher-level guards,https://github.com/pytorch/pytorch/issues/143670,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",scan when time allows,2025-02-27T21:59:13Z
issue,127225,make_graphed_callables don't work with FSDP at all even on a simple network,https://github.com/pytorch/pytorch/issues/127225,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, oncall: distributed",scan when time allows,2025-02-27T18:26:02Z
issue,140437,Let torch.compiler.allow_in_graph work in more situations,https://github.com/pytorch/pytorch/issues/140437,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",scan when time allows,2025-02-27T17:31:20Z
issue,147867,Track follow ups to #147354,https://github.com/pytorch/pytorch/issues/147867,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: internals,scan when time allows,2025-02-27T16:03:32Z
issue,64359,[feature request] `numpy.append` / `numpy.insert` / `numpy.delete` equivalents and implement dynamic arrays (reallocate storage with a surplus),https://github.com/pytorch/pytorch/issues/64359,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: numpy, module: viewing and reshaping",scan when time allows,2025-02-27T10:50:56Z
issue,8937,[caffe2] Is there any method to implement learning rate scheduler?,https://github.com/pytorch/pytorch/issues/8937,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2025-02-27T07:48:10Z
issue,148029,abs and arange Meta impls are non-standard and thus not working with other backends,https://github.com/pytorch/pytorch/issues/148029,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dispatch,scan when time allows,2025-02-27T00:28:54Z
issue,138786,Support nn.Module arguments for the function to torch.compiler.allow_in_graph,https://github.com/pytorch/pytorch/issues/138786,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",scan when time allows,2025-02-27T00:08:23Z
issue,141589,Minimizing recompiles when compiling multiple inlined `nn.Module`s,https://github.com/pytorch/pytorch/issues/141589,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamo, module: guards, oncall: pt2",scan when time allows,2025-02-26T22:41:43Z
issue,142395,[Dynamo] Do an audit on skipfiles and mark more files as inline,https://github.com/pytorch/pytorch/issues/142395,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",scan when time allows,2025-02-26T22:40:19Z
issue,133230,[dynamo] torch.compile does not support compiling a module that contains autograd within it.,https://github.com/pytorch/pytorch/issues/133230,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: compiled autograd, module: dynamo, oncall: pt2",scan when time allows,2025-02-26T22:32:30Z
issue,133117,"[dynamo] Improve support for subclasses of popular builtin types like dict, list, set",https://github.com/pytorch/pytorch/issues/133117,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",scan when time allows,2025-02-26T22:30:26Z
issue,147597,[Dynamo] WeakRefVariable doesn't use the most updated python referent when call_function is executed,https://github.com/pytorch/pytorch/issues/147597,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",scan when time allows,2025-02-26T21:24:42Z
issue,129228,CVE-2024-5480 reported by security analyzers,https://github.com/pytorch/pytorch/issues/129228,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2025-02-26T11:12:57Z
issue,147859,Exporting a PyTorch Model with Dynamic ModuleList Indexing to ONNX,https://github.com/pytorch/pytorch/issues/147859,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2025-02-26T02:41:28Z
issue,32651,[Tensorboard] Problem with subfolders from SummaryWriter,https://github.com/pytorch/pytorch/issues/32651,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tensorboard,scan when time allows,2025-02-25T17:37:13Z
issue,147682,Error in parsing arch_string in _extract_arch_version,https://github.com/pytorch/pytorch/issues/147682,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2025-02-25T15:29:10Z
issue,147692,Please generate a new version of pytorch into blackwell rtx5080 with cuda 12.8. Thank you very much.,https://github.com/pytorch/pytorch/issues/147692,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: binaries,scan when time allows,2025-02-25T15:24:41Z
issue,128015,Best approach for dynamically changing batch sizes on iterable datasets,https://github.com/pytorch/pytorch/issues/128015,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: data, module: dataloader",scan when time allows,2025-02-25T02:50:38Z
issue,129937,[tracker] Dynamo guard validation exposed issues,https://github.com/pytorch/pytorch/issues/129937,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",scan when time allows,2025-02-25T00:23:56Z
issue,130119,Compilation time for packing function is quite long,https://github.com/pytorch/pytorch/issues/130119,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: compile-time, module: dynamo, oncall: pt2",scan when time allows,2025-02-25T00:11:58Z
issue,130534,Clarify invariants established by torch.fx.passes.runtime_asserts or factor so we can have a pre-pass that reestablishes them,https://github.com/pytorch/pytorch/issues/130534,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamic shapes, oncall: pt2",scan when time allows,2025-02-24T22:45:58Z
issue,13811,Feature request: von Mises-Fisher distribution,https://github.com/pytorch/pytorch/issues/13811,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distributions,scan when time allows,2025-02-24T22:44:09Z
issue,130769,[Inductor] Add an API to register external callable candidates for inductor's Matmul/Conv tuning choices,https://github.com/pytorch/pytorch/issues/130769,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",scan when time allows,2025-02-24T22:32:09Z
issue,128095,[inductor] Graph breaks in CohereForAI/aya-23-8b,https://github.com/pytorch/pytorch/issues/128095,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamo, module: graph breaks, oncall: pt2",scan when time allows,2025-02-24T21:51:31Z
issue,128153,"torch.compile Jamba: long compilation time with backend=""eager""",https://github.com/pytorch/pytorch/issues/128153,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: compile-time, module: dynamo, module: guards, oncall: pt2",scan when time allows,2025-02-24T18:30:17Z
issue,129482,Dynamo graph breaks on boxed ProcessGroup,https://github.com/pytorch/pytorch/issues/129482,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: distributed, oncall: pt2",scan when time allows,2025-02-24T17:44:41Z
issue,147595,Non-Determinism in Faster R-CNN Despite Setting All Deterministic Flags,https://github.com/pytorch/pytorch/issues/147595,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: determinism,scan when time allows,2025-02-24T16:05:22Z
issue,147389,Fix representation of `Lazy*` modules after loading parameters,https://github.com/pytorch/pytorch/issues/147389,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2025-02-24T16:03:32Z
issue,147585,[SDPA backward error]Error detected in ScaledDotProductEfficientAttentionBackward0 when input seqlen is very long and with attn_mask,https://github.com/pytorch/pytorch/issues/147585,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sdpa,scan when time allows,2025-02-24T16:01:29Z
issue,147495,Native Node Sleep/Wake Functionality for Neural Network Layers,https://github.com/pytorch/pytorch/issues/147495,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2025-02-24T15:53:59Z
issue,147506,Unintuitive behavior and errors in F.pad,https://github.com/pytorch/pytorch/issues/147506,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: padding",scan when time allows,2025-02-24T15:52:09Z
issue,61292,torch.linspace tensor support,https://github.com/pytorch/pytorch/issues/61292,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: numpy,scan when time allows,2025-02-24T10:59:10Z
issue,130067,Meta device not supported with autocast,https://github.com/pytorch/pytorch/issues/130067,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision),scan when time allows,2025-02-24T10:58:22Z
issue,147288,Calling optimizer.load_state_dict causes Tensors lost,https://github.com/pytorch/pytorch/issues/147288,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2025-02-23T18:15:57Z
issue,146157,AdamW refactoring broke checkpoint reloading with DCP,https://github.com/pytorch/pytorch/issues/146157,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: optimizer, oncall: distributed checkpointing",scan when time allows,2025-02-22T05:02:04Z
issue,147284,torch.softmax gives wierd result under specific condition on cuda double tensors with multiple rows and 781 columns.,https://github.com/pytorch/pytorch/issues/147284,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: nn",scan when time allows,2025-02-21T23:02:47Z
issue,49844,BCEWithLogitsLoss gives out nan with -inf logits,https://github.com/pytorch/pytorch/issues/49844,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: loss, module: nans and infs",scan when time allows,2025-02-21T20:32:28Z
issue,147287,[inductor][fuzzer] legacy `torch.native_batch_norm` should be removed,https://github.com/pytorch/pytorch/issues/147287,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: decompositions, oncall: pt2",scan when time allows,2025-02-21T13:43:44Z
issue,147581,cpp libtorch transformerimpl lack some parameter between with python pytorch,https://github.com/pytorch/pytorch/issues/147581,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: nn",scan when time allows,2025-02-21T06:58:05Z
issue,59692,"cuda streams run sequentially, expected to run parallel",https://github.com/pytorch/pytorch/issues/59692,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: performance",scan when time allows,2025-02-21T05:25:52Z
issue,147510,`clamp_` and `clamp` behave differently on MPS device.,https://github.com/pytorch/pytorch/issues/147510,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: mps",scan when time allows,2025-02-21T03:33:22Z
issue,38412,No MKL Compatible Conda installation for PyTorch 1.5,https://github.com/pytorch/pytorch/issues/38412,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: mkl",scan when time allows,2025-02-21T03:21:40Z
issue,141686,[PT2] Fine-grained custom triton kernels support,https://github.com/pytorch/pytorch/issues/141686,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: user triton, oncall: pt2",scan when time allows,2025-02-20T23:11:06Z
issue,137514,Support multiple ragged dims for NJT,https://github.com/pytorch/pytorch/issues/137514,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nestedtensor,scan when time allows,2025-02-20T21:15:26Z
issue,37443,torch.clamp_max clamp_min shouldn't be there,https://github.com/pytorch/pytorch/issues/37443,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2025-02-20T17:01:42Z
issue,133586,No way for low-overhead total norm in native PyTorch with large number of tensors,https://github.com/pytorch/pytorch/issues/133586,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: mta, module: nn, oncall: pt2",scan when time allows,2025-02-20T02:55:32Z
issue,131667,[RFE] Improve the error reporting if MIG is incorrectly used with NVML,https://github.com/pytorch/pytorch/issues/131667,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2025-02-19T20:40:33Z
issue,52134,Build using Py_LIMITED_API and then build wheels with the stable ABI abi3 tag,https://github.com/pytorch/pytorch/issues/52134,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: performance, oncall: releng",scan when time allows,2025-02-19T19:16:05Z
issue,146910,"INTERNAL ASSERT FAILED at ""/pytorch/aten/src/ATen/native/sparse/SparseCsrTensorMath.cpp"":817, please report a bug to PyTorch.",https://github.com/pytorch/pytorch/issues/146910,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2025-02-19T13:56:12Z
issue,143112,FFT half precision only let CUDA pass the check,https://github.com/pytorch/pytorch/issues/143112,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: fft, module: privateuse1",scan when time allows,2025-02-19T09:50:58Z
issue,145214,Nested tensor support for pointwise matrix multiplication of nested tensor and normal tensor,https://github.com/pytorch/pytorch/issues/145214,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nestedtensor,scan when time allows,2025-02-19T09:30:15Z
issue,121668,Warning static library kineto_LIBRARY-NOTFOUND not found when trying build tutorial torch.compiler_aot_inductor.html,https://github.com/pytorch/pytorch/issues/121668,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: cpp, oncall: releng",scan when time allows,2025-02-19T04:50:58Z
issue,121485,libtorch memory leakage when inference by Multi-thread,https://github.com/pytorch/pytorch/issues/121485,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: memory usage",scan when time allows,2025-02-19T04:39:37Z
issue,146629,[inductor][idea] Defer realize/inline decisions,https://github.com/pytorch/pytorch/issues/146629,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",scan when time allows,2025-02-18T21:19:48Z
issue,147239,`DeviceCopy in input program` source hint,https://github.com/pytorch/pytorch/issues/147239,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: pt2,scan when time allows,2025-02-18T18:39:46Z
issue,147264,`Dim.DYNAMIC` inferred to be constant,https://github.com/pytorch/pytorch/issues/147264,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamic shapes, oncall: pt2",scan when time allows,2025-02-18T17:48:48Z
issue,147387,Track test count regressions,https://github.com/pytorch/pytorch/issues/147387,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: devx,scan when time allows,2025-02-18T17:46:13Z
issue,143969,[distributed] parallelize_module error with `SequenceParallel`,https://github.com/pytorch/pytorch/issues/143969,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2025-02-17T19:26:18Z
issue,81084,[ONNX] Exporting the operator `::svd` to ONNX opset version 13 is not supported.,https://github.com/pytorch/pytorch/issues/81084,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2025-02-16T13:14:45Z
issue,146708,Periodic Activations Module,https://github.com/pytorch/pytorch/issues/146708,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2025-02-14T20:06:26Z
issue,101335,fsdp training with the seq2seqTranier module gets stuck during evaluation.,https://github.com/pytorch/pytorch/issues/101335,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fsdp,scan when time allows,2025-02-14T18:42:23Z
issue,107909,Provide a `reset_parameters()` method for MultiheadAttention to support FSDP meta device initializtion,https://github.com/pytorch/pytorch/issues/107909,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: fsdp, module: nn",scan when time allows,2025-02-14T18:40:04Z
issue,109002,[FSDP] How can I wrap a model that has both nn.Parameter and nn.Module,https://github.com/pytorch/pytorch/issues/109002,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: fsdp, oncall: distributed",scan when time allows,2025-02-14T18:39:38Z
issue,63911,Add efficient symmetric tensor representations,https://github.com/pytorch/pytorch/issues/63911,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: __torch_function__, module: linear algebra",scan when time allows,2025-02-14T17:10:50Z
issue,139210,[Runtime Error] Build PyTorch with cuda12.2 on Jetson AGX Orin with jetpack 5.1.4,https://github.com/pytorch/pytorch/issues/139210,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: arm, module: cuda, module: jetson",scan when time allows,2025-02-14T04:28:52Z
issue,147118,padding fails on view from large tensor,https://github.com/pytorch/pytorch/issues/147118,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: 64-bit, module: cuda, module: edge cases, module: padding",scan when time allows,2025-02-13T18:16:46Z
issue,113560,t.contiguous() ~10 slower in eager mode compared to torch.compile,https://github.com/pytorch/pytorch/issues/113560,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: performance",scan when time allows,2025-02-13T17:59:21Z
issue,146848,_scale_dot_product_efficient_attention eager/meta computation mismatch on output tensor shape,https://github.com/pytorch/pytorch/issues/146848,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: sdpa, oncall: pt2",scan when time allows,2025-02-13T12:34:23Z
issue,132136,`torch.nn.transformer.forward` returns incorrect value inside `torch.no_grad()` blocks.,https://github.com/pytorch/pytorch/issues/132136,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2025-02-13T11:14:43Z
issue,146976,Adding sparse_coo_tensor objects together behaves differently with CPU and CUDA,https://github.com/pytorch/pytorch/issues/146976,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2025-02-13T09:00:09Z
issue,146964,Feature Request: Interface to Check Compilation Status Inside Compiled Module,https://github.com/pytorch/pytorch/issues/146964,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",scan when time allows,2025-02-13T08:58:59Z
issue,146119,File based multi gpu,https://github.com/pytorch/pytorch/issues/146119,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2025-02-13T07:14:37Z
issue,116393,Error: IndexError: map::at When using torch.distributed.all_reduce(tensor),https://github.com/pytorch/pytorch/issues/116393,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2025-02-13T02:04:22Z
issue,146944,win-vs2022-cpu-py3 test failures in test-default-2-3-lf.windows.4xlarge.nonephemeral_37054642004,https://github.com/pytorch/pytorch/issues/146944,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: windows,scan when time allows,2025-02-12T22:19:09Z
issue,146901,Trying to use forward AD with _scaled_dot_product_efficient_attention,https://github.com/pytorch/pytorch/issues/146901,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: forward ad, module: sdpa",scan when time allows,2025-02-12T22:14:49Z
issue,91439,Pytorch2.0 doesn't support compiling GRU and RNN model,https://github.com/pytorch/pytorch/issues/91439,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",scan when time allows,2025-02-12T20:03:34Z
issue,143571,Issues linking to libtorch on M2 mac,https://github.com/pytorch/pytorch/issues/143571,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: arm,scan when time allows,2025-02-12T17:15:21Z
issue,146982,Porting Pytorch to AIX Operating System.,https://github.com/pytorch/pytorch/issues/146982,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: power",scan when time allows,2025-02-12T15:03:33Z
issue,146108,[associative_scan] Support lifted arguments for inductor,https://github.com/pytorch/pytorch/issues/146108,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: pt2,scan when time allows,2025-02-12T10:07:04Z
issue,86274,TransformerEncoder/TransformerDecoder has same initial parameters for all layers,https://github.com/pytorch/pytorch/issues/86274,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2025-02-12T01:12:02Z
issue,146885,Add numpy.core.multiarray.scalar to default safe globals,https://github.com/pytorch/pytorch/issues/146885,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: serialization,scan when time allows,2025-02-11T13:15:59Z
issue,144012,[RFC] Add CPP Grouped GEMM Template for Inductor CPU,https://github.com/pytorch/pytorch/issues/144012,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",scan when time allows,2025-02-11T10:27:22Z
issue,146765,"Automatically resolve tensor mismatch issues, tensor conversion, and moving tensors to devices",https://github.com/pytorch/pytorch/issues/146765,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python frontend,scan when time allows,2025-02-11T09:33:38Z
issue,146831,Missing Typing for Loss Functions Causes Poor Code Completion in VSCode,https://github.com/pytorch/pytorch/issues/146831,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: typing",scan when time allows,2025-02-11T09:14:41Z
issue,78064,nn.CosineSimilarity returns value larger than 1,https://github.com/pytorch/pytorch/issues/78064,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: nn",scan when time allows,2025-02-10T23:54:06Z
issue,121595,Different Dropout behavior on macOS and Linux,https://github.com/pytorch/pytorch/issues/121595,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: numerical-reproducibility, module: random",scan when time allows,2025-02-10T16:23:05Z
issue,146570,inconsistency in geometric_ on CPU and GPU,https://github.com/pytorch/pytorch/issues/146570,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: edge cases,scan when time allows,2025-02-10T14:30:28Z
issue,145710,LBFGS-B with cuda implementation (and CPU one too),https://github.com/pytorch/pytorch/issues/145710,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2025-02-09T13:20:55Z
issue,146699,`#type: ignore` linter,https://github.com/pytorch/pytorch/issues/146699,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: lint, module: typing",scan when time allows,2025-02-08T13:43:48Z
issue,146711,Add different padding modes to torch.nn.utils.rnn.pad_sequence,https://github.com/pytorch/pytorch/issues/146711,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: rnn",scan when time allows,2025-02-08T01:31:43Z
issue,146697,There is a performance drop because we have not yet implemented the batching rule for aten::mse_loss_backward.,https://github.com/pytorch/pytorch/issues/146697,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: batching, module: functorch, module: vmap",scan when time allows,2025-02-08T01:31:07Z
issue,146732,increase lwork/rwork sizes for all float->int conversions,https://github.com/pytorch/pytorch/issues/146732,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: linear algebra,scan when time allows,2025-02-07T23:31:33Z
issue,146328,DeepSeek: Grouped/Ragged gemm with offsets on the GPU,https://github.com/pytorch/pytorch/issues/146328,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: linear algebra, module: nestedtensor",scan when time allows,2025-02-07T22:23:01Z
issue,130948,[inductor][cpu]opacus_cifar10 performance regression in 2024-07-15 nightly release,https://github.com/pytorch/pytorch/issues/130948,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: oncall: cpu inductor, oncall: pt2",scan when time allows,2025-02-07T21:04:35Z
issue,126211,Conflicting OMP implementations linked when building with MKLDNN,https://github.com/pytorch/pytorch/issues/126211,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: mkldnn, module: openmp",scan when time allows,2025-02-07T20:57:00Z
issue,104702,"[feature request] Specialized memory layouts and wide blocked/tiled dtypes for cublasLt (`CUBLASLT_ORDER_COL32`, `CUBLASLT_ORDER_COL4_4R2_8C`, `CUBLASLT_ORDER_COL32_2R_4R4`)/onednn: e.g. torch.float16x32 / torch.int8x32 / torch.bits1x512 (akin to torch.quint2x4)",https://github.com/pytorch/pytorch/issues/104702,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: memory format,scan when time allows,2025-02-07T18:42:39Z
issue,104591,version libcudnn_ops_infer.so.8 not defined in file libcudnn_ops_infer.so.8 with link time reference,https://github.com/pytorch/pytorch/issues/104591,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2025-02-07T16:49:18Z
issue,146701,Unable to Import PyTorch After Upgrade in Docker Environment,https://github.com/pytorch/pytorch/issues/146701,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docker,scan when time allows,2025-02-07T15:40:48Z
issue,146086,Enhancements to dim_order API for Ambiguity Detection,https://github.com/pytorch/pytorch/issues/146086,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: memory format, module: python frontend",scan when time allows,2025-02-07T09:49:15Z
issue,144955,ModuleNotFoundError: No module named 'torch.privateuseone',https://github.com/pytorch/pytorch/issues/144955,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: privateuse1",scan when time allows,2025-02-07T06:18:40Z
issue,146118,[NCCL] Segfault in topology path selection when calculating max bandwidth on nightly,https://github.com/pytorch/pytorch/issues/146118,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nccl, oncall: distributed",scan when time allows,2025-02-07T03:11:21Z
issue,32570,[feature request] Out-variant and dtype argument for torch.argmax / torch.argmin / torch.argsort (and friends),https://github.com/pytorch/pytorch/issues/32570,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: memory usage, module: reductions, module: sorting and selection",scan when time allows,2025-02-07T01:12:23Z
issue,146021,multi node Error when dist.destroy_process_group,https://github.com/pytorch/pytorch/issues/146021,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nccl, oncall: distributed",scan when time allows,2025-02-05T09:02:11Z
issue,146096,`torch.nn.functional.conv2d` 8 times slower in torch 2.5.1 compared to 2.3.1,https://github.com/pytorch/pytorch/issues/146096,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cudnn,scan when time allows,2025-02-04T23:14:28Z
issue,145388,create DISABLED issues for specific runner labels,https://github.com/pytorch/pytorch/issues/145388,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2025-02-04T21:41:25Z
issue,25162,Incorrect Validation Accuracy Due to Distributed Sampler,https://github.com/pytorch/pytorch/issues/25162,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dataloader, oncall: distributed",scan when time allows,2025-02-04T20:08:03Z
issue,145356,[autograd] inconsistent jvp results,https://github.com/pytorch/pytorch/issues/145356,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: functorch",scan when time allows,2025-02-04T19:00:46Z
issue,145947,flops counter error with PyTorch2.5 and 2.6,https://github.com/pytorch/pytorch/issues/145947,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: flop counter, oncall: distributed",scan when time allows,2025-02-04T08:02:05Z
issue,115339,Add support for measuring elapsed time in cuda-graphs,https://github.com/pytorch/pytorch/issues/115339,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda graphs,scan when time allows,2025-02-04T00:19:02Z
issue,142145,Fix disabled 3.13 cpp_extension test failing due to registration error,https://github.com/pytorch/pytorch/issues/142145,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp-extensions, module: privateuse1",scan when time allows,2025-02-03T20:16:03Z
issue,145592,"_pickle.UnpicklingError: invalid load key, ''.",https://github.com/pytorch/pytorch/issues/145592,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2025-02-03T06:55:32Z
issue,130359,FP8 Rowwise Scaling does not work on L4 GPU,https://github.com/pytorch/pytorch/issues/130359,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: floatx (formerly float8),scan when time allows,2025-02-03T03:03:09Z
issue,146034,This lib does not have new types of loss functions like focal Loss,https://github.com/pytorch/pytorch/issues/146034,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2025-02-02T18:57:45Z
issue,143864,Adaptive pool MPS,https://github.com/pytorch/pytorch/issues/143864,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: mps, module: pooling",scan when time allows,2025-02-01T11:13:39Z
issue,145851,FIRE Relative Positional Encodings,https://github.com/pytorch/pytorch/issues/145851,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2025-01-31T21:01:13Z
issue,77939,Fails to compile with GCC 12.1.0,https://github.com/pytorch/pytorch/issues/77939,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2025-01-31T08:46:35Z
issue,146040,Poor performance of any() on tensors even worse on cuda vs cpu,https://github.com/pytorch/pytorch/issues/146040,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: performance",scan when time allows,2025-01-30T22:50:37Z
issue,146056,RuntimeError: CUDA error: the provided PTX was compiled with an unsupported toolchain. Compile with `TORCH_USE_CUDA_DSA` to enable device-side assertions.,https://github.com/pytorch/pytorch/issues/146056,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2025-01-30T22:50:00Z
issue,146036,"torch/autograd/graph.py"" torch.func.{grad, vjp, jacrev, hessian} don't yet support saved tensor hooks",https://github.com/pytorch/pytorch/issues/146036,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: functorch",scan when time allows,2025-01-30T22:49:24Z
issue,145665,Pickling duplicates Storage,https://github.com/pytorch/pytorch/issues/145665,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: pickle,scan when time allows,2025-01-30T19:44:43Z
issue,92912,functorch.functionalize doesn't error out with logcumsumexp.out,https://github.com/pytorch/pytorch/issues/92912,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: functionalization, module: meta tensors, module: xla",scan when time allows,2025-01-29T20:45:02Z
issue,145944,Add align_to_window option to onnx version of torch.stft,https://github.com/pytorch/pytorch/issues/145944,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: fft, module: onnx",scan when time allows,2025-01-29T20:16:36Z
issue,145943,Add center and padding options to onnx version of torch.stft,https://github.com/pytorch/pytorch/issues/145943,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: fft, module: onnx",scan when time allows,2025-01-29T20:16:25Z
issue,134236,[CI] Binary Test Workflows are failing when updating triton hash,https://github.com/pytorch/pytorch/issues/134236,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, oncall: releng",scan when time allows,2025-01-29T17:30:45Z
issue,145907,There may be a documentation Error in torch.nn.CrossEntropyLoss Formula,https://github.com/pytorch/pytorch/issues/145907,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: loss",scan when time allows,2025-01-29T16:04:35Z
issue,75184,"PyTorch source code compile fail after ""Built target fbgemm_avx2""",https://github.com/pytorch/pytorch/issues/75184,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2025-01-29T14:32:44Z
issue,145729,KL div between t-distribution,https://github.com/pytorch/pytorch/issues/145729,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distributions,scan when time allows,2025-01-28T23:44:56Z
issue,145624,Local config flags for torch.compile,https://github.com/pytorch/pytorch/issues/145624,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: pt2,scan when time allows,2025-01-28T19:28:01Z
issue,145233,Not using set_num_threads results in very slow .all(),https://github.com/pytorch/pytorch/issues/145233,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: performance",scan when time allows,2025-01-27T18:49:00Z
issue,145716,support arm architectures in your Docker images on docker hub? (,https://github.com/pytorch/pytorch/issues/145716,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: arm, module: docker",scan when time allows,2025-01-27T15:45:37Z
issue,104113,Documentation building fails due to torchgen,https://github.com/pytorch/pytorch/issues/104113,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: docs",scan when time allows,2025-01-27T05:06:20Z
issue,72065,Batched sparse-sparse matrix multiplication/ sparse torch.einsum,https://github.com/pytorch/pytorch/issues/72065,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2025-01-25T22:13:22Z
issue,107294,"'MPS' training Issue(s) with NanoGPT: -Inf, NaN's",https://github.com/pytorch/pytorch/issues/107294,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: intel, module: mps",scan when time allows,2025-01-24T22:29:26Z
issue,91368,"PyTorch memory leak reference cycle in for loop, Mac M1",https://github.com/pytorch/pytorch/issues/91368,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: arm, module: memory usage, module: mps",scan when time allows,2025-01-24T22:11:02Z
issue,119677,Unexpected Gradient Results From conv1d on MacOS,https://github.com/pytorch/pytorch/issues/119677,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: correctness (silent), module: functorch, module: mps",scan when time allows,2025-01-24T21:58:06Z
issue,145493,Add a lint rule to avoid the word `interface` in C++,https://github.com/pytorch/pytorch/issues/145493,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: lint",scan when time allows,2025-01-24T20:12:08Z
issue,145453,[CUDA] Illegal Memory Access with `AdaptiveMaxPool2d`,https://github.com/pytorch/pytorch/issues/145453,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: edge cases, module: nn",scan when time allows,2025-01-24T20:10:08Z
issue,143955,Issue with aten::_sparse_coo_tensor_with_dims_and_tensors on Apple Silicon GPU (MPS) Backend when Using Whisper Model,https://github.com/pytorch/pytorch/issues/143955,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: mps, module: sparse",scan when time allows,2025-01-24T18:41:32Z
issue,145468,[inductor][torchbench] Unsupported operator issue when running the torch_multimodal_clip model with batch size 4.,https://github.com/pytorch/pytorch/issues/145468,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: sdpa, oncall: pt2",scan when time allows,2025-01-24T15:48:00Z
issue,145368,The possible error in the pytorch documentation of RNN.,https://github.com/pytorch/pytorch/issues/145368,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rnn,scan when time allows,2025-01-24T03:27:21Z
issue,51151,-fno-omit-frame-pointer by default in our builds,https://github.com/pytorch/pytorch/issues/51151,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2025-01-24T03:05:20Z
issue,145262,Missing create_graph arguments in torch.func apis,https://github.com/pytorch/pytorch/issues/145262,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: functorch,scan when time allows,2025-01-24T01:32:09Z
issue,144223,Support LayerNorm2d,https://github.com/pytorch/pytorch/issues/144223,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2025-01-24T01:22:24Z
issue,80827,Inconsistent computation of gradient in MaxUnPooling,https://github.com/pytorch/pytorch/issues/80827,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: correctness (silent), module: determinism",scan when time allows,2025-01-23T07:45:00Z
issue,145357,FP8: E5M2: The FP8 E5M2 result is not `inf` when casting a FP32 value larger than max normal value of FP8 E5M2 (57344),https://github.com/pytorch/pytorch/issues/145357,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: floatx (formerly float8), module: nans and infs",scan when time allows,2025-01-23T02:20:15Z
issue,143215,"Runners, torchbench, & the future",https://github.com/pytorch/pytorch/issues/143215,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2025-01-23T02:17:12Z
issue,141079,[triton 3.2] test_convolution_as_mm failure on A100,https://github.com/pytorch/pytorch/issues/141079,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2025-01-22T21:24:04Z
issue,145350,[CUDA] Illegal Memory Access with `ReplicationPad2D`,https://github.com/pytorch/pytorch/issues/145350,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: edge cases, module: nn",scan when time allows,2025-01-22T18:13:41Z
issue,145349,[CUDA] Illegal Memory Access with `AdaptiveAvgPool2d`,https://github.com/pytorch/pytorch/issues/145349,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: edge cases, module: nn",scan when time allows,2025-01-22T18:13:26Z
issue,145326,[inductor][triton] refactor ASTSource.make_ir integration,https://github.com/pytorch/pytorch/issues/145326,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, module: user triton, oncall: pt2",scan when time allows,2025-01-22T18:07:47Z
issue,145014,Add ATen functions in native_functions.yaml to torch_in_graph_functions list automatically,https://github.com/pytorch/pytorch/issues/145014,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dynamo,scan when time allows,2025-01-22T15:24:47Z
issue,145242,Expose configurable path instead of using fixed path in the inductor module for serialized pattern generation,https://github.com/pytorch/pytorch/issues/145242,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",scan when time allows,2025-01-22T15:23:15Z
issue,140296,asynchronous copies from accelerator to cpu: what should be the expected behaviour?,https://github.com/pytorch/pytorch/issues/140296,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: python frontend",scan when time allows,2025-01-21T22:35:53Z
issue,142884,DataLoader hangs when object fails during pickling,https://github.com/pytorch/pytorch/issues/142884,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2025-01-21T21:31:18Z
issue,145043,Bug when using reparameterized model evaluating with DDP,https://github.com/pytorch/pytorch/issues/145043,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2025-01-21T21:14:33Z
issue,144913,UserWarning: cuDNN SDPA backward got grad_output.strides() != output.strides(),https://github.com/pytorch/pytorch/issues/144913,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cudnn, module: sdpa",scan when time allows,2025-01-21T17:42:01Z
issue,145145,[RFC] Improve performance for layer_norm op for cuda with revectorized,https://github.com/pytorch/pytorch/issues/145145,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: nn",scan when time allows,2025-01-21T17:39:26Z
issue,145040,Issue: Illegal Memory Access in Backward Pass of `scaled_dot_product_attention` with Custom Attention Mask,https://github.com/pytorch/pytorch/issues/145040,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sdpa,scan when time allows,2025-01-21T17:38:28Z
issue,144982,[compiled autograd] It would be nice if the compiiled autograd graph was actually runnable,https://github.com/pytorch/pytorch/issues/144982,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: compiled autograd, oncall: pt2",scan when time allows,2025-01-21T17:38:01Z
issue,121961,Numerical error when using torch.nn.functional.pad with a large array on MPS device,https://github.com/pytorch/pytorch/issues/121961,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: mps, module: nn",scan when time allows,2025-01-21T15:49:59Z
issue,144928,Add API to detect if activation checkpointing is enabled in the current region or not,https://github.com/pytorch/pytorch/issues/144928,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: activation checkpointing,scan when time allows,2025-01-21T15:34:11Z
issue,143913,Set up Mac builds with clang >= 17 even though Xcode only has at most clang 16,https://github.com/pytorch/pytorch/issues/143913,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: ci",scan when time allows,2025-01-20T20:22:20Z
issue,134462,Cannot create and distribute array in torch.func.grad,https://github.com/pytorch/pytorch/issues/134462,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dtensor, module: functorch, oncall: distributed",scan when time allows,2025-01-17T20:02:47Z
issue,144671,torch.stack for sequences,https://github.com/pytorch/pytorch/issues/144671,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: python frontend, module: typing",scan when time allows,2025-01-17T14:32:57Z
issue,139582,nn.LSTM documentation,https://github.com/pytorch/pytorch/issues/139582,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2025-01-17T14:21:03Z
issue,144960,Memory-efficient attention is not selected if inputs's ndim != 4,https://github.com/pytorch/pytorch/issues/144960,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sdpa,scan when time allows,2025-01-17T00:44:04Z
issue,144690,Support new CUDA conda package layout natively in cpp_extension.CUDAExtension,https://github.com/pytorch/pytorch/issues/144690,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp-extensions, module: cuda",scan when time allows,2025-01-16T23:19:12Z
issue,93737,Improve handling of generators,https://github.com/pytorch/pytorch/issues/93737,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",scan when time allows,2025-01-16T22:48:32Z
issue,131450,Make torch.compile support inter-op parallelism,https://github.com/pytorch/pytorch/issues/131450,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2025-01-16T21:58:39Z
issue,144890,Ways the HUD compilers dashboard could be better,https://github.com/pytorch/pytorch/issues/144890,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: devx,scan when time allows,2025-01-16T18:34:01Z
issue,144969,Support something similar to export dynamic dims for torch.compile with fullgraph=True,https://github.com/pytorch/pytorch/issues/144969,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamic shapes, oncall: pt2",scan when time allows,2025-01-16T15:38:16Z
issue,23425,Hanging on when one gpu node return zero as loss in the context of distributed data parallel training,https://github.com/pytorch/pytorch/issues/23425,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2025-01-16T10:16:27Z
issue,144375,Unable to compile models using tensorrt backend: CUDNN_STATUS_BAD_PARAM_STREAM_MISMATCH,https://github.com/pytorch/pytorch/issues/144375,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",scan when time allows,2025-01-16T06:03:22Z
issue,144875,[Monitoring] Display on HUD the information about runners that failed to be created (which cause jobs to queue),https://github.com/pytorch/pytorch/issues/144875,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2025-01-15T23:46:59Z
issue,144688,topK for sparse Vectors,https://github.com/pytorch/pytorch/issues/144688,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2025-01-15T20:34:31Z
issue,141832,Optimizers' `differentiable` flag doesn't work,https://github.com/pytorch/pytorch/issues/141832,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2025-01-15T18:58:01Z
issue,144861,Region check for in-place read and write does not always work,https://github.com/pytorch/pytorch/issues/144861,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: partial aliasing,scan when time allows,2025-01-15T18:16:12Z
issue,144611,[CUDA] Illegal Memory Access with `ConvTranspose2d`,https://github.com/pytorch/pytorch/issues/144611,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2025-01-15T18:04:56Z
issue,144612,[CUDA] Illegal Memory Access with `torch.bmm`,https://github.com/pytorch/pytorch/issues/144612,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2025-01-15T18:04:45Z
issue,144711,Tensor Stride Inconsistent (?) Behavior When One of the Dimension is 1,https://github.com/pytorch/pytorch/issues/144711,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: memory format,scan when time allows,2025-01-15T16:35:57Z
issue,144633,144x less efficient CPU usage when training NN past a certain width,https://github.com/pytorch/pytorch/issues/144633,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: arm, module: performance",scan when time allows,2025-01-15T15:54:41Z
issue,129627,[RFC][DTensor] DTensor Strided Sharding: A More Flexible Way To Shard Tensors,https://github.com/pytorch/pytorch/issues/129627,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2025-01-15T05:46:53Z
issue,141789,torch.roll runs too slow at MPS backend,https://github.com/pytorch/pytorch/issues/141789,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: mps, module: performance",scan when time allows,2025-01-15T02:05:44Z
issue,141594,Marking integer inputs to a compiled region requires `torch._dynamo.assume_static_by_default = True`,https://github.com/pytorch/pytorch/issues/141594,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamic shapes, oncall: pt2",scan when time allows,2025-01-14T17:41:59Z
issue,144677,Tensorboard SummaryWriter.add_hparams doesn't log hparam metrics dictionary,https://github.com/pytorch/pytorch/issues/144677,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tensorboard,scan when time allows,2025-01-14T14:20:50Z
issue,75721,Sparse tensors: indexing with `index` and/or being able run `sparse_tensor[index]`,https://github.com/pytorch/pytorch/issues/75721,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2025-01-14T13:31:19Z
issue,144478,[dynamo] Support using UserDefinedFunction as argument (as_proxy).,https://github.com/pytorch/pytorch/issues/144478,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",scan when time allows,2025-01-14T09:12:05Z
issue,121894,TORCH_USE_CUDA_DSA compiler flag not working,https://github.com/pytorch/pytorch/issues/121894,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2025-01-14T00:18:15Z
issue,135210,Incorrect calculation of standard deviation when using complex numbers on MPS backend,https://github.com/pytorch/pytorch/issues/135210,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: complex, module: correctness (silent), module: mps",scan when time allows,2025-01-13T19:09:26Z
issue,144673,Thread-safe approach on temporarily changing the `set_default_type`,https://github.com/pytorch/pytorch/issues/144673,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python frontend,scan when time allows,2025-01-13T18:36:08Z
issue,142029,[RFC] Add Intel GPU Support to Torch Test Cases,https://github.com/pytorch/pytorch/issues/142029,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: testing, module: xpu",scan when time allows,2025-01-13T18:19:51Z
issue,144660,Implement the `mode` property for transformed distributions,https://github.com/pytorch/pytorch/issues/144660,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distributions,scan when time allows,2025-01-13T17:18:07Z
issue,144643,nn.functional.interpolate doesn't work correctly with NCW tensors when output W exceeds 2^23,https://github.com/pytorch/pytorch/issues/144643,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: edge cases, module: nn",scan when time allows,2025-01-13T17:17:28Z
issue,144639,torch.vmap + autograd.Function + current_level bug,https://github.com/pytorch/pytorch/issues/144639,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: functorch, module: vmap",scan when time allows,2025-01-13T17:13:54Z
issue,143836,Integration of AdamCPR Optimizer into PyTorch,https://github.com/pytorch/pytorch/issues/143836,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2025-01-11T22:52:49Z
issue,142467,Wrong error message,https://github.com/pytorch/pytorch/issues/142467,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: error checking",scan when time allows,2025-01-11T16:43:05Z
issue,144403,Extended functionality for torch.quantization.fuse_modules,https://github.com/pytorch/pytorch/issues/144403,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2025-01-11T01:16:40Z
issue,134430,"[MPS] `F.interpolate` with `mode=""nearest-exact""` differs from expected output",https://github.com/pytorch/pytorch/issues/134430,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: mps",scan when time allows,2025-01-10T20:09:26Z
issue,73920,"pytorch/pytorch:1.8.1-cuda10.2-cudnn7-devel docker container contains cudnn 8.2, not 7.x as the name implies",https://github.com/pytorch/pytorch/issues/73920,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docker,scan when time allows,2025-01-10T03:31:11Z
issue,60069,Symbolic trace with *args,https://github.com/pytorch/pytorch/issues/60069,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2025-01-10T02:13:04Z
issue,137484,Better error message in `torch.linalg.vector_norm`,https://github.com/pytorch/pytorch/issues/137484,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: edge cases, module: error checking",scan when time allows,2025-01-10T02:11:42Z
issue,144495,c10/util/BFloat16-math.h has undefined behavior,https://github.com/pytorch/pytorch/issues/144495,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: core aten",scan when time allows,2025-01-09T22:36:01Z
issue,94691,Nan is output by GRU on mps,https://github.com/pytorch/pytorch/issues/94691,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: mps",scan when time allows,2025-01-09T19:38:38Z
issue,78929,Add type() support for mps backend,https://github.com/pytorch/pytorch/issues/78929,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: error checking, module: mps",scan when time allows,2025-01-09T19:31:26Z
issue,104106,batch size unexpectedly affects model inference on Mac M1,https://github.com/pytorch/pytorch/issues/104106,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: mps",scan when time allows,2025-01-09T19:26:46Z
issue,104832,"MPS matmul with sliced (strided) out argument produces wrong output, may corrupt memory",https://github.com/pytorch/pytorch/issues/104832,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: mps",scan when time allows,2025-01-09T19:26:41Z
issue,103023,MPS bug: padding_idx in nn.Embedding does not prevent gradient accumulation,https://github.com/pytorch/pytorch/issues/103023,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: mps",scan when time allows,2025-01-09T19:24:12Z
issue,107455,Moving tensor to MPS using .to(torch.device('mps') deletes entries from tensor,https://github.com/pytorch/pytorch/issues/107455,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: mps",scan when time allows,2025-01-09T19:23:27Z
issue,102852,The operator 'aten::poisson' is not currently implemented for the MPS device,https://github.com/pytorch/pytorch/issues/102852,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: mps,scan when time allows,2025-01-09T19:23:00Z
issue,107694,masked_fill_ outputs incorrect results for 'mps' tensor after transpose,https://github.com/pytorch/pytorch/issues/107694,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: mps",scan when time allows,2025-01-09T19:22:25Z
issue,109943,Problems when loading PT files und Linux - Duda which are created under Mac Apple Silicon MPS,https://github.com/pytorch/pytorch/issues/109943,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: mps,scan when time allows,2025-01-09T19:19:54Z
issue,111173,unique(return_counts=True) fails on MPS for unsorted tensors with 1M+ elements,https://github.com/pytorch/pytorch/issues/111173,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: mps",scan when time allows,2025-01-09T19:18:18Z
issue,111479,multi_head_attention_forward generates different values on MPS compared to CPU,https://github.com/pytorch/pytorch/issues/111479,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: mps",scan when time allows,2025-01-09T19:16:58Z
issue,101167,round float16 calculation error in mps backend,https://github.com/pytorch/pytorch/issues/101167,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: mps",scan when time allows,2025-01-09T19:15:16Z
issue,113062,MPS device: Sample from MultivariateNormal distribution,https://github.com/pytorch/pytorch/issues/113062,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: mps,scan when time allows,2025-01-09T19:13:15Z
issue,116857,Operator 'aten::_weight_norm_interface' not implemented MPS device.,https://github.com/pytorch/pytorch/issues/116857,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: mps,scan when time allows,2025-01-09T19:10:01Z
issue,100334,MPS device inference all same value,https://github.com/pytorch/pytorch/issues/100334,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: mps",scan when time allows,2025-01-09T18:59:59Z
issue,96982,[mps] conv1d outputs zeros,https://github.com/pytorch/pytorch/issues/96982,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: mps",scan when time allows,2025-01-09T18:51:27Z
issue,96225,[MPS] F.conv1d and F.conv2d produce incorrect gradients when minibatch >= 2^16,https://github.com/pytorch/pytorch/issues/96225,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: mps",scan when time allows,2025-01-09T18:50:51Z
issue,124029,rand_like on permuted parameter not working on MPS,https://github.com/pytorch/pytorch/issues/124029,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: mps",scan when time allows,2025-01-09T18:29:04Z
issue,125098,RuntimeError: MPS: Unsupported Border padding mode,https://github.com/pytorch/pytorch/issues/125098,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: mps,scan when time allows,2025-01-09T18:28:14Z
issue,87129,[MPS] sum on a size=1 dim is ~5x slower than squeeze,https://github.com/pytorch/pytorch/issues/87129,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: mps, module: performance",scan when time allows,2025-01-09T18:21:12Z
issue,134534,MiniCPM-V 2.6 transformer model returns nonsense tokens when running on MPS backend vs CPU backend.,https://github.com/pytorch/pytorch/issues/134534,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: mps",scan when time allows,2025-01-09T18:16:43Z
issue,136692,NotImplementedError: The operator 'aten::linalg_matrix_exp' is not currently implemented for the MPS device.,https://github.com/pytorch/pytorch/issues/136692,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: mps,scan when time allows,2025-01-09T18:14:04Z
issue,83015,Incorrect tensor conversion to m1 MPS.,https://github.com/pytorch/pytorch/issues/83015,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: mps",scan when time allows,2025-01-09T18:11:19Z
issue,137581,Expose underlying cudastream_t for torch.cuda.Stream,https://github.com/pytorch/pytorch/issues/137581,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: docs, module: python frontend",scan when time allows,2025-01-09T17:13:16Z
issue,9407,[feature request] Make `torch.gather` broadcastable.,https://github.com/pytorch/pytorch/issues/9407,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2025-01-09T15:06:04Z
issue,143564,[Bug] Memory leak in C++ libtorch,https://github.com/pytorch/pytorch/issues/143564,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp,scan when time allows,2025-01-08T12:13:25Z
issue,142703,We don't have test coverage for aarch64 (ARM64) bfloat16 feature (__ARM_FEATURE_BF16),https://github.com/pytorch/pytorch/issues/142703,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: ci",scan when time allows,2025-01-07T21:52:47Z
issue,140563,NCCL hangs cause timeout,https://github.com/pytorch/pytorch/issues/140563,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2025-01-07T21:43:59Z
issue,144328,Debug build fails to compile on x86 with WERROR=1,https://github.com/pytorch/pytorch/issues/144328,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2025-01-07T15:26:31Z
issue,141869,Use Python 3.9 type annotations,https://github.com/pytorch/pytorch/issues/141869,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: typing,scan when time allows,2025-01-07T12:18:42Z
issue,144116,RNN batch_first argument only works on the input not h_0 when if should work on both,https://github.com/pytorch/pytorch/issues/144116,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: rnn",scan when time allows,2025-01-07T06:14:08Z
issue,144210,Add inheritance possibilities,https://github.com/pytorch/pytorch/issues/144210,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2025-01-07T06:11:32Z
issue,44255,Allow to use system zstd,https://github.com/pytorch/pytorch/issues/44255,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2025-01-07T00:57:37Z
issue,142414,[dynamo] Tracking issue for multiple inheritance support,https://github.com/pytorch/pytorch/issues/142414,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",scan when time allows,2025-01-06T23:23:26Z
issue,144033,Training fails with Torch 2.1.0 on Nvidia Jetpack 5.1.2,https://github.com/pytorch/pytorch/issues/144033,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: jetson,scan when time allows,2025-01-06T15:38:30Z
issue,144217,PyTorch Enum PYI stubs are invalid. Fix typing of PYI stubs,https://github.com/pytorch/pytorch/issues/144217,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: lint, module: typing",scan when time allows,2025-01-06T01:57:43Z
issue,143658,Log more contextual data when nan is detected under the anomaly mode,https://github.com/pytorch/pytorch/issues/143658,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: nans and infs",scan when time allows,2025-01-06T01:53:54Z
issue,43255,Adding entropy function analogous to SciPy,https://github.com/pytorch/pytorch/issues/43255,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: numpy,scan when time allows,2025-01-05T23:12:57Z
issue,50334,Function Request: np.interp,https://github.com/pytorch/pytorch/issues/50334,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: interpolation, module: numpy",scan when time allows,2025-01-04T12:31:41Z
issue,142250,Embedding forward performance analysis,https://github.com/pytorch/pytorch/issues/142250,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",scan when time allows,2025-01-03T22:02:29Z
issue,144109,Uneven Sharding in DTensor Leads to unexpected tensor resolution with `full_tensor`,https://github.com/pytorch/pytorch/issues/144109,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dtensor, oncall: distributed",scan when time allows,2025-01-03T20:03:03Z
issue,132605,MPS lgamma function changes results when using broadcasting,https://github.com/pytorch/pytorch/issues/132605,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: mps",scan when time allows,2025-01-03T19:18:15Z
issue,143931,The model compiled with torch.compile encounters an error when run.,https://github.com/pytorch/pytorch/issues/143931,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",scan when time allows,2025-01-03T17:43:47Z
issue,135050,Setting a scalar and a 0D tensor or a 0D tensor and a scalar to `min` and `max` argument of `clamp()` respectively works,https://github.com/pytorch/pytorch/issues/135050,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: edge cases, module: python frontend",scan when time allows,2025-01-03T09:39:38Z
issue,10978,[feature request] padding for torch.cat,https://github.com/pytorch/pytorch/issues/10978,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: viewing and reshaping,scan when time allows,2025-01-02T18:48:55Z
issue,138460,"CUDA Binary dependency chain is wrong, leading to bad binary packaging",https://github.com/pytorch/pytorch/issues/138460,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: cuda",scan when time allows,2025-01-02T18:11:57Z
issue,144025,torch.cuda.empty_cache() causes extra memory usage on 'cuda:0',https://github.com/pytorch/pytorch/issues/144025,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: cudacachingallocator",scan when time allows,2025-01-02T15:36:06Z
issue,144040,Torch.sparse.mm failing gradient computation at half precision.,https://github.com/pytorch/pytorch/issues/144040,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: half, module: sparse",scan when time allows,2025-01-02T15:25:45Z
issue,144001,torch.utils.flop_counter.FlopCounterMode,https://github.com/pytorch/pytorch/issues/144001,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: flop counter,scan when time allows,2025-01-02T15:21:14Z
issue,132218,[profiler] CUDA runtime op has wrong device time because of lazy init,https://github.com/pytorch/pytorch/issues/132218,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2025-01-02T03:18:17Z
issue,127615,TORCH_LIBRARY breaks when passing (unexpanded) macro as namespace argument,https://github.com/pytorch/pytorch/issues/127615,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: library,scan when time allows,2025-01-01T19:17:25Z
issue,141822,F.cross_entropy unexpectedly slower than F.log_softmax + torch.gather,https://github.com/pytorch/pytorch/issues/141822,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: performance",scan when time allows,2025-01-01T01:30:05Z
issue,139281,`torch._numpy.ndarray.astype()` does not accept Numpy Dtypes correctly,https://github.com/pytorch/pytorch/issues/139281,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: numpy,scan when time allows,2024-12-31T16:40:05Z
issue,85773,Conda Pytorch (Pytorch channel) in WSL2 Ubuntu can't find libcudnn shared objects,https://github.com/pytorch/pytorch/issues/85773,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cuda",scan when time allows,2024-12-31T08:25:40Z
issue,88137,"Add eq, to, masked_select, index_select, narrow to nested tensors",https://github.com/pytorch/pytorch/issues/88137,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nestedtensor,scan when time allows,2024-12-31T07:46:08Z
issue,143947,Add differentiable flag to SGD,https://github.com/pytorch/pytorch/issues/143947,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2024-12-30T22:52:50Z
issue,48273,torch.multinomial example is incorrect,https://github.com/pytorch/pytorch/issues/48273,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: distributions, module: docs",scan when time allows,2024-12-30T19:01:11Z
issue,143965,test case got killed file_based_local_timer_test.py test_get_timer_recursive,https://github.com/pytorch/pytorch/issues/143965,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tests,scan when time allows,2024-12-30T14:55:03Z
issue,126344,PyTorch: RuntimeError: device >= 0 && device < num_gpus INTERNAL ASSERT FAILED,https://github.com/pytorch/pytorch/issues/126344,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2024-12-30T02:46:09Z
issue,122973,mps training is slower than cpu,https://github.com/pytorch/pytorch/issues/122973,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: mps, module: performance",scan when time allows,2024-12-28T19:24:44Z
issue,92132,Torchrun seems to have problem with virtual environment,https://github.com/pytorch/pytorch/issues/92132,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: oncall: distributed, oncall: r2p",scan when time allows,2024-12-28T13:50:41Z
issue,106221,RProp improvement tracker,https://github.com/pytorch/pytorch/issues/106221,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2024-12-27T19:20:31Z
issue,126691,TORCH_CHECK used within torch.compile does not throw legible errors,https://github.com/pytorch/pytorch/issues/126691,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: cpu inductor,scan when time allows,2024-12-27T02:36:54Z
issue,143860,_transform_bias_rescale_qkv cpu op get error on debug build,https://github.com/pytorch/pytorch/issues/143860,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-12-27T02:14:28Z
issue,139602,[RFC]: Adding Triton Backend for Aten operators,https://github.com/pytorch/pytorch/issues/139602,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",scan when time allows,2024-12-25T19:25:00Z
issue,113750,[docs] Add reference/decomp impl snippets to functions in online docs for hackability and educational purposes / compensate for some unclear language in existing docs,https://github.com/pytorch/pytorch/issues/113750,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: norms and normalization",scan when time allows,2024-12-25T18:20:22Z
issue,143561,"""aten::any"" and ""aten::all"" behavior for unsigned tensors different than signed",https://github.com/pytorch/pytorch/issues/143561,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: reductions, module: unsigned int",scan when time allows,2024-12-25T15:06:57Z
issue,74669,String dtypes for torch Tensors,https://github.com/pytorch/pytorch/issues/74669,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2024-12-25T14:24:17Z
issue,120407,Dark mode please 🙏🏻,https://github.com/pytorch/pytorch/issues/120407,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2024-12-24T09:20:58Z
issue,143754,nn.LayerNorm is slower than naive implementation when dimension is low,https://github.com/pytorch/pytorch/issues/143754,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: norms and normalization, module: performance",scan when time allows,2024-12-23T17:12:29Z
issue,143588,Full BFGS optimizer,https://github.com/pytorch/pytorch/issues/143588,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2024-12-23T17:11:02Z
issue,143669,String representation of nn.MultiheadAttention should contain arguments,https://github.com/pytorch/pytorch/issues/143669,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-12-23T14:28:07Z
issue,139005,Dependency management corrupted in pytorch 2.5.0,https://github.com/pytorch/pytorch/issues/139005,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: binaries,scan when time allows,2024-12-21T00:25:32Z
issue,143655,反射填充的反向传播没有确定性实现,https://github.com/pytorch/pytorch/issues/143655,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: determinism,scan when time allows,2024-12-20T22:31:35Z
issue,143453,Compiler Bisector Improvements,https://github.com/pytorch/pytorch/issues/143453,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: inductor,scan when time allows,2024-12-20T17:40:37Z
issue,142301,Unfold,https://github.com/pytorch/pytorch/issues/142301,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-12-20T17:35:27Z
issue,138360,[ROCm] PyTorch Profiler Seg Fault on PyTorch Nightly,https://github.com/pytorch/pytorch/issues/138360,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: rocm, oncall: profiler",scan when time allows,2024-12-20T16:50:22Z
issue,47917,LibTorch cannot load PyTorch exported model,https://github.com/pytorch/pytorch/issues/47917,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: serialization",scan when time allows,2024-12-20T08:41:41Z
issue,143440,Locale issues in colab: after tensor(1j).cuda().abs() !commands cannot be executed.,https://github.com/pytorch/pytorch/issues/143440,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: python frontend, module: third_party",scan when time allows,2024-12-19T22:55:25Z
issue,52675,[feature request] Introduce env var `TORCH_USE_CUDA=OFF` and some `torch.cuda.enabled = False` to faithfully/completely disable cuda (torch.cuda.is_available() <- False and torch.cuda.get_device_count() <- 0) and driver loading/interaction,https://github.com/pytorch/pytorch/issues/52675,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2024-12-19T19:09:14Z
issue,89491,build: cmake: functorch.so not installed at expected location,https://github.com/pytorch/pytorch/issues/89491,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2024-12-19T07:35:15Z
issue,91524,functorch.so is installed back into the source directory,https://github.com/pytorch/pytorch/issues/91524,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: functorch",scan when time allows,2024-12-19T07:34:28Z
issue,141746,How to specify the port for processes with rank > 1 in the Gloo communication backend?,https://github.com/pytorch/pytorch/issues/141746,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2024-12-19T03:52:52Z
issue,135611,Indexing Numpy array with Tensor Gives Unexpected Results,https://github.com/pytorch/pytorch/issues/135611,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: numpy,scan when time allows,2024-12-19T00:44:29Z
issue,143424,higher rank convolution,https://github.com/pytorch/pytorch/issues/143424,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-12-18T22:58:26Z
issue,143497,No reproducibility after ONNX export of fully converted QAT model,https://github.com/pytorch/pytorch/issues/143497,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: onnx, oncall: quantization",scan when time allows,2024-12-18T16:27:44Z
issue,143334,Applying python 'next' directly on a tensor is painfully slow,https://github.com/pytorch/pytorch/issues/143334,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python frontend,scan when time allows,2024-12-18T14:48:22Z
issue,109539,Torch FX SubgraphMatcher Any / Oneof Patterns,https://github.com/pytorch/pytorch/issues/109539,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: fx, module: fx.passes",scan when time allows,2024-12-18T11:50:25Z
issue,143434,Missing nightly 20241217 on x86_64,https://github.com/pytorch/pytorch/issues/143434,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: binaries,scan when time allows,2024-12-18T00:18:37Z
issue,142025,[export] NotImplementedError: No registered serialization name for <class 'diffusers.models.modeling_outputs.Transformer2DModelOutput'>,https://github.com/pytorch/pytorch/issues/142025,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: oncall: export, oncall: pt2",scan when time allows,2024-12-17T22:09:41Z
issue,139305,torch.export torchaudio kaldi module,https://github.com/pytorch/pytorch/issues/139305,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: oncall: export, oncall: pt2",scan when time allows,2024-12-17T22:09:40Z
issue,137274,Not decomposing CompositeImplicitAutograd ops in post-dispatch export changes autocast (and possibly more) behavior when compared with eager mode,https://github.com/pytorch/pytorch/issues/137274,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: oncall: export, oncall: pt2",scan when time allows,2024-12-17T22:09:37Z
issue,143079,PyTorch depends on custom actions stored in contributor personal repos,https://github.com/pytorch/pytorch/issues/143079,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2024-12-17T21:51:46Z
issue,26340,CPU version of PyTorch on PyPI,https://github.com/pytorch/pytorch/issues/26340,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cpu, oncall: releng",scan when time allows,2024-12-17T20:19:08Z
issue,142176,torch.fx.GraphModule constructor doesn't copy all parameters/buffers from source graph module,https://github.com/pytorch/pytorch/issues/142176,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: oncall: export, oncall: fx, oncall: pt2",scan when time allows,2024-12-17T19:29:24Z
issue,139429,Add 'no_type_conversion' flag to register_buffer(),https://github.com/pytorch/pytorch/issues/139429,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2024-12-17T07:11:07Z
issue,143274,torch._logging.set_logs kind of sucks for Jupyter notebooks,https://github.com/pytorch/pytorch/issues/143274,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: logging,scan when time allows,2024-12-17T03:01:50Z
issue,143293,`bias=False` fails in `Transformer` when `batch_first=True` and in eval mode,https://github.com/pytorch/pytorch/issues/143293,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-12-17T03:00:39Z
issue,143306,Add CPU scalar support in addcdiv,https://github.com/pytorch/pytorch/issues/143306,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python frontend,scan when time allows,2024-12-17T02:41:36Z
issue,143206,Define release branch creation readiness process,https://github.com/pytorch/pytorch/issues/143206,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: releng,scan when time allows,2024-12-17T01:19:44Z
issue,141417,"Could you provide a standardized interface or documentation for mapping all versions of torch, torchvision, and torchaudio?",https://github.com/pytorch/pytorch/issues/141417,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, oncall: releng",scan when time allows,2024-12-17T01:19:43Z
issue,141675,Stuck at dynamo with dynamic shape enabled in PT2.5.1 which was not observed PT2.4.0.,https://github.com/pytorch/pytorch/issues/141675,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamic shapes, oncall: pt2",scan when time allows,2024-12-17T00:25:01Z
issue,126921,[ONNX] view(dtype=dtype) is not supported by both onnx.export and onnx.dynamo_export,https://github.com/pytorch/pytorch/issues/126921,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2024-12-16T20:53:23Z
issue,142104,[compiled autograd] Does compiled autograd under torch 2.4 support using selective checkpoint?,https://github.com/pytorch/pytorch/issues/142104,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: compiled autograd, oncall: pt2",scan when time allows,2024-12-16T20:17:29Z
issue,142392,Stft changes number of times when with center=False and n_fft > win_length,https://github.com/pytorch/pytorch/issues/142392,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: fft, module: scipy compatibility",scan when time allows,2024-12-16T18:17:50Z
issue,61309,The class weights implementation for loss functions is incorrect,https://github.com/pytorch/pytorch/issues/61309,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: loss, module: nn",scan when time allows,2024-12-16T17:43:50Z
issue,78809,Feature Request: Hessenberg and Schur decompositions,https://github.com/pytorch/pytorch/issues/78809,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: linear algebra,scan when time allows,2024-12-16T13:32:11Z
issue,143109,Failed to build Pytorch in Alpine Linux based image,https://github.com/pytorch/pytorch/issues/143109,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, oncall: distributed",scan when time allows,2024-12-16T08:21:24Z
issue,143189,`torch.linalg.qr_ex` and `torch.linalg.lstsq_ex`,https://github.com/pytorch/pytorch/issues/143189,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: linear algebra,scan when time allows,2024-12-15T10:48:42Z
issue,68703,Feature request: Extend `torch.eye` for creating a batch of eye matrices,https://github.com/pytorch/pytorch/issues/68703,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tensor creation,scan when time allows,2024-12-15T08:43:02Z
issue,89884,[RFC] PyTorch Tensor Parallel(TP) User API for Distributed Training,https://github.com/pytorch/pytorch/issues/89884,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dtensor,scan when time allows,2024-12-14T07:56:40Z
issue,142146,Inconsistent Presence of 'out' Arg in cross_entropy_loss vs Other Loss Terms in Python Bindings,https://github.com/pytorch/pytorch/issues/142146,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: loss, module: nn",scan when time allows,2024-12-13T20:55:13Z
issue,142852,Deprecate `torch.squeeze()` with unspecified `dim`,https://github.com/pytorch/pytorch/issues/142852,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: python frontend, module: viewing and reshaping",scan when time allows,2024-12-13T20:31:24Z
issue,143208,We should switch Tensor object (and other perf-critical objects) to use the new managed dict feature from CPython,https://github.com/pytorch/pytorch/issues/143208,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python frontend,scan when time allows,2024-12-13T19:38:16Z
issue,142229,[Pipelining] Backward pass tries to detach view tensor in-place,https://github.com/pytorch/pytorch/issues/142229,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2024-12-13T17:25:59Z
issue,142875,[user empathy day] No speedups for SAM2 model,https://github.com/pytorch/pytorch/issues/142875,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: performance, oncall: pt2",scan when time allows,2024-12-12T17:13:48Z
issue,142825,"Add [low, high] interval for torch.distributions.beta",https://github.com/pytorch/pytorch/issues/142825,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distributions,scan when time allows,2024-12-12T15:27:10Z
issue,115024,InstanceNorm1d throws bad warning,https://github.com/pytorch/pytorch/issues/115024,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-12-12T09:51:11Z
issue,64043,Uncleared memory use after torch.load(),https://github.com/pytorch/pytorch/issues/64043,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: memory usage, module: serialization",scan when time allows,2024-12-12T09:22:02Z
issue,142484,`RuntimeError: MKL FFT error: Intel oneMKL DFTI ERROR: Inconsistent configuration parameters` in `TestJitCPU.test_variant_consistency_jit_fft_fft2_cpu_complex64` against mkl 2024*,https://github.com/pytorch/pytorch/issues/142484,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: intel, module: linear algebra, module: mkl",scan when time allows,2024-12-12T02:34:02Z
issue,142815,torch.linalg.solve fails on CPU with multiple threads and batch dimension,https://github.com/pytorch/pytorch/issues/142815,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: linear algebra, module: mkl",scan when time allows,2024-12-12T00:08:03Z
issue,139131,[FlexAttention] Update the way we generate kernel options,https://github.com/pytorch/pytorch/issues/139131,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: flex attention, module: higher order operators, module: inductor, module: pt2-dispatcher",scan when time allows,2024-12-11T15:14:14Z
issue,142493,Custom ops are not logged to dynamo_compile/TORCH_TRACE when an error occurs,https://github.com/pytorch/pytorch/issues/142493,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: logging, oncall: pt2",scan when time allows,2024-12-11T12:15:22Z
issue,142515,[MPS] `out_channels` <= 2**16 guard for convolution is too broad,https://github.com/pytorch/pytorch/issues/142515,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: error checking, module: mps",scan when time allows,2024-12-11T02:37:47Z
issue,139908,Inductor vs. Liger Performance Track,https://github.com/pytorch/pytorch/issues/139908,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",scan when time allows,2024-12-10T23:44:27Z
issue,67598,unnecessary warning with autocast,https://github.com/pytorch/pytorch/issues/67598,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision),scan when time allows,2024-12-10T22:04:39Z
issue,58533,Migrate wheels to manylinux2010,https://github.com/pytorch/pytorch/issues/58533,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: releng,scan when time allows,2024-12-10T20:15:53Z
issue,55948,Deprecate torch.stft returning real-valued tensors and torch.istft accepting real-valued inputs,https://github.com/pytorch/pytorch/issues/55948,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: deprecation, module: fft",scan when time allows,2024-12-10T20:15:18Z
issue,129732,Torch Distributed Pipelinging API datatype mismatch,https://github.com/pytorch/pytorch/issues/129732,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: pipelining, oncall: distributed",scan when time allows,2024-12-10T19:56:02Z
issue,114153,[Tracker] Inconsistencies between CPU and GPU computation,https://github.com/pytorch/pytorch/issues/114153,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: numerical-stability",scan when time allows,2024-12-10T18:26:21Z
issue,138800,Batching rule not defined for `aten::_make_dual`.,https://github.com/pytorch/pytorch/issues/138800,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: functorch, module: vmap",scan when time allows,2024-12-10T18:17:27Z
issue,115618,"RuntimeError: handle_0 INTERNAL ASSERT FAILED at ""../c10/cuda/driver_api.cpp"":15",https://github.com/pytorch/pytorch/issues/115618,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2024-12-10T13:25:57Z
issue,140222,inconsistency in ```torch.nn.functional.log_softmax``` on CPU and GPU,https://github.com/pytorch/pytorch/issues/140222,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bfloat16, module: numerical-stability",scan when time allows,2024-12-10T07:03:05Z
issue,142469,Using DeviceMesh Slicing with torch.compile(),https://github.com/pytorch/pytorch/issues/142469,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: devicemesh, oncall: distributed",scan when time allows,2024-12-10T06:33:19Z
issue,142331,torch.is_vulkan_available() false in windows,https://github.com/pytorch/pytorch/issues/142331,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: vulkan,scan when time allows,2024-12-09T22:19:04Z
issue,141963,Do an audit on skipfiles and mark more files as inline,https://github.com/pytorch/pytorch/issues/141963,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tests,scan when time allows,2024-12-09T21:45:39Z
issue,142315,Better fusion on triton templates,https://github.com/pytorch/pytorch/issues/142315,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",scan when time allows,2024-12-09T21:31:26Z
issue,141472,[sparse] `mul`: missing support with CSC arguments,https://github.com/pytorch/pytorch/issues/141472,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2024-12-09T17:36:30Z
issue,142353,Extend custom op fwd/bwd functions to support data types beyond the basic set,https://github.com/pytorch/pytorch/issues/142353,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: custom-operators,scan when time allows,2024-12-09T06:05:46Z
issue,22049,Cannot update part of the parameters in DistributedDataParallel.,https://github.com/pytorch/pytorch/issues/22049,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2024-12-08T11:39:42Z
issue,141934,[Dynamo] Symbolic variables are reduced to Ints in compiled graph inputs,https://github.com/pytorch/pytorch/issues/141934,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamic shapes, oncall: pt2",scan when time allows,2024-12-06T21:59:29Z
issue,142171,[3.13] Fix failing module tracker dynamo-wrapped test due to recursion error,https://github.com/pytorch/pytorch/issues/142171,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",scan when time allows,2024-12-06T21:12:42Z
issue,140881,Implement Cut Cross Entropy,https://github.com/pytorch/pytorch/issues/140881,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: python frontend",scan when time allows,2024-12-06T20:06:01Z
issue,137182,torch_np tests failing with Numpy 2.1.1,https://github.com/pytorch/pytorch/issues/137182,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: numpy, module: tests",scan when time allows,2024-12-06T07:35:54Z
issue,141964,Handling custom context managers,https://github.com/pytorch/pytorch/issues/141964,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",scan when time allows,2024-12-06T05:26:16Z
issue,141216,Implement ADOPT Optimizer into torch.optim module,https://github.com/pytorch/pytorch/issues/141216,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2024-12-05T22:46:57Z
issue,142174,[3.13] Fix skipped numpy 2.0 dynamo-wrapped tests,https://github.com/pytorch/pytorch/issues/142174,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: numpy, module: python frontend, module: tests",scan when time allows,2024-12-05T21:53:40Z
issue,128395,_foreach_atan2,https://github.com/pytorch/pytorch/issues/128395,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: mta, module: optimizer",scan when time allows,2024-12-05T21:21:58Z
issue,142129,Feature Request - Backport torch.special.gammainc for Backward Compatibility,https://github.com/pytorch/pytorch/issues/142129,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python frontend,scan when time allows,2024-12-05T19:21:25Z
issue,124216,[NestedTensor] NJT with lengths set fails torch.compile(),https://github.com/pytorch/pytorch/issues/124216,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nestedtensor,scan when time allows,2024-12-05T18:59:41Z
issue,142127,[Bug]need check for duplicate malloc calls in CUDAPluggableAllocator,https://github.com/pytorch/pytorch/issues/142127,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: cudacachingallocator",scan when time allows,2024-12-05T16:41:01Z
issue,141018,The description of key_padding_mask in float mode is incorrect,https://github.com/pytorch/pytorch/issues/141018,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-12-05T12:55:00Z
issue,12449,"Could not find a package configuration file provided by ""Torch"" with any of the following names:",https://github.com/pytorch/pytorch/issues/12449,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp-extensions,scan when time allows,2024-12-04T08:17:16Z
issue,134682,RuntimeError: cuDNN version incompatibility: happened when I have lstm layer,https://github.com/pytorch/pytorch/issues/134682,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cudnn",scan when time allows,2024-12-04T05:46:18Z
issue,28829,torch.cuda.close request to be able to reset communications,https://github.com/pytorch/pytorch/issues/28829,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2024-12-04T03:49:33Z
issue,141977,frozen modules,https://github.com/pytorch/pytorch/issues/141977,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: binaries,scan when time allows,2024-12-03T23:39:16Z
issue,141801,[typing] add `nn.Module` overload type hint to `torch.compile`.,https://github.com/pytorch/pytorch/issues/141801,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: typing",scan when time allows,2024-12-03T23:02:23Z
issue,141828,Adding to Private Type Stubs,https://github.com/pytorch/pytorch/issues/141828,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: typing",scan when time allows,2024-12-03T23:01:28Z
issue,141758,Test in `test/test_overrides.py` failed with AssertionError,https://github.com/pytorch/pytorch/issues/141758,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: testing,scan when time allows,2024-12-03T23:00:57Z
issue,139643,[triton 3.2] test_convolution2,https://github.com/pytorch/pytorch/issues/139643,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",scan when time allows,2024-12-03T22:47:35Z
issue,141764,Replace Miniconda with minforge as option,https://github.com/pytorch/pytorch/issues/141764,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2024-12-03T22:07:24Z
issue,141558,Incremental CI testing,https://github.com/pytorch/pytorch/issues/141558,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2024-12-03T21:49:15Z
issue,141690,Reworking is_big_gpu check for inductor GEMMs,https://github.com/pytorch/pytorch/issues/141690,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",scan when time allows,2024-12-03T19:18:46Z
issue,141560,[Feature Request] Add source Inductor Python file path to each Triton kernel,https://github.com/pytorch/pytorch/issues/141560,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: pt2,scan when time allows,2024-12-03T19:13:25Z
issue,141727,"inductor and eager gives slightly different results for rms_norm, with `emulate_precision_casts=True`",https://github.com/pytorch/pytorch/issues/141727,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: pt2,scan when time allows,2024-12-03T18:26:43Z
issue,141139,Consider noting the PyTorch version for BC-breaking error messages,https://github.com/pytorch/pytorch/issues/141139,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: bc-breaking,scan when time allows,2024-12-03T15:54:06Z
issue,141890,Does vmap accept None out_dims?,https://github.com/pytorch/pytorch/issues/141890,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: vmap,scan when time allows,2024-12-03T15:50:16Z
issue,141802,inconsistency in ```torch.Tensor.logcumsumexp``` on CPU and GPU,https://github.com/pytorch/pytorch/issues/141802,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: numerical-stability, module: python frontend",scan when time allows,2024-12-03T15:47:08Z
issue,140640,"Identical Formulas for BatchNorm, LayerNorm, InstanceNorm, GroupNorm",https://github.com/pytorch/pytorch/issues/140640,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2024-12-03T12:44:53Z
issue,141915,Test failures when migrating to SM89,https://github.com/pytorch/pytorch/issues/141915,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: tests",scan when time allows,2024-12-03T04:59:14Z
issue,139381,Improved definition for Hardsigmoid,https://github.com/pytorch/pytorch/issues/139381,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-12-03T00:32:14Z
issue,140345,Flaky bot might be missing some data when making a judgement,https://github.com/pytorch/pytorch/issues/140345,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2024-12-02T23:31:00Z
issue,133833,Unexpected behavior of `CrossEntropyLoss()`,https://github.com/pytorch/pytorch/issues/133833,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-12-02T23:17:09Z
issue,141845,Using torch.func with saved tensor hooks,https://github.com/pytorch/pytorch/issues/141845,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: functorch",scan when time allows,2024-12-02T16:33:39Z
issue,141749,Incorrect error message on incorrect byte boundries on _scaled_mm on fp8,https://github.com/pytorch/pytorch/issues/141749,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: floatx (formerly float8),scan when time allows,2024-12-02T16:02:11Z
issue,112376,Contradictory Error Message for stride Argument in torch.conv_transpose3d(),https://github.com/pytorch/pytorch/issues/112376,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: error checking",scan when time allows,2024-12-02T14:14:16Z
issue,125297,Make the `sccache` cache easily available to all pytorch contributors in readonly mode,https://github.com/pytorch/pytorch/issues/125297,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: devx,scan when time allows,2024-12-02T12:05:18Z
issue,128412,A Permute Layer for torch.nn.Sequential,https://github.com/pytorch/pytorch/issues/128412,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-12-02T10:18:02Z
issue,139019,Stabilizing the Kumaraswamy Distribution,https://github.com/pytorch/pytorch/issues/139019,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: distributions, module: numerical-stability",scan when time allows,2024-11-30T18:38:17Z
issue,133304,The performance regression of `conv_transpose1d` on CPU with variable-shaped inputs,https://github.com/pytorch/pytorch/issues/133304,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: intel",scan when time allows,2024-11-29T07:59:42Z
issue,113128,NCCL watchdog thread terminated with exception,https://github.com/pytorch/pytorch/issues/113128,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nccl, oncall: distributed",scan when time allows,2024-11-28T08:31:07Z
issue,78109,Doesn't work when register hook to torch.nn.MultiheadAttention.out_proj,https://github.com/pytorch/pytorch/issues/78109,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2024-11-27T21:52:16Z
issue,72450,Memory parity with JAX attention,https://github.com/pytorch/pytorch/issues/72450,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: memory usage,scan when time allows,2024-11-27T21:52:06Z
issue,57117,MultiheadAttention.out_proj.weight is not explicitly initialized,https://github.com/pytorch/pytorch/issues/57117,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2024-11-27T21:51:57Z
issue,139191,"When I use register_full_backward_hook and register_forward_hook in nn.MultiheadAttention, The MultiheadAttention executed wrong conditional branch and get wrong backward grads.",https://github.com/pytorch/pytorch/issues/139191,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-11-27T21:51:48Z
issue,124931,"Missing description of Transformer argument ""memory_mask"" shape in 3D (including the batch dimension) case",https://github.com/pytorch/pytorch/issues/124931,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-11-27T21:51:43Z
issue,91435,Batch_first attribute in quantizable multiheadattention,https://github.com/pytorch/pytorch/issues/91435,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-11-27T21:51:34Z
issue,80738,Output for `aten::_native_multi_head_attention` appears inconsistent with entry in `native_functions.yaml`,https://github.com/pytorch/pytorch/issues/80738,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-11-27T21:51:16Z
issue,96633,Request for adding Warning/Error feature when dropout set to 1.0 in Transformer layer,https://github.com/pytorch/pytorch/issues/96633,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-11-27T21:51:04Z
issue,94428,nn.TransformerEncoderLayer fastpath (BetterTransformer) is much slower with src_key_padding_mask,https://github.com/pytorch/pytorch/issues/94428,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-11-27T21:50:55Z
issue,95210,Add parallel attention layers and Multi-Query Attention (MQA) from PaLM to the fast path for transformers,https://github.com/pytorch/pytorch/issues/95210,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-11-27T21:50:51Z
issue,90842,nn.MultiheadAttention softmax inconsistent in training mode,https://github.com/pytorch/pytorch/issues/90842,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-11-27T21:50:48Z
issue,92990,Feature request: access to variable,https://github.com/pytorch/pytorch/issues/92990,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-11-27T21:50:47Z
issue,115362,torch._fused_sdp_choice for fake and real tensors returns different results,https://github.com/pytorch/pytorch/issues/115362,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: faketensor, module: sdpa",scan when time allows,2024-11-27T21:50:29Z
issue,120793,"libtorch MultiheadAttention doesn't accept different dimensions for Q, K, V",https://github.com/pytorch/pytorch/issues/120793,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-11-27T21:50:26Z
issue,113713,[cuDNN][cuDNN V8 API] cuDNN Flash-Attention Upstreaming RFC/tracking issue,https://github.com/pytorch/pytorch/issues/113713,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: cudnn, module: sdpa",scan when time allows,2024-11-27T21:50:21Z
issue,119188,scaled_dot_product_attention: different output between the reference implementation and the one from torch.nn.functional,https://github.com/pytorch/pytorch/issues/119188,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sdpa,scan when time allows,2024-11-27T21:50:17Z
issue,57721,nn.TransformerEncoder cannot deal with large negative value even when this value is masked by src_key_padding_mask,https://github.com/pytorch/pytorch/issues/57721,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: nn",scan when time allows,2024-11-27T21:49:37Z
issue,64006,Missing device and dtype description in torch.nn.MultiheadAttention,https://github.com/pytorch/pytorch/issues/64006,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: nn",scan when time allows,2024-11-27T21:49:14Z
issue,67999,Improved Transformer and MultiHeadAttention design,https://github.com/pytorch/pytorch/issues/67999,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-11-27T21:49:04Z
issue,66678,Adding modern linear transformer variants to `transformer` module,https://github.com/pytorch/pytorch/issues/66678,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-11-27T21:48:47Z
issue,64509,Return attention weights in nn.Transformer,https://github.com/pytorch/pytorch/issues/64509,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-11-27T21:43:25Z
issue,64525,nn.TransformerEncoder - all nan values issues when src_key_padding_mask provided,https://github.com/pytorch/pytorch/issues/64525,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nans and infs, module: nn",scan when time allows,2024-11-27T21:43:14Z
issue,60939,Functional multi_head_attention_forward softmax get nan for fp16 mode,https://github.com/pytorch/pytorch/issues/60939,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-11-27T21:43:06Z
issue,112015,op scaled_dot_product_attention case different results,https://github.com/pytorch/pytorch/issues/112015,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sdpa,scan when time allows,2024-11-27T21:36:53Z
issue,120790,MultiheadAttention returns incorrect results for windowed long causal mask for long input,https://github.com/pytorch/pytorch/issues/120790,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-11-27T21:36:48Z
issue,118972,Inconsistent usage of `attn_mask` and `is_causal` arguments,https://github.com/pytorch/pytorch/issues/118972,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-11-27T21:36:39Z
issue,118628,TransformerEncoderLayer fast path predicts NaN when provided attention bias,https://github.com/pytorch/pytorch/issues/118628,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-11-27T21:36:34Z
issue,116077,TransformerEncoderLayer and TransformerDecoderLayer do not work,https://github.com/pytorch/pytorch/issues/116077,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-11-27T21:36:31Z
issue,100347,nn.MultiheadAttention doesn't use efficient scaled_dot_product_attention,https://github.com/pytorch/pytorch/issues/100347,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: nn",scan when time allows,2024-11-27T21:36:25Z
issue,100087,TransformerEncoderLayer behavior inconsistent between training and evaluation mode,https://github.com/pytorch/pytorch/issues/100087,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-11-27T21:36:21Z
issue,98291,Make BetterTransformer implementation non-blocking,https://github.com/pytorch/pytorch/issues/98291,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-11-27T21:35:31Z
issue,96883,Expected scalar type Half but found Float when running nn.MultiheadAttention with AMP,https://github.com/pytorch/pytorch/issues/96883,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-11-27T21:35:27Z
issue,98203,"AssertionError: was expecting embedding dimension of 22, but got 1320",https://github.com/pytorch/pytorch/issues/98203,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-11-27T21:35:23Z
issue,116472,Using a 4D attention mask in MultiheadAttention forward,https://github.com/pytorch/pytorch/issues/116472,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-11-27T21:34:53Z
issue,105925,Possible speed up of nn.MultiheadAttention,https://github.com/pytorch/pytorch/issues/105925,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-11-27T21:34:42Z
issue,106630,no_grad() changes output of TransformerDecoder module during evaluation,https://github.com/pytorch/pytorch/issues/106630,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-11-27T21:34:37Z
issue,110702,sliding_window attention in scaled_dot_product,https://github.com/pytorch/pytorch/issues/110702,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sdpa,scan when time allows,2024-11-27T21:34:24Z
issue,124937,Conflict between bias=False and why_not_sparsity_fast_path in Transformer Module,https://github.com/pytorch/pytorch/issues/124937,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-11-27T21:33:43Z
issue,138552,The default setting of PyTorch should show warning always.,https://github.com/pytorch/pytorch/issues/138552,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nestedtensor, module: nn",scan when time allows,2024-11-27T21:33:40Z
issue,138993,Support `attn_mask` in `jagged_scaled_dot_product_attention`,https://github.com/pytorch/pytorch/issues/138993,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nestedtensor, module: sdpa",scan when time allows,2024-11-27T21:33:19Z
issue,140930,[cudnn_frontend] Error: No execution plans support the graph when using CUDNN_ATTENTION in SDPA,https://github.com/pytorch/pytorch/issues/140930,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: cudnn, module: sdpa",scan when time allows,2024-11-27T21:32:49Z
issue,69321,Poor Transformer inference scalability on CPU,https://github.com/pytorch/pytorch/issues/69321,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: performance",scan when time allows,2024-11-27T20:13:22Z
issue,135150,Undocumented fast pass behavior in nn.TransformerEncoderLayer causes failures in test_transformerencoderlayer_cuda_float32 (__main__.TestNNDeviceTypeCUDA),https://github.com/pytorch/pytorch/issues/135150,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-11-27T20:12:24Z
issue,121193,Detailed documentation with an example for the causal mask in nn.TransformerEncoder,https://github.com/pytorch/pytorch/issues/121193,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-11-27T20:12:24Z
issue,118853,MultiheadAttention returns different results for binary mask and float mask,https://github.com/pytorch/pytorch/issues/118853,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-11-27T20:12:23Z
issue,113590,Implement FlashFFTConv algorithm,https://github.com/pytorch/pytorch/issues/113590,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cudnn, module: nn, module: sdpa",scan when time allows,2024-11-27T20:12:23Z
issue,115472,Potential issue with custom transformer masks when using fast path and batch_first=True,https://github.com/pytorch/pytorch/issues/115472,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-11-27T20:12:22Z
issue,132700,Forward Hooks in Transformer Not Working When `model.eval()` and `torch.no_grad()`,https://github.com/pytorch/pytorch/issues/132700,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sdpa,scan when time allows,2024-11-27T20:12:11Z
issue,138907,Select SDPA backend smartly by `sdp_params` and benchmark results,https://github.com/pytorch/pytorch/issues/138907,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sdpa,scan when time allows,2024-11-27T20:11:44Z
issue,138649,SDPA 2.5 Issue tracking,https://github.com/pytorch/pytorch/issues/138649,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sdpa,scan when time allows,2024-11-27T20:11:43Z
issue,124718,"Add scaled_dot_product_attention ""scale"" argument to nn.MultiHeadAttention",https://github.com/pytorch/pytorch/issues/124718,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sdpa,scan when time allows,2024-11-27T20:11:41Z
issue,113524,_fused_sdp_choice returns wrong backend in pt2,https://github.com/pytorch/pytorch/issues/113524,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sdpa,scan when time allows,2024-11-27T20:11:41Z
issue,110154,"ValueError: args contained 2 None's after flattening. When exporting a ScriptModule or ScriptFunction, no args may be None because that breaks type propagation.",https://github.com/pytorch/pytorch/issues/110154,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: onnx",scan when time allows,2024-11-27T20:10:40Z
issue,135590,Discussion about the testing methods in test_transformers.py after math backend always uses fp32,https://github.com/pytorch/pytorch/issues/135590,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sdpa,scan when time allows,2024-11-27T20:10:03Z
issue,133780,2.4.0: sdpa disallows batch-of-0 in Flash (default backend) and cuDNN.,https://github.com/pytorch/pytorch/issues/133780,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cudnn, module: sdpa",scan when time allows,2024-11-27T20:10:03Z
issue,129523,"`test_dummy_mha_with_nt_cuda` fails on `sm70`, `sm75`",https://github.com/pytorch/pytorch/issues/129523,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: nestedtensor, module: sdpa",scan when time allows,2024-11-27T20:10:01Z
issue,123911,"`torch.softmax(inp, dtype=torch.float32).to(torch.float16)` is not equivalent to `torch.softmax(inp)` for fp16 input",https://github.com/pytorch/pytorch/issues/123911,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: half, module: nn, module: numerical-stability, module: sdpa",scan when time allows,2024-11-27T20:09:58Z
issue,121465,[RFC] PagedAttention Support,https://github.com/pytorch/pytorch/issues/121465,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sdpa,scan when time allows,2024-11-27T20:09:55Z
issue,119131,"`scaled_dot_product_attention` produces NaNs for math backend, but not memory efficient backend when used in mixed precision.",https://github.com/pytorch/pytorch/issues/119131,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: numerical-stability, module: sdpa",scan when time allows,2024-11-27T20:09:55Z
issue,116237,support Tensor type for scale in scaled_dot_product_attention,https://github.com/pytorch/pytorch/issues/116237,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sdpa,scan when time allows,2024-11-27T20:09:53Z
issue,110681,[RFC] Scaled Dot Product Attention API Changes,https://github.com/pytorch/pytorch/issues/110681,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: python frontend, module: sdpa",scan when time allows,2024-11-27T20:09:53Z
issue,103484,No backward implementation for `torch._native_multi_head_attention`,https://github.com/pytorch/pytorch/issues/103484,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-11-27T20:08:28Z
issue,104739,Error reporting uses formal parameter names of downstream C++ function,https://github.com/pytorch/pytorch/issues/104739,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-11-27T20:08:16Z
issue,138420,batch_first support for multi_head_attention_forward,https://github.com/pytorch/pytorch/issues/138420,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: sdpa",scan when time allows,2024-11-27T20:05:15Z
issue,137486,"[feature request] [ux] Introduce special value `SDPBackend.ALL` (or similar functionality) to mean ""all available backends"" for `torch.nn.attention.sdpa_kernel(...)` (some special value `SDPBackend.ERROR` already exists)",https://github.com/pytorch/pytorch/issues/137486,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sdpa,scan when time allows,2024-11-27T20:05:13Z
issue,137192,Setting `nn.TransformerEncoder()` and `nn.TransformerDecoder()` to `nn.Transformer()` doesn't get any errors with `nn.Transformer()`'s arguments having wrong type values,https://github.com/pytorch/pytorch/issues/137192,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-11-27T20:04:53Z
issue,137173,`batch_first` argument of `nn.Transformer()` should be `True` by default because it's not convenient to get a warning with the default settings,https://github.com/pytorch/pytorch/issues/137173,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-11-27T20:04:52Z
issue,131254,Masked Attention has no effect in ``TransformerEncoderLayer``,https://github.com/pytorch/pytorch/issues/131254,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-11-27T20:02:10Z
issue,126973,Performance Degradation in F.linear with Batch Size > 1 in Multi-Head Attention,https://github.com/pytorch/pytorch/issues/126973,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: performance",scan when time allows,2024-11-27T20:02:10Z
issue,123642,TransformerDecoderLayer fails when on cuda and batch dimension is zero (on cpu is fine),https://github.com/pytorch/pytorch/issues/123642,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-11-27T20:02:05Z
issue,122660,Deprecate `torch.nn.MultiHeadAttention` and `torch.nn.Transformer`-related modules,https://github.com/pytorch/pytorch/issues/122660,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-11-27T20:02:05Z
issue,121632,"[BUG] Given boolean tgt_mask, TransformerDecoder produces wrong results with MPS backend",https://github.com/pytorch/pytorch/issues/121632,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: mps, module: nn",scan when time allows,2024-11-27T20:02:03Z
issue,119653,I hope PyTorch will open the sliding window local attention feature as soon as possible.,https://github.com/pytorch/pytorch/issues/119653,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sdpa,scan when time allows,2024-11-27T19:56:30Z
issue,119371,Nan output after masked TransforrmerDecoder,https://github.com/pytorch/pytorch/issues/119371,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-11-27T19:55:44Z
issue,116175,Performance for `TransformerEncoderLayer` and `TransformerDecoderLayer` drops severely if pytorch version changed from `pt1.10` to `pt2.0`,https://github.com/pytorch/pytorch/issues/116175,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: performance",scan when time allows,2024-11-27T19:55:40Z
issue,114796,Add head_mask for transformers,https://github.com/pytorch/pytorch/issues/114796,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-11-27T19:55:36Z
issue,103836,Please consider the SCFA/dynamic flash attention for your implementation of scaled dot product attention,https://github.com/pytorch/pytorch/issues/103836,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: sdpa, module: sparse",scan when time allows,2024-11-27T19:53:57Z
issue,109987,Static quantization for Transformer block : AttributeError 'function' object has no attribute 'is_cuda',https://github.com/pytorch/pytorch/issues/109987,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-11-27T19:52:31Z
issue,108522,nn.Transformer has dropout layers that BERT / GPT-2 do not have,https://github.com/pytorch/pytorch/issues/108522,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: nn",scan when time allows,2024-11-27T19:52:30Z
issue,107663,`RuntimeError: expected scalar type BFloat16 but found Float` with `torch.nn.TransformerEncoder`,https://github.com/pytorch/pytorch/issues/107663,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision), module: nn",scan when time allows,2024-11-27T19:52:28Z
issue,103668,MultiheadAttention should split embed_dim into four parameters,https://github.com/pytorch/pytorch/issues/103668,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-11-27T19:52:25Z
issue,102333,"transformer encoder-layer, the sample-Independent attn_mask(dim=3) has different behaviors when training and validating",https://github.com/pytorch/pytorch/issues/102333,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-11-27T19:52:25Z
issue,99012,"[BUG]Float32 attention mask not working with torch.autocast(""cpu"")",https://github.com/pytorch/pytorch/issues/99012,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-11-27T19:50:43Z
issue,97128,"MultiHeadAttention, fast path broken with `bias=False` or uneven number of heads",https://github.com/pytorch/pytorch/issues/97128,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-11-27T19:50:41Z
issue,97111,TransformerEncoder truncates output when some token positions are masked by `src_key_padding_mask` across batch,https://github.com/pytorch/pytorch/issues/97111,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nestedtensor, module: nn",scan when time allows,2024-11-27T19:50:40Z
issue,90866,Support arbitrary masks for _nested_tensor_from_mask in nn.TransformerEncoder,https://github.com/pytorch/pytorch/issues/90866,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nestedtensor, module: nn",scan when time allows,2024-11-27T19:50:35Z
issue,89394,torch.nn.TransformerEncoderLayer missing exception description information.,https://github.com/pytorch/pytorch/issues/89394,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: error checking, module: nn",scan when time allows,2024-11-27T19:50:34Z
issue,82919,Adding a warning of non-compatibility with forward hooks for the fast path of TransformerEncoderLayer,https://github.com/pytorch/pytorch/issues/82919,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-11-27T19:50:31Z
issue,80494,[bug] libtorch bug in nn::MultiheadAttention and nn::Transformer,https://github.com/pytorch/pytorch/issues/80494,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: cpp, module: nn",scan when time allows,2024-11-27T19:48:44Z
issue,78954,Can we have Additive Attention?,https://github.com/pytorch/pytorch/issues/78954,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2024-11-27T19:48:43Z
issue,78754,Getting NotImplementedError when trying to implement E2E support for `prim::is_nested` Op in torch-mlir.,https://github.com/pytorch/pytorch/issues/78754,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nestedtensor,scan when time allows,2024-11-27T19:48:42Z
issue,74092,Two consecutive nn.LayerNorm are used in transformer model when norm_first is False,https://github.com/pytorch/pytorch/issues/74092,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-11-27T19:48:41Z
issue,69353,Pruning `torch.nn.MultiheadAttention` causes RuntimeError,https://github.com/pytorch/pytorch/issues/69353,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: pruning,scan when time allows,2024-11-27T19:48:39Z
issue,67651,MultiHeadAttention in quantizable seems incorrect with batch_first=True,https://github.com/pytorch/pytorch/issues/67651,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2024-11-27T19:48:38Z
issue,57230,Implementation of Self Attention vs Encoder Decoder Attention Causing Early Evaluation for PyTorch/XLA,https://github.com/pytorch/pytorch/issues/57230,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2024-11-27T19:48:34Z
issue,55228,Requesting a feature to calculate attention weights.,https://github.com/pytorch/pytorch/issues/55228,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2024-11-27T19:48:32Z
issue,50258,More general MultiHeadAttention and Transformer modules,https://github.com/pytorch/pytorch/issues/50258,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2024-11-27T19:48:31Z
issue,49555,Gunicorn preload flag not working with PyTorch library,https://github.com/pytorch/pytorch/issues/49555,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-11-27T19:47:17Z
issue,46031,Custom attention,https://github.com/pytorch/pytorch/issues/46031,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-11-27T19:47:17Z
issue,49791,Caching for autoregressive decoding of Transformer,https://github.com/pytorch/pytorch/issues/49791,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: performance",scan when time allows,2024-11-27T19:47:16Z
issue,43865,"Allow custom kwargs for forward method of nn.TransformerEncoderLayer, nn.TransformerDecoderLayer.",https://github.com/pytorch/pytorch/issues/43865,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-11-27T19:47:15Z
issue,40932,MultiheadAttention set(-inf) cause 'Nan' in loss computing,https://github.com/pytorch/pytorch/issues/40932,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nans and infs, module: nn",scan when time allows,2024-11-27T19:47:13Z
issue,41070,Cannot define v_dim in MultiheadAttention,https://github.com/pytorch/pytorch/issues/41070,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-11-27T19:47:12Z
issue,34760,incomplete implementation doubt,https://github.com/pytorch/pytorch/issues/34760,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-11-27T19:47:11Z
issue,34573,Restructure `multi_head_attention_forward`,https://github.com/pytorch/pytorch/issues/34573,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-11-27T19:47:11Z
issue,34453,[Feature request] Query padding mask for nn.MultiheadAttention,https://github.com/pytorch/pytorch/issues/34453,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-11-27T19:47:08Z
issue,32590,[FYI] MultiheadAttention / Transformer,https://github.com/pytorch/pytorch/issues/32590,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-11-27T19:47:08Z
issue,32529,addition of attention based techniques to pytorch,https://github.com/pytorch/pytorch/issues/32529,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: nn",scan when time allows,2024-11-27T19:47:08Z
issue,31277,nn.MultiHeadAttention with different similarity measures,https://github.com/pytorch/pytorch/issues/31277,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-11-27T19:47:08Z
issue,28884,Why attn_mask is not 3D tensor in nn.MultiheadAttention?,https://github.com/pytorch/pytorch/issues/28884,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-11-27T19:47:05Z
issue,28657,[docs] Unclear input/output format for TransformerEncoderLayer,https://github.com/pytorch/pytorch/issues/28657,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-11-27T19:47:05Z
issue,25132,Feature request: Fix dimension convention for masks in transformer,https://github.com/pytorch/pytorch/issues/25132,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-11-27T19:47:04Z
issue,24930,Successive Layer Normalization in nn.Transformer,https://github.com/pytorch/pytorch/issues/24930,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-11-27T19:47:04Z
issue,21518,`attn_mask` in nn.MultiheadAttention is additive,https://github.com/pytorch/pytorch/issues/21518,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: nn",scan when time allows,2024-11-27T19:45:19Z
issue,21876,nn.TransformerLayer,https://github.com/pytorch/pytorch/issues/21876,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-11-27T19:45:09Z
issue,138399,`dtype` promotion of `out=` functions on meta inputs not consistent.,https://github.com/pytorch/pytorch/issues/138399,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: meta tensors, module: type promotion",scan when time allows,2024-11-27T12:30:01Z
issue,141467,import torch failed after installing nightly wheel,https://github.com/pytorch/pytorch/issues/141467,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: binaries,scan when time allows,2024-11-27T12:05:09Z
issue,130470,nested_tensor jagged_from_list doesn't correctly handle corner case of empty list,https://github.com/pytorch/pytorch/issues/130470,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nestedtensor,scan when time allows,2024-11-27T09:12:53Z
issue,118991,Memory Leak in v2.0.1,https://github.com/pytorch/pytorch/issues/118991,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: memory usage",scan when time allows,2024-11-27T05:46:21Z
issue,130137,How to get stream operators in custom backend compiler ?,https://github.com/pytorch/pytorch/issues/130137,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: oncall: distributed, oncall: pt2",scan when time allows,2024-11-27T05:20:33Z
issue,98668,Cannot use `checkpoint_sequential` with `torch.compile`,https://github.com/pytorch/pytorch/issues/98668,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: activation checkpointing, oncall: distributed, oncall: pt2",scan when time allows,2024-11-27T05:11:45Z
issue,133092,Umbrella issue for Dynamo tracing through ProcessGroup related torch.distributed APIs,https://github.com/pytorch/pytorch/issues/133092,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: oncall: distributed, oncall: pt2",scan when time allows,2024-11-27T05:02:01Z
issue,140416,[DDP + Dynamo] Failed to get compiled graphs,https://github.com/pytorch/pytorch/issues/140416,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: distributed, oncall: pt2",scan when time allows,2024-11-26T23:54:49Z
issue,141422,"What is ""recompilation profiler"" in doc? (Seems to have a dangling link)",https://github.com/pytorch/pytorch/issues/141422,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",scan when time allows,2024-11-26T23:22:21Z
issue,132940,Enabling GPU mixed precision algorithms,https://github.com/pytorch/pytorch/issues/132940,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: linear algebra,scan when time allows,2024-11-26T21:23:39Z
issue,140712,No fake impl or Meta kernel for Communication Operator,https://github.com/pytorch/pytorch/issues/140712,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2024-11-26T13:16:19Z
issue,49291,add inverse cdf for Chi-square Distribution,https://github.com/pytorch/pytorch/issues/49291,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distributions,scan when time allows,2024-11-26T10:59:15Z
issue,141335,Rounding issue of tensor sizes when sharding,https://github.com/pytorch/pytorch/issues/141335,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2024-11-26T09:33:08Z
issue,112443,torch.export emits node outside of Core ATen IR,https://github.com/pytorch/pytorch/issues/112443,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: export,scan when time allows,2024-11-26T02:02:47Z
issue,141507,[AudioLM] Inductor `LoweringException: NotImplementedError: View`,https://github.com/pytorch/pytorch/issues/141507,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",scan when time allows,2024-11-26T00:39:33Z
issue,138217,[PT2] Don't graph break on loss.backward(),https://github.com/pytorch/pytorch/issues/138217,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: compiled autograd, module: dynamo, oncall: pt2",scan when time allows,2024-11-25T22:57:16Z
issue,141343,numpy's division compiled but returning Different results,https://github.com/pytorch/pytorch/issues/141343,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: numpy, oncall: pt2",scan when time allows,2024-11-25T21:26:20Z
issue,141350,Support saving pytorch models in OCI registries,https://github.com/pytorch/pytorch/issues/141350,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: serialization,scan when time allows,2024-11-25T16:55:11Z
issue,141295,Memory efficient reshape for stride-0 tensors,https://github.com/pytorch/pytorch/issues/141295,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python frontend,scan when time allows,2024-11-25T16:53:13Z
issue,62475,Learning rate scheduler list index out of range,https://github.com/pytorch/pytorch/issues/62475,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: lrscheduler, module: optimizer",scan when time allows,2024-11-25T15:36:52Z
issue,141177,`CrossEntropyLoss` much slower with class probabilities vs. class indices target,https://github.com/pytorch/pytorch/issues/141177,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: loss, module: nn",scan when time allows,2024-11-25T15:06:49Z
issue,141449,Segmentation fault (core dumped) in `_fft_r2c`,https://github.com/pytorch/pytorch/issues/141449,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: edge cases, module: error checking",scan when time allows,2024-11-25T01:40:48Z
issue,141450,Segmentation fault (core dumped) in `_fft_c2c`,https://github.com/pytorch/pytorch/issues/141450,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: edge cases, module: error checking",scan when time allows,2024-11-25T01:40:33Z
issue,120003,[FSDP2] Eager-Mode Execution Tracker,https://github.com/pytorch/pytorch/issues/120003,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fsdp,scan when time allows,2024-11-24T19:33:37Z
issue,141203,SEGV in `torch.native_batch_norm`,https://github.com/pytorch/pytorch/issues/141203,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: edge cases, module: error checking, module: sanitizers",scan when time allows,2024-11-22T23:37:36Z
issue,140879,fr_trace usability issues,https://github.com/pytorch/pytorch/issues/140879,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2024-11-22T21:08:34Z
issue,141337,`torch.det` with python 3.10 and torch 2.5.0+cu121 raises RuntimeError,https://github.com/pytorch/pytorch/issues/141337,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: cuda",scan when time allows,2024-11-22T21:04:01Z
issue,140301,[RFC] Distributed manual_seed API for N-D parallelism,https://github.com/pytorch/pytorch/issues/140301,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: c10d, oncall: distributed",scan when time allows,2024-11-22T21:03:43Z
issue,88320,`torch.load()` cannot load data saved at non-zero position in a file (`failed finding central directory`),https://github.com/pytorch/pytorch/issues/88320,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: serialization,scan when time allows,2024-11-22T08:48:57Z
issue,141125,[Dynamo] Dynamo is not aware of the function users override,https://github.com/pytorch/pytorch/issues/141125,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",scan when time allows,2024-11-22T03:13:20Z
issue,141277,Multiple torch.{*}_copy variants are not in generated docs/docs skiplist,https://github.com/pytorch/pytorch/issues/141277,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2024-11-21T20:21:26Z
issue,93743,Support data-dependent control flow operators,https://github.com/pytorch/pytorch/issues/93743,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamo, module: onnx, oncall: pt2",scan when time allows,2024-11-21T19:22:04Z
issue,74235,RuntimeError: CUDA error: unspecified launch failure,https://github.com/pytorch/pytorch/issues/74235,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: cudnn",scan when time allows,2024-11-21T02:09:20Z
issue,140237,Cannot export FeaStConv-based network to onnx via torch.jit.script,https://github.com/pytorch/pytorch/issues/140237,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2024-11-21T01:56:00Z
issue,140787,Apple Silicon Unified Memory usage,https://github.com/pytorch/pytorch/issues/140787,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: mps,scan when time allows,2024-11-21T01:22:46Z
issue,140706,"MPS random number generation is slow, if not hanging forever",https://github.com/pytorch/pytorch/issues/140706,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: mps, module: performance",scan when time allows,2024-11-21T01:20:44Z
issue,98925,Request for deterministic support for reflection_pad2d_backward_cuda,https://github.com/pytorch/pytorch/issues/98925,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: determinism, module: padding",scan when time allows,2024-11-20T21:21:32Z
issue,141144,"Spsolve for sparse, complex matrices",https://github.com/pytorch/pytorch/issues/141144,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2024-11-20T19:06:58Z
issue,140703,[Feature Request] The `.forward()`/etc. shape API,https://github.com/pytorch/pytorch/issues/140703,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-11-19T21:49:44Z
issue,140892,inconsistency in ```torch.special.xlogy``` on CPU and GPU,https://github.com/pytorch/pytorch/issues/140892,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bfloat16, module: numerical-stability",scan when time allows,2024-11-19T14:43:15Z
issue,68009,Properly document the `to.dtype_layout` overload of `Tensor.to`,https://github.com/pytorch/pytorch/issues/68009,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2024-11-19T06:22:18Z
issue,140596,[export][non-strict] Passing in kwargs torch.export fails at non_strict_utils.fakify(),https://github.com/pytorch/pytorch/issues/140596,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, oncall: export, oncall: pt2",scan when time allows,2024-11-18T21:37:55Z
issue,140592,Inductor pattern matcher - missing val,https://github.com/pytorch/pytorch/issues/140592,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",scan when time allows,2024-11-18T20:22:26Z
issue,41485,GLOO infiniband with PyTorch,https://github.com/pytorch/pytorch/issues/41485,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2024-11-18T18:31:56Z
issue,140560,`skipIfXYZ` decorators will skip CPU tests as well,https://github.com/pytorch/pytorch/issues/140560,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: testing,scan when time allows,2024-11-18T18:11:57Z
issue,138957,`DataLoader` probably shouldn't use `fork` by default,https://github.com/pytorch/pytorch/issues/138957,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: data, module: dataloader",scan when time allows,2024-11-18T18:08:21Z
issue,50346,how to save weights when using RPC framework,https://github.com/pytorch/pytorch/issues/50346,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: rpc, oncall: distributed",scan when time allows,2024-11-18T17:04:45Z
issue,139324,[FR] Support sub-group partition ProcessGroup,https://github.com/pytorch/pytorch/issues/139324,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2024-11-17T02:17:44Z
issue,124016,Expand stale management to Issues,https://github.com/pytorch/pytorch/issues/124016,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: devx,scan when time allows,2024-11-16T17:16:34Z
issue,138248,"Adam optimizer without first moment estimate, for less vram",https://github.com/pytorch/pytorch/issues/138248,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2024-11-16T13:36:16Z
issue,140711,"after around 13 iteration, tensor img_syn contains many ""NaN"" value when trained on cifari10, but on minist dataset, it works fine",https://github.com/pytorch/pytorch/issues/140711,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2024-11-16T05:54:47Z
issue,140701,inconsistency in ```torch.special.polygamma``` on CPU and GPU,https://github.com/pytorch/pytorch/issues/140701,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: special,scan when time allows,2024-11-16T00:33:49Z
issue,140796,inconsistency in ```torch.special.xlog1py``` on CPU and GPU,https://github.com/pytorch/pytorch/issues/140796,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: special,scan when time allows,2024-11-16T00:32:32Z
issue,139520,inductor/test_move_constructors_to_cuda.py::TestMoveConstructorsToCuda::test_multi_gpu unit test failure,https://github.com/pytorch/pytorch/issues/139520,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2024-11-15T21:17:45Z
issue,140570,"On the Apple M1 GPU, torch.where() finds tensor entries in the wrong place",https://github.com/pytorch/pytorch/issues/140570,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: 64-bit, module: correctness (silent), module: mps",scan when time allows,2024-11-15T18:55:57Z
issue,140513,"Using @torch.compile, there are slight differences in output when run on different GPUs on the same machine.",https://github.com/pytorch/pytorch/issues/140513,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: determinism, oncall: pt2",scan when time allows,2024-11-15T17:36:52Z
issue,140704,vmap with mps tensor fails to recognized batched shape,https://github.com/pytorch/pytorch/issues/140704,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: functorch, module: mps, module: vmap",scan when time allows,2024-11-15T16:29:42Z
issue,140687,“Can't swap an already initialized allocator” when calling torch.cuda.memory.change_current_allocator,https://github.com/pytorch/pytorch/issues/140687,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: cudacachingallocator",scan when time allows,2024-11-15T15:33:54Z
issue,140778,Floating point exception (core dumped) in `torch.sparse.sampled_addmm`,https://github.com/pytorch/pytorch/issues/140778,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2024-11-15T15:32:51Z
issue,92134,[Bug][Dataloader] unable to mmap 2048 bytes from file <filename not specified>: Cannot allocate memory (12),https://github.com/pytorch/pytorch/issues/92134,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dataloader, module: multiprocessing",scan when time allows,2024-11-15T05:09:14Z
issue,115500,Batchnorm NAN in amp autocast mode.,https://github.com/pytorch/pytorch/issues/115500,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision), module: cuda, module: norms and normalization",scan when time allows,2024-11-14T10:22:41Z
issue,103150,Build fails at linking torch_shm_manager on aarch64,https://github.com/pytorch/pytorch/issues/103150,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2024-11-14T09:09:28Z
issue,127388,"torch=2.1, import torch; fails AttributeError: '_OpNamespace' 'aten' object has no attribute 'sym_constrain_range_for_size'",https://github.com/pytorch/pytorch/issues/127388,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, oncall: quantization",scan when time allows,2024-11-14T08:27:10Z
issue,125134,binaries/dump_operator_names.cc missing iostream include,https://github.com/pytorch/pytorch/issues/125134,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: wsl",scan when time allows,2024-11-14T05:45:36Z
issue,140516,[autograd engine && compiled autograd] support for partial backward compilation,https://github.com/pytorch/pytorch/issues/140516,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: compiled autograd, oncall: pt2",scan when time allows,2024-11-14T03:09:27Z
issue,140602,"torch.from_numpy() returns a cpu tensor, even under `with torch.device(""meta"")`",https://github.com/pytorch/pytorch/issues/140602,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: meta tensors, module: numpy",scan when time allows,2024-11-14T03:04:09Z
issue,112727,[dynamo] Implement iter fallback (and possibly all iters/generators) as polyfill,https://github.com/pytorch/pytorch/issues/112727,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",scan when time allows,2024-11-13T22:42:36Z
issue,109309,Support the `ExitStack` context manager (or a simplified version),https://github.com/pytorch/pytorch/issues/109309,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",scan when time allows,2024-11-13T22:17:35Z
issue,134067,subprocess tensor creation gets stuck,https://github.com/pytorch/pytorch/issues/134067,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: deadlock, module: multiprocessing",scan when time allows,2024-11-13T02:56:55Z
issue,140396,Optional ```clamp_k``` Argument for Handling Out-of-Range Values in ```torch.topk```,https://github.com/pytorch/pytorch/issues/140396,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sorting and selection,scan when time allows,2024-11-13T02:36:49Z
issue,140348,inconsistency in ```torch.special.i0e``` on CPU and GPU,https://github.com/pytorch/pytorch/issues/140348,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bfloat16, module: cuda, module: numerical-stability, module: special",scan when time allows,2024-11-13T01:19:05Z
issue,140017,ghstack-mergebility-check often gives false positives,https://github.com/pytorch/pytorch/issues/140017,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2024-11-12T21:49:31Z
issue,140018,TargetDeterminator skips op_info tests for operator directly changed there,https://github.com/pytorch/pytorch/issues/140018,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2024-11-12T21:47:15Z
issue,73764,Group convolution slower than manually running separate convolutions in CUDA streams,https://github.com/pytorch/pytorch/issues/73764,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: cuda, module: cudnn, module: performance",scan when time allows,2024-11-12T18:27:54Z
issue,140252,inconsistency in ```torch.special.logit``` on CPU and GPU,https://github.com/pytorch/pytorch/issues/140252,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bfloat16, module: cuda, module: special",scan when time allows,2024-11-11T19:54:47Z
issue,140214,torch.svd_lowrank - non reproducible,https://github.com/pytorch/pytorch/issues/140214,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: linear algebra, module: sparse",scan when time allows,2024-11-11T19:47:56Z
issue,116102,torch.func.grad with respect to intermediate outputs,https://github.com/pytorch/pytorch/issues/116102,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: functorch,scan when time allows,2024-11-11T16:12:03Z
issue,140012,Default value for get_num_threads,https://github.com/pytorch/pytorch/issues/140012,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: multithreading,scan when time allows,2024-11-11T01:51:32Z
issue,139962,fbgemm build errors on warning that VLAs are a clang extension when building with clang18,https://github.com/pytorch/pytorch/issues/139962,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: third_party",scan when time allows,2024-11-09T00:01:41Z
issue,140112,inconsistency in ```torch.nn.BatchNorm1d``` on CPU and GPU,https://github.com/pytorch/pytorch/issues/140112,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: numerical-stability",scan when time allows,2024-11-08T17:12:56Z
issue,100741,Reading `state_dict()` in C++,https://github.com/pytorch/pytorch/issues/100741,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: nn, module: serialization",scan when time allows,2024-11-08T16:49:26Z
issue,140090,Add Android Support (Python Wheel),https://github.com/pytorch/pytorch/issues/140090,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: android, module: binaries",scan when time allows,2024-11-08T14:58:44Z
issue,140088,Building PyTorch from source fails when magma.h is in /usr/include,https://github.com/pytorch/pytorch/issues/140088,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: magma",scan when time allows,2024-11-08T13:46:24Z
issue,139958,Can't seem to run tip of tree on Linux,https://github.com/pytorch/pytorch/issues/139958,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2024-11-08T13:38:26Z
issue,139910,Setting a `str` value to `weight` and `pos_weight` argument of `BCEWithLogitsLoss()` gets an indirect error message,https://github.com/pytorch/pytorch/issues/139910,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-11-08T13:32:50Z
issue,38419,[RFC] Add tar-based IterableDataset implementation to PyTorch,https://github.com/pytorch/pytorch/issues/38419,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2024-11-08T04:24:27Z
issue,75577,Cannot use socks5h proxy because of urllib: `urlopen error Remote end closed connection without response`,https://github.com/pytorch/pytorch/issues/75577,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: hub, module: vision",scan when time allows,2024-11-08T02:18:29Z
issue,139876,Allow using context managers in torch.fx,https://github.com/pytorch/pytorch/issues/139876,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: fx, module: python frontend",scan when time allows,2024-11-07T15:57:34Z
issue,139916,Both `input` and `target` argument or at least `target` argument of `BCELoss()` shoud accept the values out of `0 <= x <= 1` so that `BCELoss()` with `torch.sigmoid()` and `nn.Sigmoid()` gets the same results as `BCEWithLogitsLoss()`.,https://github.com/pytorch/pytorch/issues/139916,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: loss, module: nn",scan when time allows,2024-11-07T15:57:16Z
issue,139609,There is a bug in torch.cat for non-standard boolean,https://github.com/pytorch/pytorch/issues/139609,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: boolean tensor, module: edge cases, module: python frontend",scan when time allows,2024-11-07T15:55:15Z
issue,32421,Segmentation fault in C++ API torch::from_blob(...).clone(),https://github.com/pytorch/pytorch/issues/32421,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp,scan when time allows,2024-11-07T12:32:57Z
issue,46027,Padding mode for ConvTransposeNd,https://github.com/pytorch/pytorch/issues/46027,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: convolution,scan when time allows,2024-11-07T03:51:39Z
issue,139874,Modify CMake files to support Clang for Windows when building Libtorch-dependent apps,https://github.com/pytorch/pytorch/issues/139874,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: windows",scan when time allows,2024-11-06T23:47:57Z
issue,139521,boxing-unboxing overhead seems significant,https://github.com/pytorch/pytorch/issues/139521,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dispatch,scan when time allows,2024-11-06T17:03:20Z
issue,139701,`softmax` fails to commute with `permute` / `transpose`,https://github.com/pytorch/pytorch/issues/139701,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-11-06T16:28:46Z
issue,139472,_refs.div.floor_rounding returns NaN instead of +- inf when a divide by 0 occurs,https://github.com/pytorch/pytorch/issues/139472,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpu,scan when time allows,2024-11-06T16:28:24Z
issue,65936,[POLL][RFC] DataParallel Deprecation,https://github.com/pytorch/pytorch/issues/65936,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: deprecation,scan when time allows,2024-11-06T12:56:48Z
issue,59786,Support negative step sizes for slicing,https://github.com/pytorch/pytorch/issues/59786,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python array api,scan when time allows,2024-11-06T02:09:55Z
issue,132386,torch cpu float16 range is not aligned with scipy on polygamma.,https://github.com/pytorch/pytorch/issues/132386,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: edge cases, module: half, module: special",scan when time allows,2024-11-06T02:08:22Z
issue,113456,Expose FakeTensor and FakeTensorMode as public APIs,https://github.com/pytorch/pytorch/issues/113456,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: faketensor, module: onnx",scan when time allows,2024-11-06T00:16:01Z
issue,139795,Support scalar tensor in `value` arg of foreach op,https://github.com/pytorch/pytorch/issues/139795,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: mta, module: optimizer",scan when time allows,2024-11-05T22:34:00Z
issue,122094,Investigate torch.compile Windows support.,https://github.com/pytorch/pytorch/issues/122094,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, module: windows, oncall: pt2",scan when time allows,2024-11-05T18:35:24Z
issue,139648,[Tracker] complex128 tests are disabled for all foreach internally,https://github.com/pytorch/pytorch/issues/139648,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: mta, module: tests",scan when time allows,2024-11-04T18:47:38Z
issue,117348,The bitwise_XX.Scalar_Tensor should be dispatched to core IR but is currently not,https://github.com/pytorch/pytorch/issues/117348,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: decompositions,scan when time allows,2024-11-04T06:47:38Z
issue,138773,[funcol] functional collectives are 67% slower than torch.distributed collectives,https://github.com/pytorch/pytorch/issues/138773,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: performance, oncall: distributed",scan when time allows,2024-11-03T02:06:37Z
issue,139304,"[RFC][Pipelining] RNG state communication, avoid seed checkpoint",https://github.com/pytorch/pytorch/issues/139304,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2024-11-01T21:44:49Z
issue,139333,torch.nn.ReflectionPad2d throws CUDA error on large tensors,https://github.com/pytorch/pytorch/issues/139333,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: nn",scan when time allows,2024-11-01T18:07:52Z
issue,139342,torch.logcumsumexp runtime error inconsistency on CPU and CUDA (also documentation missing constraint),https://github.com/pytorch/pytorch/issues/139342,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: python frontend",scan when time allows,2024-11-01T18:07:29Z
issue,139245,ncclInternalError: Internal check failed,https://github.com/pytorch/pytorch/issues/139245,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nccl, oncall: distributed",scan when time allows,2024-11-01T18:05:53Z
issue,139241,Please implement batching rule for torch.nn.functional.multi_margin_loss,https://github.com/pytorch/pytorch/issues/139241,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: functorch, module: vmap",scan when time allows,2024-11-01T13:24:25Z
issue,139178,"the mpi backend not support to transfer float16/bfloat16 tensor and would post ""IndexError: map::at"" when transferring float16/bfloat16 tensor",https://github.com/pytorch/pytorch/issues/139178,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: mpi,scan when time allows,2024-11-01T09:25:58Z
issue,139382,"RuntimeError: free_upper_bound + pytorch_used_bytes[device] <= device_total INTERNAL ASSERT FAILED at ""../c10/cuda/CUDAMallocAsyncAllocator.cpp"":527",https://github.com/pytorch/pytorch/issues/139382,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2024-10-31T21:47:41Z
issue,139200,find_spec does something weird to Python Path when loading modules,https://github.com/pytorch/pytorch/issues/139200,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python frontend,scan when time allows,2024-10-31T21:46:00Z
issue,54137,Add additive angular margin loss,https://github.com/pytorch/pytorch/issues/54137,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: loss,scan when time allows,2024-10-31T19:15:39Z
issue,68291,"INTERNAL ASSERT FAILED at ""/opt/conda/conda-bld/pytorch_1634272068185/work/aten/src/ATen/native/LinearAlgebraUtils.h"":244",https://github.com/pytorch/pytorch/issues/68291,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: linear algebra,scan when time allows,2024-10-31T19:07:17Z
issue,120735,"`torch.optim.lr_scheduler.{LRScheduler,*}`: lack of documentation",https://github.com/pytorch/pytorch/issues/120735,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: lrscheduler",scan when time allows,2024-10-31T16:02:21Z
issue,137368,Fused AdamW causing Illegal memory access on H100 for large models,https://github.com/pytorch/pytorch/issues/137368,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: 64-bit, module: cuda, module: optimizer",scan when time allows,2024-10-31T15:48:33Z
issue,127736,torch.nn.BatchNorm1d missing check on values of parameters eps and momentum,https://github.com/pytorch/pytorch/issues/127736,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-10-31T01:50:07Z
issue,139252,Ban relative imports in test/,https://github.com/pytorch/pytorch/issues/139252,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: testing, module: tests",scan when time allows,2024-10-30T17:16:40Z
issue,139269,TensorBoard images loading error,https://github.com/pytorch/pytorch/issues/139269,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: tensorboard, module: windows",scan when time allows,2024-10-30T16:46:29Z
issue,61173,Increased memory usage with AMP,https://github.com/pytorch/pytorch/issues/61173,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision), module: cuda",scan when time allows,2024-10-30T06:37:34Z
issue,139185,Operators being traced as method calls in torch.fx,https://github.com/pytorch/pytorch/issues/139185,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2024-10-29T21:31:49Z
issue,139170,`pkg_resources` module is deprecated by setuptools,https://github.com/pytorch/pytorch/issues/139170,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2024-10-29T21:30:51Z
issue,139163,SubgraphMatcher may fail to match when the matching pattern having call_module IR,https://github.com/pytorch/pytorch/issues/139163,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2024-10-29T21:29:49Z
issue,138042,`Calculate Docker Image` step times out if run on LF runners,https://github.com/pytorch/pytorch/issues/138042,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2024-10-29T20:48:02Z
issue,139188,PCH build fail with sccache-v0.8.2,https://github.com/pytorch/pytorch/issues/139188,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ci, module: third_party",scan when time allows,2024-10-29T19:49:38Z
issue,106455,"vmap, jacrev, jacfwd, hessian, etc., in libTorch",https://github.com/pytorch/pytorch/issues/106455,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: functorch, module: vmap",scan when time allows,2024-10-29T17:19:41Z
issue,123628,`torch.onnx.export` doesn't correctly constfold constants and DequantLinear,https://github.com/pytorch/pytorch/issues/123628,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2024-10-29T16:47:49Z
issue,139201,First run lint on just the changes in the PR before running it over the entire PR,https://github.com/pytorch/pytorch/issues/139201,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2024-10-29T16:45:44Z
issue,139115,torch.nn.AvgPool2d works with long on cpu but not gpu,https://github.com/pytorch/pytorch/issues/139115,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: pooling",scan when time allows,2024-10-29T12:09:46Z
issue,139066,"Covert Conv3d to Conv2d, But output different",https://github.com/pytorch/pytorch/issues/139066,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: convolution,scan when time allows,2024-10-29T06:49:56Z
issue,138829,torch.dot has inconsistent support for int64 (long),https://github.com/pytorch/pytorch/issues/138829,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2024-10-29T06:38:41Z
issue,138981,The error compiling problem of Qt and libtorch(my solution to eliminate the bug),https://github.com/pytorch/pytorch/issues/138981,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: dependency bug",scan when time allows,2024-10-29T02:25:20Z
issue,138811,"same data all reduce on H20, but results are different",https://github.com/pytorch/pytorch/issues/138811,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2024-10-29T02:07:39Z
issue,138743,Cleanup the scaling logic in runtime.triton_heuristics.triton_config,https://github.com/pytorch/pytorch/issues/138743,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",scan when time allows,2024-10-28T18:01:24Z
issue,134399,Build errors while building PyTorch with BLIS,https://github.com/pytorch/pytorch/issues/134399,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2024-10-28T05:03:51Z
issue,122089,Infinite cycle on BSC tensor: base -> values -> base,https://github.com/pytorch/pytorch/issues/122089,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2024-10-28T02:21:09Z
issue,136216,[torch.export] Detect internal constrains,https://github.com/pytorch/pytorch/issues/136216,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: oncall: export, oncall: pt2",scan when time allows,2024-10-27T16:09:45Z
issue,8104,load_state_dict unexpectedly does not load Tensor to buffers that currently have None value,https://github.com/pytorch/pytorch/issues/8104,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bc-breaking, module: docs, module: nn",scan when time allows,2024-10-27T10:55:30Z
issue,138940,Libtorch torch::Tensor default constructor does not initialize on CPU like in python,https://github.com/pytorch/pytorch/issues/138940,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp,scan when time allows,2024-10-25T23:47:13Z
issue,138898,cpu faster than mps for both GRUCell and LSTMCell on apple silicon M3 max: MPS Metal,https://github.com/pytorch/pytorch/issues/138898,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: mps, module: performance",scan when time allows,2024-10-25T13:56:08Z
issue,131312,"Pytorch Error: ImportError: /home/dp/anaconda3/envs/ComfyUI/lib/python3.11/site-packages/torch/lib/../../nvidia/cusparse/lib/libcusparse.so.12: undefined symbol: __nvJitLinkComplete_12_4, version libnvJitLink.so.12",https://github.com/pytorch/pytorch/issues/131312,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: cuda",scan when time allows,2024-10-25T07:01:49Z
issue,73131,Support writing tensorboard traces to AWS S3 (and other cloud storage services) in profiler,https://github.com/pytorch/pytorch/issues/73131,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tensorboard,scan when time allows,2024-10-25T06:15:11Z
issue,116242,"Building project using libtorch results in ""Failed to find nvToolsExt""",https://github.com/pytorch/pytorch/issues/116242,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2024-10-24T22:31:25Z
issue,45910,with torch.cuda.amp.autocast() get out of memory error when using with torch.no_grad() during validation,https://github.com/pytorch/pytorch/issues/45910,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision), module: autograd, module: docs",scan when time allows,2024-10-24T13:22:44Z
issue,35527,`torch.bernoulli` and `torch.rand` to support `dtype=torch.bool` kwarg (and sypport operation of `tensor.uniform_` on bool tensors),https://github.com/pytorch/pytorch/issues/35527,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distributions,scan when time allows,2024-10-24T12:59:01Z
issue,138564,[CI] Continuous spam of PyTorch CI build status notifications in fork,https://github.com/pytorch/pytorch/issues/138564,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2024-10-24T09:34:07Z
issue,138673,dead code in test that can be removed,https://github.com/pytorch/pytorch/issues/138673,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tests,scan when time allows,2024-10-23T20:22:05Z
issue,138726,[PP] Add unit tests to check for memory regressions,https://github.com/pytorch/pytorch/issues/138726,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: pipelining,scan when time allows,2024-10-23T18:46:17Z
issue,137387,Simple explanation should be added to `capturable` parameter of `optim.RMSprop()`,https://github.com/pytorch/pytorch/issues/137387,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2024-10-23T18:00:49Z
issue,138468,os.environ modification is not thread safe,https://github.com/pytorch/pytorch/issues/138468,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: multithreading,scan when time allows,2024-10-23T08:36:34Z
issue,13130,"arm64 port for PyTorch, libtorch",https://github.com/pytorch/pytorch/issues/13130,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: arm, module: ci",scan when time allows,2024-10-23T06:50:46Z
issue,138570,`torch.std_mean` returns `NaN` as mean of an `inf` array.,https://github.com/pytorch/pytorch/issues/138570,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nans and infs, module: python frontend",scan when time allows,2024-10-23T05:43:35Z
issue,125862,Pytorch nightly docker image invalidated layers,https://github.com/pytorch/pytorch/issues/125862,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: docker",scan when time allows,2024-10-22T17:33:24Z
issue,138192,"Inconsistent behavior of `torch.tensor` in converting `NaN` to int32 on normal list input and numpy array, and different architectures.",https://github.com/pytorch/pytorch/issues/138192,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: error checking,scan when time allows,2024-10-22T15:55:19Z
issue,78262,Request for adding the possibility for training on sparse tensors,https://github.com/pytorch/pytorch/issues/78262,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2024-10-22T15:45:02Z
issue,137974,Can FlightRecorder set the stream key of collective and P2P OP into the entry.,https://github.com/pytorch/pytorch/issues/137974,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2024-10-22T04:18:08Z
issue,133877,Torch Distributed Elastic has no way to address non-retriable errors,https://github.com/pytorch/pytorch/issues/133877,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2024-10-21T20:39:04Z
issue,138396,`out=` meta device support.,https://github.com/pytorch/pytorch/issues/138396,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: meta tensors,scan when time allows,2024-10-21T20:24:34Z
issue,138197,Compile error with -WERROR on clang-12,https://github.com/pytorch/pytorch/issues/138197,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2024-10-21T18:05:54Z
issue,138320,TeLU activation function,https://github.com/pytorch/pytorch/issues/138320,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-10-21T18:02:29Z
issue,138453,`torch.__config__.show()` silently initialises CUDA (?); forked processes fail with uninitialised CUDA,https://github.com/pytorch/pytorch/issues/138453,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: multiprocessing",scan when time allows,2024-10-21T15:24:44Z
issue,138418,"Internal assert triggered at ivalue_inl.h:1966, Expected IntList but got Int",https://github.com/pytorch/pytorch/issues/138418,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: error checking, module: onnx, oncall: jit",scan when time allows,2024-10-21T15:20:05Z
issue,138387,torch::distributions module for C++ frontend API,https://github.com/pytorch/pytorch/issues/138387,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp,scan when time allows,2024-10-21T15:09:38Z
issue,69858,Feature Request: Bayesian Personalized Ranking Loss,https://github.com/pytorch/pytorch/issues/69858,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: loss, module: nn",scan when time allows,2024-10-21T07:21:03Z
issue,138299,"Cache enablement should be tested inside of cache load() function, not at call site",https://github.com/pytorch/pytorch/issues/138299,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: pt2,scan when time allows,2024-10-19T00:24:47Z
issue,138353,Empty Dimensions returned in `aten::_trilinear`,https://github.com/pytorch/pytorch/issues/138353,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: linear algebra,scan when time allows,2024-10-18T20:09:26Z
issue,138337,quantile does not work with float16: throws RuntimeError: quantile() input tensor must be either float or double dtype,https://github.com/pytorch/pytorch/issues/138337,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: cuda, module: half",scan when time allows,2024-10-18T20:02:44Z
issue,122701,Unable to call len on nested tensor,https://github.com/pytorch/pytorch/issues/122701,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nestedtensor,scan when time allows,2024-10-18T18:20:44Z
issue,138180,NestedTensor doesn't support elementwise operations?,https://github.com/pytorch/pytorch/issues/138180,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nestedtensor,scan when time allows,2024-10-18T16:43:38Z
issue,95108,`torch.nn.LazyLinear` crash when using torch.bfloat16 dtype in pytorch 1.12.0 and 1.13.0,https://github.com/pytorch/pytorch/issues/95108,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-10-18T08:06:58Z
issue,138199,Faster dropout (aten.bernoulli) on cpu,https://github.com/pytorch/pytorch/issues/138199,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: mps, module: performance",scan when time allows,2024-10-18T02:23:11Z
issue,117546,interpolate::trilinear has different result between 1.8.0 and the newer version,https://github.com/pytorch/pytorch/issues/117546,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-10-18T02:01:18Z
issue,138168,nan in sparse mv result when I use cpu,https://github.com/pytorch/pytorch/issues/138168,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2024-10-18T00:51:21Z
issue,111003,Dynamo inlining should compile partial subgraphs / improve graph break with recursive calls,https://github.com/pytorch/pytorch/issues/111003,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",scan when time allows,2024-10-17T23:50:44Z
issue,78484,torch.lerp: discrepancy between CUDA and CPU (with extremal inputs),https://github.com/pytorch/pytorch/issues/78484,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nans and infs,scan when time allows,2024-10-17T10:42:27Z
issue,137969,Flight recorder cannot mark the state of the batch P2P OP entry as completed.,https://github.com/pytorch/pytorch/issues/137969,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2024-10-17T03:16:32Z
issue,137482,"`torch.ops.aten.divide_` leads to ""RuntimeError: result type Float can't be cast to the desired output type Int""",https://github.com/pytorch/pytorch/issues/137482,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: core aten,scan when time allows,2024-10-17T02:13:10Z
issue,136918,"ONNX export: w=tensor.shape() in torch.arange(end=w...) causes ""tensor does not have a device"" error",https://github.com/pytorch/pytorch/issues/136918,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2024-10-16T21:55:45Z
issue,136681,[triton_op] Automatically `tl.constexpr` user-written kernel params when they are static integers,https://github.com/pytorch/pytorch/issues/136681,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamic shapes, module: inductor, module: user triton, oncall: pt2",scan when time allows,2024-10-16T21:41:42Z
issue,137951,Better document that non-blocking GPU -> CPU memory requires device sync,https://github.com/pytorch/pytorch/issues/137951,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: cuda, module: docs",scan when time allows,2024-10-16T20:43:25Z
issue,136736,torch.unravel_index does not check out-of-bounds,https://github.com/pytorch/pytorch/issues/136736,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: advanced indexing, module: numpy",scan when time allows,2024-10-16T17:57:54Z
issue,137167,`empty_like` with NJT on a different device fails because the offsets are not cast appropriately,https://github.com/pytorch/pytorch/issues/137167,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nestedtensor, module: tensor creation",scan when time allows,2024-10-16T17:32:16Z
issue,138017,`_disable_byteorder_record` parameter should be added to the doc of `save()`,https://github.com/pytorch/pytorch/issues/138017,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: serialization",scan when time allows,2024-10-16T13:27:14Z
issue,137872,[Perf]The delay between inference request impact the performance,https://github.com/pytorch/pytorch/issues/137872,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: performance,scan when time allows,2024-10-16T13:26:30Z
issue,137968,inconsistency in ```torch.special.entr``` on CPU and GPU,https://github.com/pytorch/pytorch/issues/137968,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: special",scan when time allows,2024-10-16T07:41:58Z
issue,130644,Add aten::_weight_norm_interface as an ONNX supported TorchScript operator,https://github.com/pytorch/pytorch/issues/130644,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2024-10-15T23:43:57Z
issue,130539,"RuntimeException: [ONNXRuntimeError] : 6 : RUNTIME_EXCEPTION : Non-zero status code returned while running Reshape node. Name:'/Reshape_5' Status Message: /onnxruntime_src/onnxruntime/core/providers/cpu/tensor/reshape_helper.h:45 onnxruntime::ReshapeHelper::ReshapeHelper(const onnxruntime::TensorShape&, onnxruntime::TensorShapeVector&, bool) input_shape_size == size was false. The input tensor cannot be reshaped to the requested shape. Input shape:{1,32,512}, requested shape:{1,1,8,64}",https://github.com/pytorch/pytorch/issues/130539,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2024-10-15T23:39:37Z
issue,130366,Deadcode emitted when using `jit.script` with `onnx.export`.,https://github.com/pytorch/pytorch/issues/130366,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2024-10-15T23:34:59Z
issue,137931,TD skips relevant tests,https://github.com/pytorch/pytorch/issues/137931,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ci, module: correctness (silent)",scan when time allows,2024-10-15T20:47:53Z
issue,86128,[TorchDispatch] Scalar Only Inputs Gets Matched To Tensor Schema,https://github.com/pytorch/pytorch/issues/86128,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: __torch_dispatch__,scan when time allows,2024-10-15T14:21:34Z
issue,137882,Take device of input tensor into consideration for torch::data::transforms::Normalize and possibly other transformations,https://github.com/pytorch/pytorch/issues/137882,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2024-10-15T13:51:51Z
issue,133665,[PixelShuffle] PixelShuffle doesn't support channels_last feature for 5D inputs,https://github.com/pytorch/pytorch/issues/133665,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: memory format, module: nn",scan when time allows,2024-10-15T05:45:34Z
issue,137941,Bazel builds intermittently fail while trying to download mkl from anaconda %),https://github.com/pytorch/pytorch/issues/137941,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2024-10-14T23:48:16Z
issue,137811,Mechanism to attempt to create a VariableTracker but graph break and turn Dynamo off if we fail,https://github.com/pytorch/pytorch/issues/137811,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",scan when time allows,2024-10-14T18:13:58Z
issue,132579,Enable F16C instructions in CUTLASS? At least for the inductor backend?,https://github.com/pytorch/pytorch/issues/132579,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: cuda, oncall: pt2",scan when time allows,2024-10-14T18:01:26Z
issue,68114,CPU Memory Deallocation,https://github.com/pytorch/pytorch/issues/68114,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: memory usage,scan when time allows,2024-10-14T14:22:00Z
issue,124454,torch.utils.cpp_extension.load recompiling every time,https://github.com/pytorch/pytorch/issues/124454,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp-extensions,scan when time allows,2024-10-14T14:03:32Z
issue,117357,[RFC] CI test stats,https://github.com/pytorch/pytorch/issues/117357,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2024-10-13T01:58:23Z
issue,135409,DataLoader + IterableDataset held up by slowest worker!?,https://github.com/pytorch/pytorch/issues/135409,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2024-10-12T12:53:37Z
issue,51075,RFC: Private CUDA memory pools,https://github.com/pytorch/pytorch/issues/51075,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: cuda graphs",scan when time allows,2024-10-12T09:58:19Z
issue,101798,Fix absolute links in pytorch repository and allow it to be proxied,https://github.com/pytorch/pytorch/issues/101798,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: releng,scan when time allows,2024-10-12T06:38:12Z
issue,137777,Using torch from multiple processes is slow,https://github.com/pytorch/pytorch/issues/137777,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: multiprocessing,scan when time allows,2024-10-11T19:47:33Z
issue,46642,Loss functions for complex tensors,https://github.com/pytorch/pytorch/issues/46642,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: complex, module: nn",scan when time allows,2024-10-11T18:51:03Z
issue,137373,"`dampening`, `maximize`, `foreach`, `differentiable` and `fused` parameter of `optim.SGD()` work with `str` values",https://github.com/pytorch/pytorch/issues/137373,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2024-10-11T18:23:47Z
issue,137367,"`lr`, `momentum`, `weight_decay` and `dampening` parameter of `optim.SGD()` work with `bool` values",https://github.com/pytorch/pytorch/issues/137367,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2024-10-11T18:18:05Z
issue,39395,Morphological operations,https://github.com/pytorch/pytorch/issues/39395,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2024-10-11T18:00:13Z
issue,68861,[In Gunicorn & multiprocessing environment] Cannot re-initialize CUDA in forked subprocess,https://github.com/pytorch/pytorch/issues/68861,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: multiprocessing,scan when time allows,2024-10-11T10:44:05Z
issue,137542,Sparse BSR matrix returns `False` is `.is_sparse()`,https://github.com/pytorch/pytorch/issues/137542,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2024-10-11T09:35:15Z
issue,134978,[RFC] A device-agnostic Python device memory related API design for stream-based accelerators,https://github.com/pytorch/pytorch/issues/134978,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python frontend,scan when time allows,2024-10-11T05:32:36Z
issue,89688,Enable NCCL for PyTorch on Windows,https://github.com/pytorch/pytorch/issues/89688,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: windows, oncall: distributed",scan when time allows,2024-10-11T04:18:02Z
issue,39757,max_unpool2d and max_unpool3d cpp signature should be similar,https://github.com/pytorch/pytorch/issues/39757,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: pooling,scan when time allows,2024-10-10T17:17:14Z
issue,126394,Building libtorch for android fails,https://github.com/pytorch/pytorch/issues/126394,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: vulkan, oncall: mobile",scan when time allows,2024-10-10T17:08:06Z
issue,137370,[CD] Implement a fix for: add +PTX to CUDA 12.4 nightly binaries,https://github.com/pytorch/pytorch/issues/137370,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: releng,scan when time allows,2024-10-10T14:37:16Z
issue,136691,"Segfaulting/aborting unit tests do not show up in ""Show Additional Test Info"" section",https://github.com/pytorch/pytorch/issues/136691,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: releng,scan when time allows,2024-10-10T14:37:15Z
issue,129176,extending forward-mode AD docs should really point to an example,https://github.com/pytorch/pytorch/issues/129176,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: forward ad,scan when time allows,2024-10-09T20:52:47Z
issue,137610,[RFC] Make Flight Recorder device generic,https://github.com/pytorch/pytorch/issues/137610,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2024-10-09T18:31:29Z
issue,136994,Add Mcdropout layer,https://github.com/pytorch/pytorch/issues/136994,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-10-09T16:15:38Z
issue,137012,[RFC] Use NCCL topology detection for Inductor communication time estimation,https://github.com/pytorch/pytorch/issues/137012,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, module: nccl, oncall: distributed, oncall: pt2",scan when time allows,2024-10-09T15:52:55Z
issue,114565,Create a reproducible build for LibTorch x64 on VS2022,https://github.com/pytorch/pytorch/issues/114565,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: windows,scan when time allows,2024-10-09T15:48:42Z
issue,137564,Implement Incremental PCA with GPU support,https://github.com/pytorch/pytorch/issues/137564,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2024-10-09T13:33:36Z
issue,120897,[memory profiling] about supporting the C++ Call Stack on the ARM Architecture,https://github.com/pytorch/pytorch/issues/120897,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: arm, module: cuda, oncall: profiler",scan when time allows,2024-10-09T09:58:00Z
issue,137464,Add support for using NestedTensors with linalg operations,https://github.com/pytorch/pytorch/issues/137464,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: linear algebra, module: nestedtensor",scan when time allows,2024-10-09T08:53:07Z
issue,134388,Remove redundant copy in functional collectives,https://github.com/pytorch/pytorch/issues/134388,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: performance, oncall: distributed",scan when time allows,2024-10-09T06:57:23Z
issue,133823,Issues compiling `torch` with `mkl`,https://github.com/pytorch/pytorch/issues/133823,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: intel, module: mkl",scan when time allows,2024-10-09T02:00:22Z
issue,137086,conv2d boundaries,https://github.com/pytorch/pytorch/issues/137086,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-10-08T19:57:35Z
issue,137419,Convert LSTM quantized by QAT to onnx,https://github.com/pytorch/pytorch/issues/137419,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2024-10-08T12:14:10Z
issue,126032,all_gather with gloo backend does not work in inference mode,https://github.com/pytorch/pytorch/issues/126032,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: c10d, oncall: distributed",scan when time allows,2024-10-08T10:16:10Z
issue,136578,Bool convolutions (and other integral types),https://github.com/pytorch/pytorch/issues/136578,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: boolean tensor, module: nn",scan when time allows,2024-10-07T20:59:02Z
issue,87514,invalid_arguments.cpp is busted,https://github.com/pytorch/pytorch/issues/87514,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, module: python frontend",scan when time allows,2024-10-07T19:33:08Z
issue,137412,Public API comparison between v2.4.1 and v2.5 release branch,https://github.com/pytorch/pytorch/issues/137412,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: python frontend",scan when time allows,2024-10-07T17:10:37Z
issue,137409,Unable to save model after removing weight normalization due to remaining hooks,https://github.com/pytorch/pytorch/issues/137409,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: serialization,scan when time allows,2024-10-07T16:37:28Z
issue,116416,"error: cast from 'PyObject *(*)(THPCppFunction *, void *)' (aka '_object *(*)(torch::autograd::THPCppFunction *, void *)') to 'getter' (aka '_object *(*)(_object *, void *)') converts to incompatible function type [-Werror,-Wcast-function-type-strict]",https://github.com/pytorch/pytorch/issues/116416,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2024-10-07T13:54:42Z
issue,110285,TypeError: Got unsupported ScalarType BFloat16,https://github.com/pytorch/pytorch/issues/110285,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: numpy,scan when time allows,2024-10-07T09:23:02Z
issue,55879,Vulkan backend on desktop platforms,https://github.com/pytorch/pytorch/issues/55879,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: backend, module: vulkan",scan when time allows,2024-10-05T07:17:34Z
issue,137172,`torch._batch_norm_impl_index` performs inconsistently on CPU and CUDA,https://github.com/pytorch/pytorch/issues/137172,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: norms and normalization,scan when time allows,2024-10-04T21:37:22Z
issue,137336,NJTs with length doesn't support H2D transfer,https://github.com/pytorch/pytorch/issues/137336,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nestedtensor,scan when time allows,2024-10-04T21:23:38Z
issue,137261,list index out of range in cpp_extension.py,https://github.com/pytorch/pytorch/issues/137261,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2024-10-04T21:20:10Z
issue,137174,`torch.vdot` performs inconsistently on CPU and CUDA,https://github.com/pytorch/pytorch/issues/137174,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2024-10-04T21:11:46Z
issue,137245,[RFC] Retiring ProcessGroupGloo's CUDA support,https://github.com/pytorch/pytorch/issues/137245,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2024-10-04T18:43:13Z
issue,137006,[RFC] Flight Precheck: ahead-of-hang collective mismatch detection,https://github.com/pytorch/pytorch/issues/137006,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2024-10-04T18:28:02Z
issue,134396,torch.sparse.softmax missing support for sparse CSR,https://github.com/pytorch/pytorch/issues/134396,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2024-10-04T15:10:24Z
issue,116182,[ONNX] GFPGANv1.pth to onnx conversion error,https://github.com/pytorch/pytorch/issues/116182,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2024-10-04T01:25:14Z
issue,135048,Setting a `bool` tensor both to `min` and `max` argument of `clamp()` gets error while setting a `bool` tensor only to `min` or `max` argument of `clamp()` works,https://github.com/pytorch/pytorch/issues/135048,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: edge cases, module: python frontend",scan when time allows,2024-10-03T16:50:43Z
issue,137089,Select parameters and buffers by pattern (using fnmatch or re),https://github.com/pytorch/pytorch/issues/137089,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: optimizer",scan when time allows,2024-10-03T03:13:34Z
issue,135927,[Discuss] Enable Windows inductor UTs and fix its timeout issue.,https://github.com/pytorch/pytorch/issues/135927,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, module: windows, oncall: pt2",scan when time allows,2024-10-02T16:42:33Z
issue,112872,libtorch exports miniz symbols,https://github.com/pytorch/pytorch/issues/112872,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: cpp",scan when time allows,2024-10-02T11:02:27Z
issue,136647,torch.compile HUD dashboard should have repro commands,https://github.com/pytorch/pytorch/issues/136647,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2024-10-01T20:36:22Z
issue,137045,Namespace error when using torch::unravel_index to implement a custom function in cuda.,https://github.com/pytorch/pytorch/issues/137045,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: numpy,scan when time allows,2024-10-01T17:26:15Z
issue,107298,dist.destroy_process_group did not destroy the process group well,https://github.com/pytorch/pytorch/issues/107298,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2024-10-01T12:48:50Z
issue,136733,"Be smart about autograd formulas saving either the input or output, depending on context",https://github.com/pytorch/pytorch/issues/136733,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, module: performance, oncall: pt2",scan when time allows,2024-09-30T20:01:12Z
issue,136913,[BUG] torch.tril return ZERO Tensor when tensor is large and in CUDA,https://github.com/pytorch/pytorch/issues/136913,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: linear algebra,scan when time allows,2024-09-30T19:56:09Z
issue,136926,Python implementation of RNN in docstring is broken,https://github.com/pytorch/pytorch/issues/136926,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: rnn",scan when time allows,2024-09-30T19:54:50Z
issue,136937,Setting the boolean value `True` to `hidden_size` argument of `nn.RNN()` gets a indirect error message,https://github.com/pytorch/pytorch/issues/136937,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: rnn",scan when time allows,2024-09-30T19:54:29Z
issue,136938,Setting wrong values to `num_layers` of `nn.RNN()` gets indirect error messages,https://github.com/pytorch/pytorch/issues/136938,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: rnn",scan when time allows,2024-09-30T19:54:07Z
issue,136955,Setting a wrong value to `dropout` argument of `nn.RNN()` gets an indirect error message,https://github.com/pytorch/pytorch/issues/136955,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: rnn",scan when time allows,2024-09-30T19:53:29Z
issue,109401,Interleaved isend and irecv causes hang,https://github.com/pytorch/pytorch/issues/109401,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2024-09-30T12:41:09Z
issue,135542,torch.nn.InstanceNorm2d and torch.nn.InstanceNorm3d returns nan with tensors of float16 dtype on cpu,https://github.com/pytorch/pytorch/issues/135542,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: nn",scan when time allows,2024-09-30T06:25:07Z
issue,136118,[RFC] Default torch.compile backend customization,https://github.com/pytorch/pytorch/issues/136118,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",scan when time allows,2024-09-28T21:39:58Z
issue,91630,Periodic ROCM distribtued jobs are broken,https://github.com/pytorch/pytorch/issues/91630,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rocm,scan when time allows,2024-09-27T15:59:47Z
issue,47996,GPU Vendor-Agnosticism via Vulkan,https://github.com/pytorch/pytorch/issues/47996,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: vulkan",scan when time allows,2024-09-27T15:39:04Z
issue,136747,Provide `gather_mm` functionality and/or expand nested tensor support,https://github.com/pytorch/pytorch/issues/136747,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nestedtensor,scan when time allows,2024-09-27T08:25:26Z
issue,136680,torch.nn.InstanceNorm3d producing inconsistent output for float16 tensors on CPU and GPU,https://github.com/pytorch/pytorch/issues/136680,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: half, module: nn",scan when time allows,2024-09-27T02:29:53Z
issue,136815,Aborted (core dumped) in `torch.cuda.caching_allocator_delete`,https://github.com/pytorch/pytorch/issues/136815,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: cudacachingallocator",scan when time allows,2024-09-27T02:16:57Z
issue,136699,false INTERNAL ASSERT FAILED in `torch._add_batch_dim`,https://github.com/pytorch/pytorch/issues/136699,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: edge cases,scan when time allows,2024-09-27T02:16:17Z
issue,136812,"`CUDNN_BACKEND_OPERATION: cudnnFinalize Failed cudnn_status: CUDNN_STATUS_BAD_PARAM` when using `float16`, works for `bfloat16` / `float32`",https://github.com/pytorch/pytorch/issues/136812,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: nn",scan when time allows,2024-09-27T02:15:38Z
issue,136717,Floating point exception (core dumped) in `torch.ao.nn.intrinsic.quantized.ConvReLU1d/ConvReLU2d/ConvReLU3d` when stride=0,https://github.com/pytorch/pytorch/issues/136717,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: edge cases,scan when time allows,2024-09-27T01:57:40Z
issue,136083,ONNX Export Fails with Dynamic Slicing on Data-Dependent Value,https://github.com/pytorch/pytorch/issues/136083,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2024-09-27T01:39:39Z
issue,136592,[ONNX] Single model export for HF models prompt and token phase,https://github.com/pytorch/pytorch/issues/136592,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2024-09-27T01:18:05Z
issue,131883,Iterating dataloader fails on sliced dataset,https://github.com/pytorch/pytorch/issues/131883,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2024-09-26T19:28:21Z
issue,132492,Wonder why _MultiProcessingDataLoaderIter.__next__ too slow？,https://github.com/pytorch/pytorch/issues/132492,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: data, module: dataloader",scan when time allows,2024-09-26T19:09:46Z
issue,136588,DistributedSampler shuffle option doesn't work as expected,https://github.com/pytorch/pytorch/issues/136588,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2024-09-26T18:57:55Z
issue,136664,[BUG] torch/extension.h: undefined symbol,https://github.com/pytorch/pytorch/issues/136664,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp-extensions, module: pybind",scan when time allows,2024-09-26T17:00:57Z
issue,106634,Confusing error message for DataLoader with num_workers=0 and non-zero timeout,https://github.com/pytorch/pytorch/issues/106634,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2024-09-26T10:06:46Z
issue,136621,A `complex` tensor with `linalg.matrix_norm()` returns a `float` tensor is returned even if I set `dtype=torch.complex64`,https://github.com/pytorch/pytorch/issues/136621,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: complex, module: linear algebra",scan when time allows,2024-09-26T02:56:49Z
issue,136656,[NJT] Dropout(0.0) with NJT increments cuda rng_state (only for no-compile),https://github.com/pytorch/pytorch/issues/136656,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nestedtensor,scan when time allows,2024-09-26T02:33:01Z
issue,98164,forward AD implimentation : _scaled_dot_product_efficient_attention,https://github.com/pytorch/pytorch/issues/98164,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: forward ad,scan when time allows,2024-09-25T06:32:49Z
issue,133040,Add Test for Inductor and Dynamo Config BC Breakages,https://github.com/pytorch/pytorch/issues/133040,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamo, module: inductor, oncall: pt2",scan when time allows,2024-09-25T00:30:26Z
issue,136568,Setting a `complex` tensor to `linalg.vector_norm()` returns a `float` tensor,https://github.com/pytorch/pytorch/issues/136568,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: complex, module: norms and normalization",scan when time allows,2024-09-24T21:31:58Z
issue,136479,The small shape change of input tensor leads to a significant increase in GPU memory usage in Conv3D,https://github.com/pytorch/pytorch/issues/136479,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: memory usage, module: nn",scan when time allows,2024-09-24T15:17:40Z
issue,136409,Linear Sum Assignment,https://github.com/pytorch/pytorch/issues/136409,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: optimizer",scan when time allows,2024-09-24T13:31:45Z
issue,73646,Pure C binding/wrapper with libtorch for inference applications,https://github.com/pytorch/pytorch/issues/73646,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2024-09-24T12:59:46Z
issue,136452,Setting a `complex` tensor to `linalg.norm()` returns a `float` tensor,https://github.com/pytorch/pytorch/issues/136452,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: linear algebra,scan when time allows,2024-09-24T04:48:13Z
issue,135724,[Inductor] Device-specific padding settings,https://github.com/pytorch/pytorch/issues/135724,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",scan when time allows,2024-09-24T04:44:29Z
issue,136410,Implement kaiming_uniform_ for sparse tensors.,https://github.com/pytorch/pytorch/issues/136410,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: initialization, module: nn, module: sparse",scan when time allows,2024-09-23T23:09:34Z
issue,135305,Add restrictions so that Inductor never uses more memory than eager,https://github.com/pytorch/pytorch/issues/135305,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",scan when time allows,2024-09-23T18:24:42Z
issue,22788,Pytorch deadlock from distributed multiprocessing,https://github.com/pytorch/pytorch/issues/22788,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2024-09-23T05:53:31Z
issue,132388,`amp.custom_fwd` interacts poorly with custom `autograd.Function` with `setup_context` separated from `forward`,https://github.com/pytorch/pytorch/issues/132388,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision), module: autograd",scan when time allows,2024-09-22T17:45:27Z
issue,86494,Automatic broadcasting for batch addition for sparse tensors,https://github.com/pytorch/pytorch/issues/86494,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2024-09-22T12:31:28Z
issue,134459,linalg.lu_factor: LU without pivoting is not implemented on the CPU,https://github.com/pytorch/pytorch/issues/134459,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: linear algebra",scan when time allows,2024-09-21T15:03:11Z
issue,136357,"torch.func.vjp and sparse tensors: ""Sparse CSR tensors do not have strides""",https://github.com/pytorch/pytorch/issues/136357,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: functorch, module: sparse",scan when time allows,2024-09-20T16:07:25Z
issue,50341,Interpolation tracking issue,https://github.com/pytorch/pytorch/issues/50341,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: interpolation, module: numpy",scan when time allows,2024-09-20T15:43:26Z
issue,136305,Numpy related failure of gradient and sobolengine distribution testcases on POWER and x86,https://github.com/pytorch/pytorch/issues/136305,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: numpy, module: power",scan when time allows,2024-09-19T16:55:45Z
issue,136250,A question about `getDLDevice`,https://github.com/pytorch/pytorch/issues/136250,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dlpack,scan when time allows,2024-09-19T16:49:35Z
issue,97106,slow torch import on macos,https://github.com/pytorch/pytorch/issues/97106,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: macos, module: performance",scan when time allows,2024-09-19T08:55:40Z
issue,134700,"[c10d][MPI] Attempting to create a new group after MPI causes ""RuntimeError: Underlying Non-PrefixStore shouldn't be null""",https://github.com/pytorch/pytorch/issues/134700,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: mpi, oncall: distributed",scan when time allows,2024-09-17T21:57:45Z
issue,21553,Feature Request: beta cdf,https://github.com/pytorch/pytorch/issues/21553,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distributions,scan when time allows,2024-09-17T20:06:14Z
issue,135930,Integer overflow while creating nested tensors,https://github.com/pytorch/pytorch/issues/135930,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nestedtensor,scan when time allows,2024-09-17T18:55:12Z
issue,136186,"""bmm_sparse_cuda"" not implemented for 'BFloat16'",https://github.com/pytorch/pytorch/issues/136186,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bfloat16, module: sparse",scan when time allows,2024-09-17T18:32:55Z
issue,136201,Compile with USE_CPP_CODE_COVERAGE=ON throw erros,https://github.com/pytorch/pytorch/issues/136201,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2024-09-17T18:19:17Z
issue,94397,jacfwd and jacrev are fundamentally broken for complex inputs,https://github.com/pytorch/pytorch/issues/94397,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: complex, module: functorch",scan when time allows,2024-09-17T17:51:16Z
issue,69471,Include Declarations.yaml in Libtorch distributions,https://github.com/pytorch/pytorch/issues/69471,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: codegen, module: cpp",scan when time allows,2024-09-17T13:04:24Z
issue,124056,MPS slows down after sleep,https://github.com/pytorch/pytorch/issues/124056,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: mps, module: performance",scan when time allows,2024-09-17T06:39:01Z
issue,98675,torch.matmul with batched CSR matrix,https://github.com/pytorch/pytorch/issues/98675,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2024-09-17T05:16:39Z
issue,127517,"With pytest8.2.0 or later, test cases under test/distributed/ execute will meet issue ""object has no attribute 'runTest'. Did you mean: 'run_test'""",https://github.com/pytorch/pytorch/issues/127517,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2024-09-17T05:06:52Z
issue,114097,[RFC] Support for Redundant Hosts in TorchElastic,https://github.com/pytorch/pytorch/issues/114097,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: elastic, oncall: distributed",scan when time allows,2024-09-16T20:03:55Z
issue,101293,Operations to shared tensors in the forked process could lead to silent crash,https://github.com/pytorch/pytorch/issues/101293,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: multiprocessing,scan when time allows,2024-09-16T19:08:41Z
issue,135487,No error raised when trying in-place operation on tensor of shape[0] > 1 (single memory location),https://github.com/pytorch/pytorch/issues/135487,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: advanced indexing, module: cpu, module: edge cases, module: error checking",scan when time allows,2024-09-16T18:21:24Z
issue,136031,Conv vs. Scipy tests fail on MacOS,https://github.com/pytorch/pytorch/issues/136031,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: mps",scan when time allows,2024-09-16T16:03:19Z
issue,136138,`num_features` of `nn.BatchNorm1d()` with a tuple of `int` works,https://github.com/pytorch/pytorch/issues/136138,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-09-16T15:58:14Z
issue,136072,"`padding_mode` parameter of `nn.ConvTranspose1d()`, `nn.ConvTranspose2d()` and `nn.ConvTranspose3d()` is not explained in the docs",https://github.com/pytorch/pytorch/issues/136072,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-09-16T15:57:48Z
issue,136073,The real AttributeError information,https://github.com/pytorch/pytorch/issues/136073,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-09-16T15:57:22Z
issue,136086,Same token different output from `Conv1d`,https://github.com/pytorch/pytorch/issues/136086,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: convolution,scan when time allows,2024-09-16T15:54:44Z
issue,136071,"Setting wrong type values to `stride`, `padding`, `output_padding` and `dilation` argument of `nn.ConvTranspose1d()` gets the wrong error messages saying only `tuple of ints`",https://github.com/pytorch/pytorch/issues/136071,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-09-16T15:52:45Z
issue,136096,"Each parameter of `nn.MaxPool1d()`, `nn.MaxPool2d()` and `nn.MaxPool3d()` should have `required` or `optional`",https://github.com/pytorch/pytorch/issues/136096,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-09-16T15:52:19Z
issue,136098,"`int`, `float` and `complex` type with `return_indices` of `nn.MaxPool1d()` also work",https://github.com/pytorch/pytorch/issues/136098,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-09-16T15:52:00Z
issue,136110,Setting `True` to `return_indices` argument of `nn.MaxPool1d()` with the tensor of `int` works,https://github.com/pytorch/pytorch/issues/136110,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-09-16T15:49:52Z
issue,136145,MultiLabelMarginLoss with customized margin.,https://github.com/pytorch/pytorch/issues/136145,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: loss, module: nn",scan when time allows,2024-09-16T15:33:31Z
issue,135095,Add MSCCL++ as a communication backend for PyTorch,https://github.com/pytorch/pytorch/issues/135095,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nccl, oncall: distributed",scan when time allows,2024-09-16T07:22:22Z
issue,126381,All processes running torch.distributed.destroy_process_group() create CUDA context on device 0,https://github.com/pytorch/pytorch/issues/126381,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: c10d, oncall: distributed",scan when time allows,2024-09-15T22:10:03Z
issue,121333,Automatic graph break bug reporter,https://github.com/pytorch/pytorch/issues/121333,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamo, module: logging, oncall: pt2",scan when time allows,2024-09-14T19:31:17Z
issue,55884,[Bug] RuntimeError: could not create a primitive on Xeon,https://github.com/pytorch/pytorch/issues/55884,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: backend, module: intel",scan when time allows,2024-09-14T06:45:21Z
issue,135705,Multiprocessing with workers returning `torch.Tensor` and limiting of the number of tasks per worker (`maxtasksperchild=1`) hangs,https://github.com/pytorch/pytorch/issues/135705,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: multiprocessing,scan when time allows,2024-09-13T23:13:10Z
issue,135992,"When calculating the loss, the input data does not contain NaN, but the output contains NaN.",https://github.com/pytorch/pytorch/issues/135992,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nans and infs,scan when time allows,2024-09-13T19:36:59Z
issue,135969,Add AT_DISPATCH for layernorm gamma and beta,https://github.com/pytorch/pytorch/issues/135969,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: half,scan when time allows,2024-09-13T19:36:35Z
issue,135837,cudaFree will be called when the reserved memory > 70G for A100 80G,https://github.com/pytorch/pytorch/issues/135837,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: memory usage",scan when time allows,2024-09-13T19:35:57Z
issue,135826,(L)-BFGS with Hager-Zhang line-search,https://github.com/pytorch/pytorch/issues/135826,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2024-09-13T19:35:28Z
issue,135640,Issue UserWarning for missing ModuleList/ModuleDict in Module classes,https://github.com/pytorch/pytorch/issues/135640,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: error checking, module: molly-guard, module: nn",scan when time allows,2024-09-13T19:30:21Z
issue,110866,Gradients (Jacobian) in inference,https://github.com/pytorch/pytorch/issues/110866,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2024-09-13T13:49:19Z
issue,123047,Alternate algorithm for computing MaxPool2D under specific condition.,https://github.com/pytorch/pytorch/issues/123047,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: nn, module: performance",scan when time allows,2024-09-13T09:41:25Z
issue,119549,"Way of mocking collectives, especially data-dependent ones like all2all_single",https://github.com/pytorch/pytorch/issues/119549,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamic shapes, module: minifier, oncall: distributed, oncall: pt2",scan when time allows,2024-09-13T03:21:35Z
issue,118798,Automatic dynamic shapes should give up if a symbolic variable has too many guards on it,https://github.com/pytorch/pytorch/issues/118798,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamic shapes, oncall: pt2",scan when time allows,2024-09-13T03:04:56Z
issue,132617,conda pytorch does not ship libnccl.so,https://github.com/pytorch/pytorch/issues/132617,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: nccl",scan when time allows,2024-09-13T02:27:21Z
issue,130968,float8 types cannot be used under deterministic mode,https://github.com/pytorch/pytorch/issues/130968,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: determinism, module: floatx (formerly float8)",scan when time allows,2024-09-12T18:36:33Z
issue,135660,torch.cat and torch.stack does not invoke `__torch_function__` API,https://github.com/pytorch/pytorch/issues/135660,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: __torch_function__, module: nestedtensor",scan when time allows,2024-09-12T17:58:19Z
issue,66806,Missing doc for `torch.segment_reduce`,https://github.com/pytorch/pytorch/issues/66806,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: numpy",scan when time allows,2024-09-12T14:45:54Z
issue,33122,[feature request] make torch.multinomial behaviour compliant with rnn output dimension,https://github.com/pytorch/pytorch/issues/33122,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bc-breaking, module: distributions",scan when time allows,2024-09-12T04:31:12Z
issue,135648,"Setting a `float` value to `in_channels`, `out_channels` and `kernel_size` argument of `nn.Conv1d()` gets indirect errors and in_channels and kernel_size with with non-bools",https://github.com/pytorch/pytorch/issues/135648,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: error checking, module: nn",scan when time allows,2024-09-11T16:54:57Z
issue,96265,RuntimeError: one of the variables needed for gradient computation has been modified by an inplace operation,https://github.com/pytorch/pytorch/issues/96265,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2024-09-11T08:52:12Z
issue,135135,TORCH_CPP_LOG_LEVEL cannot enable VLOG,https://github.com/pytorch/pytorch/issues/135135,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: logging,scan when time allows,2024-09-11T00:13:31Z
issue,135604,"Generic BLAS provider incorrectly considered ""found""",https://github.com/pytorch/pytorch/issues/135604,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2024-09-10T23:07:06Z
issue,135514,add torch.linalg stubs,https://github.com/pytorch/pytorch/issues/135514,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: linear algebra,scan when time allows,2024-09-10T22:07:57Z
issue,135484,"Boolean indexing not working correctly for a numpy array indexed with a boolean pytorch tensor in the case of 2d singletons (shape of (1, 1))",https://github.com/pytorch/pytorch/issues/135484,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: numpy, module: python frontend",scan when time allows,2024-09-10T15:47:41Z
issue,134851,GPU Precision drops significantly when kernel size is increased from 3 to 5 in Conv3D,https://github.com/pytorch/pytorch/issues/134851,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: cudnn, module: performance",scan when time allows,2024-09-10T14:11:38Z
issue,135359,"binary_cross_entropy_with_logits outputs NaN on -inf, which is inconsistent with documentation",https://github.com/pytorch/pytorch/issues/135359,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bc-breaking, module: edge cases, module: loss, module: nans and infs",scan when time allows,2024-09-10T11:19:48Z
issue,134827,ONNX NMS Operator Output Different From Torchvision Implementation,https://github.com/pytorch/pytorch/issues/134827,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2024-09-10T09:22:32Z
issue,122626,Significant performance degradation with multiprocessing in PyTorch 2.x compared to 1.13.1,https://github.com/pytorch/pytorch/issues/122626,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: multiprocessing,scan when time allows,2024-09-10T08:36:47Z
issue,135267,First all_to_all_single / barrier takes a large amount of time.,https://github.com/pytorch/pytorch/issues/135267,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2024-09-10T01:24:02Z
issue,52718,Discrepancy between CPU->GPU and GPU->CPU data transfer speeds,https://github.com/pytorch/pytorch/issues/52718,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: performance",scan when time allows,2024-09-09T23:14:40Z
issue,135489,Sympy bottleneck in eval_relation/assumptions getit,https://github.com/pytorch/pytorch/issues/135489,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: compile-time, module: dynamic shapes, oncall: pt2",scan when time allows,2024-09-09T19:13:12Z
issue,135153,TestSDPAPrivateUse1Only tests causes irreversible changes to PyTorch internal state,https://github.com/pytorch/pytorch/issues/135153,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: privateuse1, module: tests",scan when time allows,2024-09-09T19:05:49Z
issue,134664,"Questions about CVE-2022-3171, CVE-2022-3509 and CVE-2022-3510",https://github.com/pytorch/pytorch/issues/134664,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: onnx, module: protobuf, module: third_party",scan when time allows,2024-09-09T17:24:56Z
issue,135229,Disable `retain_grad` on non-leaf tensors?,https://github.com/pytorch/pytorch/issues/135229,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2024-09-09T14:39:05Z
issue,134948,Add Validity Checks for Parameter Combinations and Input Shapes in torch.nn.MaxPool2d to Prevent Unexpected Outputs,https://github.com/pytorch/pytorch/issues/134948,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-09-09T13:36:04Z
issue,134945,Broadcasting Parameter destroys hooks in PyTorch 2.4,https://github.com/pytorch/pytorch/issues/134945,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2024-09-09T10:00:59Z
issue,135382,Stop tracking FunctionalTensor bases in Python,https://github.com/pytorch/pytorch/issues/135382,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: functionalization,scan when time allows,2024-09-09T01:43:57Z
issue,135437,"Cannot build libtorch from source with NVHPC compilers - ""Could NOT find Threads""",https://github.com/pytorch/pytorch/issues/135437,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cuda",scan when time allows,2024-09-09T01:33:00Z
issue,135447,[MPS] Correctness issue in backward pass of `nn.ReplicationPad1d` and `nn.ReplicationPad2d`,https://github.com/pytorch/pytorch/issues/135447,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: correctness (silent), module: mps",scan when time allows,2024-09-09T01:29:18Z
issue,65198,Multi-processing leaking file descriptors,https://github.com/pytorch/pytorch/issues/65198,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dataloader, module: multiprocessing",scan when time allows,2024-09-08T15:08:29Z
issue,99781,2.0.0+cu118 package missing proper libnvrtc-builtins.so.11.8,https://github.com/pytorch/pytorch/issues/99781,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: cpp",scan when time allows,2024-09-07T21:22:38Z
issue,135398,"Models that have non-tensor elements in state_dict, are not ONNX-exportable (and not JIT-traceable)",https://github.com/pytorch/pytorch/issues/135398,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2024-09-07T01:02:04Z
issue,134885,SyntaxError when Running repro.py,https://github.com/pytorch/pytorch/issues/134885,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamo, module: minifier, oncall: pt2",scan when time allows,2024-09-06T22:36:58Z
issue,135255,Log missed optimization opportunities due to dynamic shapes in Inductor,https://github.com/pytorch/pytorch/issues/135255,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamic shapes, oncall: pt2",scan when time allows,2024-09-06T22:26:10Z
issue,133281,"Make CUDAPluggableAllocator support getDeviceStats, resetPeakStats",https://github.com/pytorch/pytorch/issues/133281,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: cudacachingallocator",scan when time allows,2024-09-06T21:06:59Z
issue,133280,Pluggable Allocator Does not raise torch.cuda.OutOfMemoryError,https://github.com/pytorch/pytorch/issues/133280,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: error checking, module: memory usage",scan when time allows,2024-09-06T21:03:41Z
issue,120585,Functionalization of `as_strided` disregards its base tensor storage.,https://github.com/pytorch/pytorch/issues/120585,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: functionalization,scan when time allows,2024-09-06T21:03:40Z
issue,135330,torch.floor_divide not triggering ZeroDivisionError on GPU,https://github.com/pytorch/pytorch/issues/135330,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: edge cases, module: error checking",scan when time allows,2024-09-06T19:07:22Z
issue,119971,Non core ATen ops are getting dispatched as it is instead of mapping/docmposing to other ATen ops.,https://github.com/pytorch/pytorch/issues/119971,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dispatch,scan when time allows,2024-09-06T19:04:39Z
issue,128910,[pipelining] Free memory in stage after use,https://github.com/pytorch/pytorch/issues/128910,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: pipelining,scan when time allows,2024-09-06T18:42:06Z
issue,134960,Issue with Weight Synchronization When Using Consecutive DDP Instances in the Same torch.distributed Environment,https://github.com/pytorch/pytorch/issues/134960,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2024-09-06T16:50:23Z
issue,135362,MacOS runner fails at `Complete runner step`,https://github.com/pytorch/pytorch/issues/135362,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: infra,scan when time allows,2024-09-06T16:26:31Z
issue,135198,lstm quantize issues,https://github.com/pytorch/pytorch/issues/135198,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: rnn, oncall: quantization",scan when time allows,2024-09-06T15:29:01Z
issue,134860,JIT tracing a quantized model with hooks is broken,https://github.com/pytorch/pytorch/issues/134860,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: oncall: jit, oncall: quantization",scan when time allows,2024-09-06T15:26:09Z
issue,43369,"torch.load(.., map_location='cpu') fails when unserializing cuda tensors on a cpu-only device serialized with pickle",https://github.com/pytorch/pytorch/issues/43369,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: pickle, module: serialization",scan when time allows,2024-09-06T13:55:23Z
issue,129634,Type Error when using from_numpy,https://github.com/pytorch/pytorch/issues/129634,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: numpy,scan when time allows,2024-09-06T12:49:46Z
issue,123973,Dynamo-based ONNX Export: Failed to produce a graph during tracing as no tensor operations were found.,https://github.com/pytorch/pytorch/issues/123973,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: oncall: export, oncall: pt2",scan when time allows,2024-09-05T23:55:37Z
issue,44710,AMP support for libtorch/c++,https://github.com/pytorch/pytorch/issues/44710,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision),scan when time allows,2024-09-05T19:23:52Z
issue,50092,batch_isend_irecv: the receiving end cannot receive large tensors from the sending end correctly,https://github.com/pytorch/pytorch/issues/50092,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2024-09-05T04:04:34Z
issue,99807,AOTAutograd/Inductor file system cache,https://github.com/pytorch/pytorch/issues/99807,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",scan when time allows,2024-09-05T02:17:51Z
issue,135091,Any support for sparse linear layer in the future?,https://github.com/pytorch/pytorch/issues/135091,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2024-09-04T23:15:30Z
issue,123971,Dynamo Export: Support for mutating module attributes,https://github.com/pytorch/pytorch/issues/123971,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: oncall: export, oncall: pt2",scan when time allows,2024-09-04T23:14:19Z
issue,128325,Cannot export onnx model with autograd.functional.jacobian,https://github.com/pytorch/pytorch/issues/128325,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2024-09-04T21:46:03Z
issue,134926,register_after_fork fails on python built without multiprocessing support,https://github.com/pytorch/pytorch/issues/134926,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: multiprocessing, module: python frontend",scan when time allows,2024-09-04T20:57:30Z
issue,57118,How to view VLOG information,https://github.com/pytorch/pytorch/issues/57118,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: logging,scan when time allows,2024-09-04T19:25:04Z
issue,36380,Population Count Op,https://github.com/pytorch/pytorch/issues/36380,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python frontend,scan when time allows,2024-09-04T17:09:07Z
issue,123353,Type parameter (T) of Optype (Where) bound to different types (tensor(int64) and tensor(float) in node (/Where),https://github.com/pytorch/pytorch/issues/123353,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2024-09-04T14:40:59Z
issue,127875,[ShapeInferenceError] Attribute dilations has incorrect size,https://github.com/pytorch/pytorch/issues/127875,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2024-09-04T14:17:48Z
issue,129200,"Export model using onnx.dynamo_export has bug of ""torch._dynamo.exc.Unsupported: call_method TupleVariable() size [ConstantVariable(int)] {}""",https://github.com/pytorch/pytorch/issues/129200,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2024-09-04T14:09:09Z
issue,128414,How to enable XNNPACK instead of NNPACK/MKLDNN in Windows?,https://github.com/pytorch/pytorch/issues/128414,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: arm, module: windows, module: xnnpack",scan when time allows,2024-09-04T10:33:25Z
issue,134326,Very large memory allocation by torch.linalg.qr,https://github.com/pytorch/pytorch/issues/134326,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: linear algebra,scan when time allows,2024-09-04T08:14:31Z
issue,134901,How to calculate second derivative using PyTorch with GPU (cuda),https://github.com/pytorch/pytorch/issues/134901,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: cudnn, module: double backwards, module: functorch",scan when time allows,2024-09-04T01:48:21Z
issue,134977,[BUG] found a weird problem of nn.conv2d in pytorch 2.2.0,https://github.com/pytorch/pytorch/issues/134977,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cudnn, module: numerical-stability",scan when time allows,2024-09-04T01:45:13Z
issue,121482,torch.linalg.eigh is very slow compared to tf.linalg.eigh,https://github.com/pytorch/pytorch/issues/121482,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: linear algebra, module: mkl, module: performance",scan when time allows,2024-09-03T21:29:18Z
issue,134803,enhanced debugging prints in torch.distributed,https://github.com/pytorch/pytorch/issues/134803,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2024-09-03T21:22:58Z
issue,108968,Unnecessary cuda synchronizations that we should remove in PyTorch,https://github.com/pytorch/pytorch/issues/108968,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: performance",scan when time allows,2024-09-03T20:23:02Z
issue,134947,torch.hub.load will raise RuntimeError: operator torchvision::nms does not exist,https://github.com/pytorch/pytorch/issues/134947,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: hub,scan when time allows,2024-09-03T19:13:31Z
issue,134953,_rebuild_sparse_tensor,https://github.com/pytorch/pytorch/issues/134953,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2024-09-03T19:11:14Z
issue,134996,Setting CPU affinity doesn't work for torch.softmax when using a thread,https://github.com/pytorch/pytorch/issues/134996,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpu,scan when time allows,2024-09-03T19:03:45Z
issue,122641,[vulkan] Vulkan backend fails creating tensor on aarch64 (Raspberry Pi 5),https://github.com/pytorch/pytorch/issues/122641,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: vulkan,scan when time allows,2024-09-01T12:03:10Z
issue,122694,elastic agent kill hang process when other processes timeout,https://github.com/pytorch/pytorch/issues/122694,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: elastic, oncall: distributed",scan when time allows,2024-08-31T02:35:36Z
issue,117736,Illegal memory access for nn.Conv2d when intermediate output has size larger than 2**31,https://github.com/pytorch/pytorch/issues/117736,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: 64-bit, module: cuda, module: edge cases",scan when time allows,2024-08-30T23:04:27Z
issue,134641,nn.Module.to(memory_format= channels_last format) failed if containing 5D parameters,https://github.com/pytorch/pytorch/issues/134641,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: nn",scan when time allows,2024-08-30T21:23:25Z
issue,134668,"Whether tensor parallelism supports the overlap of communication calculations for gradient computation, and how to implement it",https://github.com/pytorch/pytorch/issues/134668,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2024-08-30T17:54:43Z
issue,101878,torch.quantile on MPS doesn't sort values when dim is not None,https://github.com/pytorch/pytorch/issues/101878,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: mps",scan when time allows,2024-08-30T17:33:04Z
issue,134637,LSTM inference c++ threads block on DropoutState,https://github.com/pytorch/pytorch/issues/134637,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: multithreading, module: rnn",scan when time allows,2024-08-30T15:06:04Z
issue,134607,torch.angle cuda version not implemented for bfloat16 and half (float16),https://github.com/pytorch/pytorch/issues/134607,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2024-08-30T15:04:58Z
issue,134821,Add decompositions for nanmedian and median,https://github.com/pytorch/pytorch/issues/134821,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: decompositions,scan when time allows,2024-08-30T14:42:31Z
issue,15738,Beta Distribution values wrong for a=b---> 0,https://github.com/pytorch/pytorch/issues/15738,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distributions,scan when time allows,2024-08-30T13:56:34Z
issue,120996,"Unnecessary CUDA synchronization, or missing warnings",https://github.com/pytorch/pytorch/issues/120996,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2024-08-30T08:22:50Z
issue,10684,make install error： [third_party/gloo/gloo/CMakeFiles/gloo.dir/all] Error 2,https://github.com/pytorch/pytorch/issues/10684,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: third_party",scan when time allows,2024-08-30T08:00:29Z
issue,119447,Cannot install or import PyTorch Nightly,https://github.com/pytorch/pytorch/issues/119447,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: binaries,scan when time allows,2024-08-30T06:54:43Z
issue,134755,"""Attempted to send CUDA tensor received from another process"" even though that was not the case",https://github.com/pytorch/pytorch/issues/134755,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: multiprocessing,scan when time allows,2024-08-30T05:04:53Z
issue,134612,torch.func.grad uses more memory than expected,https://github.com/pytorch/pytorch/issues/134612,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: functorch, module: memory usage",scan when time allows,2024-08-30T04:57:30Z
issue,134691,[docs] We should add a lint for bullet points and multiple lines,https://github.com/pytorch/pytorch/issues/134691,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: lint",scan when time allows,2024-08-29T19:03:24Z
issue,133761,Incorrect GPU management and deadlocks without torch.cuda.set_device,https://github.com/pytorch/pytorch/issues/133761,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2024-08-29T18:42:16Z
issue,132231,torch.nn.Bilinear weights initialization reason,https://github.com/pytorch/pytorch/issues/132231,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-08-28T23:32:48Z
issue,117209,"torch.onnx.export the torch,nn.TransformerEncoderLayer, the dynamic_axes work not right",https://github.com/pytorch/pytorch/issues/117209,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2024-08-28T22:33:33Z
issue,125173,Flight Recorder Sequence IDs are insufficient,https://github.com/pytorch/pytorch/issues/125173,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2024-08-28T22:25:01Z
issue,121519,Avoid shard 0 download on all ranks,https://github.com/pytorch/pytorch/issues/121519,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2024-08-28T22:24:19Z
issue,121311,In-place native functional collectives right after a custom torch op fail to codegen in inductor,https://github.com/pytorch/pytorch/issues/121311,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: distributed, oncall: pt2",scan when time allows,2024-08-28T19:07:03Z
issue,118472,[c10d] Avoid busy looping in C10D_NCCL_CHECK_TIMEOUT,https://github.com/pytorch/pytorch/issues/118472,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nccl, oncall: distributed",scan when time allows,2024-08-28T18:14:10Z
issue,134680,Use `torch._C._stash_obj_in_tls` for global state in serialization,https://github.com/pytorch/pytorch/issues/134680,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: serialization,scan when time allows,2024-08-28T14:52:32Z
issue,30968,Categorical.sample too slow,https://github.com/pytorch/pytorch/issues/30968,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distributions,scan when time allows,2024-08-28T14:04:56Z
issue,126523,Summary of tests in machine readable format,https://github.com/pytorch/pytorch/issues/126523,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tests,scan when time allows,2024-08-28T12:06:45Z
issue,116322,[RFC] Intel GPU Runtime Upstreaming for Allocator,https://github.com/pytorch/pytorch/issues/116322,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: intel,scan when time allows,2024-08-28T10:47:30Z
issue,134570,histogramdd 'Cuda' backend,https://github.com/pytorch/pytorch/issues/134570,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2024-08-28T09:39:35Z
issue,125182,QR Decomposition for Sparse Matrix,https://github.com/pytorch/pytorch/issues/125182,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: linear algebra, module: sparse",scan when time allows,2024-08-28T09:22:41Z
issue,33296,Training got stuck due to timeout from dataloader,https://github.com/pytorch/pytorch/issues/33296,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dataloader, module: performance",scan when time allows,2024-08-28T07:46:46Z
issue,133849,CatArrayBatchedCopy and AllGather don't overlap during FSDP backward,https://github.com/pytorch/pytorch/issues/133849,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fsdp,scan when time allows,2024-08-28T05:05:49Z
issue,56136,"Please add a tab of ""all"" packages in `Docs` option",https://github.com/pytorch/pytorch/issues/56136,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2024-08-28T02:44:46Z
issue,134190,ProcessGroupNCCL::WorkNCCL only records seq_id which is not enough to be matched with comm op in Execution Trace replay,https://github.com/pytorch/pytorch/issues/134190,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2024-08-28T00:41:17Z
issue,74537,ComplexHalf Coverage Tracker,https://github.com/pytorch/pytorch/issues/74537,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: complex, module: half",scan when time allows,2024-08-27T23:47:04Z
issue,123532,Request: Restrict use of tags to strictly formal release versions,https://github.com/pytorch/pytorch/issues/123532,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ci, oncall: releng",scan when time allows,2024-08-27T21:55:26Z
issue,112459,GRUCell batching rule for vmap,https://github.com/pytorch/pytorch/issues/112459,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: functorch,scan when time allows,2024-08-27T19:51:27Z
issue,131650,"Pytorch 2.4, Cuda 12.4: RuntimeError: device >= 0 && device < num_gpus INTERNAL ASSERT FAILED at ""/opt/conda/conda-bld/pytorch_1720538439675/work/aten/src/ATen/cuda/CUDAContext.cpp"":49, please report a bug to PyTorch. device=^A, num_gpus=^A",https://github.com/pytorch/pytorch/issues/131650,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2024-08-27T11:30:19Z
issue,37324,Multiprocessing: model shared between processes hangs during copy.deepcopy,https://github.com/pytorch/pytorch/issues/37324,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: multiprocessing",scan when time allows,2024-08-27T09:58:53Z
issue,134180,[export] Incorrect behavior with torch.LongTensor and torch.Tensor,https://github.com/pytorch/pytorch/issues/134180,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: export,scan when time allows,2024-08-27T05:17:35Z
issue,133941,"Why core aten IR doesn’t contain scalar_tensor overload version of some ops, like bitwise-like ops, remainder and so on?",https://github.com/pytorch/pytorch/issues/133941,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: core aten,scan when time allows,2024-08-27T03:28:53Z
issue,134403,RuntimeError: cuFFT error: CUFFT_INTERNAL_ERROR,https://github.com/pytorch/pytorch/issues/134403,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: fft",scan when time allows,2024-08-26T18:30:35Z
issue,133985,Feat: output dtype for count_nonzero,https://github.com/pytorch/pytorch/issues/133985,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python frontend,scan when time allows,2024-08-26T17:57:31Z
issue,134183,storage_module && PyModule_Check(storage_module) INTERNAL ASSERT FAILED at DynamicTypes.cpp,https://github.com/pytorch/pytorch/issues/134183,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: edge cases, module: error checking, module: python frontend",scan when time allows,2024-08-26T17:29:01Z
issue,134441,Cuda memory error when testing GroupedGemm on a multi-Gpu environment,https://github.com/pytorch/pytorch/issues/134441,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2024-08-26T17:01:17Z
issue,134439,configure ignores USE_QNNPACK=0 and still looks for third_party/QNNPACK,https://github.com/pytorch/pytorch/issues/134439,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2024-08-26T17:00:29Z
issue,134069,Kron with gradient for sparse tensors,https://github.com/pytorch/pytorch/issues/134069,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: sparse",scan when time allows,2024-08-26T15:51:04Z
issue,32575,Interrupting a DDP worker while using the CUDA MPS server causes CUDA to hang until reboot,https://github.com/pytorch/pytorch/issues/32575,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, oncall: distributed",scan when time allows,2024-08-26T13:37:21Z
issue,126045,A scalar with `elements` or `test_elements` argument of `isin()` doesn't work,https://github.com/pytorch/pytorch/issues/126045,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python frontend,scan when time allows,2024-08-26T09:39:37Z
issue,134293,[PrivateUse1 Backend] Unable to register backend to PrivateUse1 dispatch key: ModuleNotFoundError: No module named 'torch.privateuseone',https://github.com/pytorch/pytorch/issues/134293,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: privateuse1,scan when time allows,2024-08-26T03:25:02Z
issue,126972,ONNX Exporter Fails with Handling Complex Tensors,https://github.com/pytorch/pytorch/issues/126972,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2024-08-25T22:03:02Z
issue,42223,Implementing packbits,https://github.com/pytorch/pytorch/issues/42223,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: numpy,scan when time allows,2024-08-25T16:48:25Z
issue,125683,[RFC] Add Cpp Template for GEMM related ops via max-autotune for Inductor CPU,https://github.com/pytorch/pytorch/issues/125683,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: cpu inductor, oncall: pt2",scan when time allows,2024-08-25T10:33:02Z
issue,70513,Conda Repodata.json file not found in Pytorch channel,https://github.com/pytorch/pytorch/issues/70513,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: binaries,scan when time allows,2024-08-25T01:58:30Z
issue,134249,Footgun: Dynamo x tensors created inside or outside of inference_mode,https://github.com/pytorch/pytorch/issues/134249,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dynamo,scan when time allows,2024-08-24T20:45:23Z
issue,84994,Cuda tensor is zero when passed through multiprocessing queue,https://github.com/pytorch/pytorch/issues/84994,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: multiprocessing,scan when time allows,2024-08-23T20:23:28Z
issue,134051,FSDP activation CPU offload memory not cleaned up,https://github.com/pytorch/pytorch/issues/134051,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: memory usage, oncall: distributed",scan when time allows,2024-08-23T19:00:37Z
issue,20248,Sparse tensors can't be used in DataLoader running many workers,https://github.com/pytorch/pytorch/issues/20248,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: multiprocessing, module: sparse",scan when time allows,2024-08-23T18:27:45Z
issue,134201,Python310 trying to run GFPGAN comes: ImportError: cannot import name 'sym_float' from 'torch' (unknown location),https://github.com/pytorch/pytorch/issues/134201,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2024-08-23T14:55:05Z
issue,82443,Quantization issue in transformers,https://github.com/pytorch/pytorch/issues/82443,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2024-08-23T02:23:55Z
issue,55910,Allow variable intermediate hidden dimensions for stacked RNN/LSTM layers,https://github.com/pytorch/pytorch/issues/55910,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: rnn",scan when time allows,2024-08-22T23:44:07Z
issue,134117,Add aarch64 and aarch64 cuda version to conda,https://github.com/pytorch/pytorch/issues/134117,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: arm, module: binaries",scan when time allows,2024-08-22T22:41:01Z
issue,134244,Setting a 1D tensor as `input` and a 0D tensor as `index` and vice versa for `gather()` works against the error message,https://github.com/pytorch/pytorch/issues/134244,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: edge cases, module: error checking, module: scatter & gather ops",scan when time allows,2024-08-22T19:30:58Z
issue,132412,[CI] PyTorch eager is slow in pytorch-linux-jammy-aarch64-py3.10-gcc11-inductor-benchmarks docker,https://github.com/pytorch/pytorch/issues/132412,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: arm, module: ci",scan when time allows,2024-08-22T04:16:48Z
issue,79715,[FX] TypeError when tracing cat taking split's output as input,https://github.com/pytorch/pytorch/issues/79715,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2024-08-22T03:58:00Z
issue,100656,"On UMA systems, pytorch fails to reserve memory exceeding the initial memory size",https://github.com/pytorch/pytorch/issues/100656,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: memory usage, module: rocm",scan when time allows,2024-08-22T02:32:52Z
issue,134079,operator benchmark diag test cuda problem,https://github.com/pytorch/pytorch/issues/134079,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: benchmark,scan when time allows,2024-08-21T23:03:26Z
issue,134098,libtorch reasoned that the crnn model was out of speed the first few times,https://github.com/pytorch/pytorch/issues/134098,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, oncall: jit",scan when time allows,2024-08-21T22:57:40Z
issue,133380,AttributeError: 'NoneType' object has no attribute 'is_failed',https://github.com/pytorch/pytorch/issues/133380,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2024-08-21T16:21:41Z
issue,134038,vmap tests aren't erroring on fallback usage,https://github.com/pytorch/pytorch/issues/134038,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: functorch, module: vmap",scan when time allows,2024-08-20T22:28:10Z
issue,133998,`torch.quantile`: `nan` instead of `inf`,https://github.com/pytorch/pytorch/issues/133998,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python frontend,scan when time allows,2024-08-20T21:39:30Z
issue,133754,sample of Binomial should return int instead of float,https://github.com/pytorch/pytorch/issues/133754,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distributions,scan when time allows,2024-08-20T18:17:31Z
issue,133818,"Warning when initialize `torch.nn.Linear(0, d)`",https://github.com/pytorch/pytorch/issues/133818,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-08-20T18:15:16Z
issue,128046,Future LRScheduler implementation with Tensor support,https://github.com/pytorch/pytorch/issues/128046,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: lrscheduler,scan when time allows,2024-08-20T15:07:43Z
issue,86989,Error: unknown architecture `armv7-a;' and Error: selected processor does not support `command' in ARM mode,https://github.com/pytorch/pytorch/issues/86989,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: arm, oncall: quantization",scan when time allows,2024-08-20T11:38:24Z
issue,131691,TCP socket keeps trying IPv6 address and doesn't try IPv4,https://github.com/pytorch/pytorch/issues/131691,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2024-08-20T01:18:56Z
issue,133383,[Inductor] Remove dead code from triton kernels,https://github.com/pytorch/pytorch/issues/133383,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",scan when time allows,2024-08-19T19:18:27Z
issue,133210,IPC feature supports third-party devices,https://github.com/pytorch/pytorch/issues/133210,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: multiprocessing, module: privateuse1, module: third_party",scan when time allows,2024-08-19T17:24:38Z
issue,89051,caffe2_interface_library CMake macro prevents linking to LibTorch as a transitive dependency,https://github.com/pytorch/pytorch/issues/89051,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp,scan when time allows,2024-08-19T08:09:08Z
issue,51836,TypeError: can't pickle _thread.lock objects when using spawn / forkserver,https://github.com/pytorch/pytorch/issues/51836,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: pickle,scan when time allows,2024-08-18T09:06:45Z
issue,45414,Module __call__ typing,https://github.com/pytorch/pytorch/issues/45414,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: typing,scan when time allows,2024-08-17T21:09:17Z
issue,133522,"`init` of `PReLU()` works with `int`, `complex` and `bool` against what the doc says.",https://github.com/pytorch/pytorch/issues/133522,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-08-17T18:28:49Z
issue,108977,[feature request] Provide some sparse eigen solver(s) for PyTorch (maybe via `ARPACK` as in scipy) + SPD sparse / laplace linear system solver - maybe NVidia AMGx library?,https://github.com/pytorch/pytorch/issues/108977,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: linear algebra, module: sparse",scan when time allows,2024-08-17T11:54:54Z
issue,132428,"MultiProcTest Exception ""None is not -6""",https://github.com/pytorch/pytorch/issues/132428,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2024-08-16T21:16:29Z
issue,133453,[randperm] Add argument to disallow identiy mappings,https://github.com/pytorch/pytorch/issues/133453,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: random,scan when time allows,2024-08-16T21:01:56Z
issue,133564,torch.cond example fails in PyTorch2.4,https://github.com/pytorch/pytorch/issues/133564,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python frontend,scan when time allows,2024-08-16T21:01:27Z
issue,130238,how to simplify torch.fx like using onnxsim?,https://github.com/pytorch/pytorch/issues/130238,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2024-08-16T13:40:42Z
issue,133527,Using `PReLU()` with a `complex` tensor and without `dtype=torch.complex64` should return a simple error message,https://github.com/pytorch/pytorch/issues/133527,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-08-16T13:31:30Z
issue,133535,"When using a wrong value for `PReLU()`, a PReLU's error message should be returned directly instead of `empty()` error.",https://github.com/pytorch/pytorch/issues/133535,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-08-16T13:31:13Z
issue,53441,"Conjugate gradient Descent, and Linear operator are not present in pytorch.",https://github.com/pytorch/pytorch/issues/53441,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: complex, module: lazy, module: linear algebra, module: sparse",scan when time allows,2024-08-15T23:32:28Z
issue,26288,Inplace and out arguments for BatchNorm (and other norm layers: InstanceNorm / LayerNorm / GroupNorm ...),https://github.com/pytorch/pytorch/issues/26288,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: norms and normalization, module: performance",scan when time allows,2024-08-15T21:12:42Z
issue,132596,MPS backend leaks memory when input sizes vary,https://github.com/pytorch/pytorch/issues/132596,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: memory usage, module: mps",scan when time allows,2024-08-15T20:24:46Z
issue,123761,Support for no-frills FP8 matmuls,https://github.com/pytorch/pytorch/issues/123761,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: floatx (formerly float8),scan when time allows,2024-08-15T13:33:17Z
issue,129803,3D matrix support in _scaled_mm,https://github.com/pytorch/pytorch/issues/129803,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: floatx (formerly float8),scan when time allows,2024-08-15T13:30:36Z
issue,133224,"When using parallelize_module to do tp parallel and pp parallel mixing, the SendRecv time is too long",https://github.com/pytorch/pytorch/issues/133224,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2024-08-14T12:54:14Z
issue,123072,Saving & Loading Model Across [In-tree and out-of-tree] Devices,https://github.com/pytorch/pytorch/issues/123072,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: backend, module: privateuse1, module: serialization",scan when time allows,2024-08-14T06:00:03Z
issue,120814,torch.compile() and optimizer UX issues,https://github.com/pytorch/pytorch/issues/120814,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: optimizer, module: pt2 optimizer, oncall: pt2",scan when time allows,2024-08-13T19:37:29Z
issue,133100,MPS clamp_min on older MacOS,https://github.com/pytorch/pytorch/issues/133100,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: mps",scan when time allows,2024-08-13T17:50:41Z
issue,133262,export error when tracing tensor subclass tensors: `'FakeTensor' object has no attribute 'get_plain'`,https://github.com/pytorch/pytorch/issues/133262,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: export,scan when time allows,2024-08-13T17:13:38Z
issue,119474,Extend functionality of torch.nn.functional.fold to apply to sequences (1D tensors + batch & channels) not just images (2D tensors + batch & channels),https://github.com/pytorch/pytorch/issues/119474,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-08-13T14:11:08Z
issue,92701,set_default_device/torch.device has performance impact for non-factory functions,https://github.com/pytorch/pytorch/issues/92701,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: __torch_function__, module: performance",scan when time allows,2024-08-13T13:15:03Z
issue,106971,Found two conflicting CUDA installs,https://github.com/pytorch/pytorch/issues/106971,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2024-08-13T09:39:08Z
issue,133006,Precision Differences in Using `dropout` and `linear`,https://github.com/pytorch/pytorch/issues/133006,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: numerical-stability,scan when time allows,2024-08-13T08:30:30Z
issue,133183,convolution_overrideable not implemented when executing VAEEncode in ComfyUI,https://github.com/pytorch/pytorch/issues/133183,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: privateuse1,scan when time allows,2024-08-12T20:51:31Z
issue,133138,Layer normalization for Nested Tensors does not return the correct tensor types,https://github.com/pytorch/pytorch/issues/133138,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nestedtensor,scan when time allows,2024-08-12T20:50:20Z
issue,133114,Conv1D: `CUDNN_BACKEND_OPERATION: cudnnFinalize Failed cudnn_status: CUDNN_STATUS_BAD_PARAM`,https://github.com/pytorch/pytorch/issues/133114,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: cuda, module: cudnn",scan when time allows,2024-08-12T20:49:02Z
issue,132696,Hanging at waitForFutureOrTimeout when actively destroying ProcessGroups,https://github.com/pytorch/pytorch/issues/132696,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2024-08-12T18:52:44Z
issue,133119,MultiProcContinuousTest doesn't work with pytest/buck,https://github.com/pytorch/pytorch/issues/133119,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2024-08-12T17:52:33Z
issue,132747,Why is kvcache transfer slow using peer-to-peer communication (two GPUs on one machine).,https://github.com/pytorch/pytorch/issues/132747,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nccl, oncall: distributed",scan when time allows,2024-08-12T17:40:37Z
issue,9873,"Pytorch is slow when only using CPU, and cannot utilize multicore of CPU",https://github.com/pytorch/pytorch/issues/9873,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: multithreading, module: performance",scan when time allows,2024-08-12T09:11:30Z
issue,40581,Libtorch C++ multiple GPU performance slower than single GPU,https://github.com/pytorch/pytorch/issues/40581,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: data parallel, module: performance",scan when time allows,2024-08-12T06:33:30Z
issue,132693,INTERNAL ASSERT FAILED in BatchLinearAlgebra.cpp,https://github.com/pytorch/pytorch/issues/132693,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: linear algebra, module: mkl",scan when time allows,2024-08-10T17:20:53Z
issue,63558,universal binaries for libtorch on mac (x86_64+arm),https://github.com/pytorch/pytorch/issues/63558,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: arm, module: binaries, module: macos",scan when time allows,2024-08-10T13:37:11Z
issue,3898,Sparse matrices in dataloader error,https://github.com/pytorch/pytorch/issues/3898,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2024-08-09T23:32:29Z
issue,60294,Add a NumPy-like `pad` function,https://github.com/pytorch/pytorch/issues/60294,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: numpy, module: padding",scan when time allows,2024-08-09T22:00:15Z
issue,57911,Circular padding in Convolution layers should not only be wrap for once.,https://github.com/pytorch/pytorch/issues/57911,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: nn, module: padding",scan when time allows,2024-08-09T21:59:23Z
issue,132991,"_pickle.UnpicklingError: invalid load key, '\x00'. during torch load",https://github.com/pytorch/pytorch/issues/132991,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: serialization,scan when time allows,2024-08-09T17:47:22Z
issue,133084,Addition of Visual Information Fidelity loss,https://github.com/pytorch/pytorch/issues/133084,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: loss,scan when time allows,2024-08-09T12:33:19Z
issue,68616,"`torch.flip`, `torch.roll`, `torch.tile` has arguments `dims=` for both dim/multidim while other functions have argument `dim=` for the same usage",https://github.com/pytorch/pytorch/issues/68616,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: internals, module: python frontend",scan when time allows,2024-08-08T20:46:38Z
issue,132212,NotImplementedError in some APIs with Nested Jagged Tensors,https://github.com/pytorch/pytorch/issues/132212,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nestedtensor,scan when time allows,2024-08-08T20:23:26Z
issue,119072,Importing PyTorch 2.2 fails with undefined symbol error: ncclCommRegister,https://github.com/pytorch/pytorch/issues/119072,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: nccl, module: third_party",scan when time allows,2024-08-08T15:39:42Z
issue,132483,Non-equivalent behaviour of deprecated set_default_tensor_type and set_default_dtype/set_default_device,https://github.com/pytorch/pytorch/issues/132483,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python frontend,scan when time allows,2024-08-08T12:18:23Z
issue,61244,error: invalid conversion in vec256_bfloat16.h,https://github.com/pytorch/pytorch/issues/61244,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2024-08-08T10:38:04Z
issue,132951,torchelastic multiprocessing start_processes documentation out of date,https://github.com/pytorch/pytorch/issues/132951,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: elastic,scan when time allows,2024-08-08T01:06:20Z
issue,132919,Add a warning for `drop_last` argument in `DataLoader`,https://github.com/pytorch/pytorch/issues/132919,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2024-08-08T01:04:12Z
issue,128779,The name of the function `nn.L1Loss()` should be `nn.MAE()` or the name of the function `MSELoss()` should be `nn.L2Loss()`,https://github.com/pytorch/pytorch/issues/128779,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-08-07T20:55:00Z
issue,107896,[feature request] [ux proposal] Min-max linear normalization to be supported in F.normalize (or in a new function),https://github.com/pytorch/pytorch/issues/107896,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-08-07T20:52:47Z
issue,128064,"Improve debugability of warnings/errors ""Triggered internally at""",https://github.com/pytorch/pytorch/issues/128064,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: error checking",scan when time allows,2024-08-07T20:37:30Z
issue,129118,interpolate nearest get values zero when outputs over 4G elements,https://github.com/pytorch/pytorch/issues/129118,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: 64-bit, module: correctness (silent), module: cuda, module: interpolation",scan when time allows,2024-08-07T19:52:50Z
issue,132028,torch.sspaddmm: self.is_sparse() INTERNAL ASSERT FAILED on torch 2.5.0.dev20240708+cu121,https://github.com/pytorch/pytorch/issues/132028,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2024-08-07T15:21:34Z
issue,132759,Delete ProxyTorchDispatchMode.enable_tracing,https://github.com/pytorch/pytorch/issues/132759,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: proxytensor,scan when time allows,2024-08-07T13:37:30Z
issue,124351,BUG Report using `backend:cudaMallocAsync`,https://github.com/pytorch/pytorch/issues/124351,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: cudacachingallocator",scan when time allows,2024-08-07T00:49:02Z
issue,129122,Questions about CVE-2024-31583 and CVE-2024-31580,https://github.com/pytorch/pytorch/issues/129122,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2024-08-06T12:51:25Z
issue,132612,bitwise_xor seems to be not optimized,https://github.com/pytorch/pytorch/issues/132612,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: performance, module: python frontend",scan when time allows,2024-08-06T08:51:21Z
issue,132614,BUG: `torch.special.gammainc`: returns finite results with NaN input,https://github.com/pytorch/pytorch/issues/132614,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nans and infs, module: special",scan when time allows,2024-08-05T20:21:24Z
issue,132600,Inquiry about Test Error after Static Compilation,https://github.com/pytorch/pytorch/issues/132600,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: static linking",scan when time allows,2024-08-05T20:20:26Z
issue,131463,Cannot subclass _TensorBase directly error,https://github.com/pytorch/pytorch/issues/131463,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python frontend,scan when time allows,2024-08-05T20:15:23Z
issue,132608,topk with mask - efficent,https://github.com/pytorch/pytorch/issues/132608,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: masked operators, module: performance, module: sorting and selection",scan when time allows,2024-08-05T19:17:30Z
issue,132599,[BUG+FIX] LibTorch silently fails to find cublas_v2.h for NVHPC,https://github.com/pytorch/pytorch/issues/132599,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cublas, module: cuda",scan when time allows,2024-08-05T18:27:00Z
issue,132602,"Is torch.compile imprecise, as this 4-year-old comment in amp.grad_scaler claims?",https://github.com/pytorch/pytorch/issues/132602,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision), module: numerical-stability",scan when time allows,2024-08-05T15:56:26Z
issue,132534,triton import inside function doesn't work,https://github.com/pytorch/pytorch/issues/132534,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: custom-operators,scan when time allows,2024-08-05T15:55:53Z
issue,132649,nn.Parallel,https://github.com/pytorch/pytorch/issues/132649,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2024-08-05T15:50:48Z
issue,113743,The cuda batched GEMM has a poor performance for bigger batch size with smaller matrix size,https://github.com/pytorch/pytorch/issues/113743,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: performance",scan when time allows,2024-08-05T10:53:12Z
issue,127391,Unable to quantize resnet50 model with post training static quantization,https://github.com/pytorch/pytorch/issues/127391,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2024-08-05T06:43:58Z
issue,91566,Build Error: OpenMP library could not be found. Proceeding might lead to highly sub-optimal performance.,https://github.com/pytorch/pytorch/issues/91566,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: mkldnn, module: third_party",scan when time allows,2024-08-04T08:53:19Z
issue,74454,torch has no attribute sparse_csr_tensor,https://github.com/pytorch/pytorch/issues/74454,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2024-08-04T08:25:51Z
issue,131307,allow tuples as keys for ModuleDict,https://github.com/pytorch/pytorch/issues/131307,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-08-04T06:56:55Z
issue,105728,Compile error PyTorch 2.0.1 / GCC 13.1.0,https://github.com/pytorch/pytorch/issues/105728,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2024-08-02T19:09:56Z
issue,132485,lazy_clone does not work for pinned memory,https://github.com/pytorch/pytorch/issues/132485,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: copy on write, module: cuda",scan when time allows,2024-08-02T14:27:30Z
issue,73829,"Run libtorch examples, export error ""undefined reference to xxx""",https://github.com/pytorch/pytorch/issues/73829,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: abi, module: binaries, module: build",scan when time allows,2024-08-02T14:04:00Z
issue,132026,"Incompatibility with Numpy,JAX and MLX for int and bool type inputs",https://github.com/pytorch/pytorch/issues/132026,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: numpy, module: python frontend",scan when time allows,2024-08-02T08:23:07Z
issue,25888,Forward/backward hooks for C++ torch::nn modules,https://github.com/pytorch/pytorch/issues/25888,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: cpp",scan when time allows,2024-08-02T08:04:10Z
issue,132130,BuildExtension breaks build because of absolute path,https://github.com/pytorch/pytorch/issues/132130,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp-extensions,scan when time allows,2024-08-02T07:42:04Z
issue,108158,TORCHELASTIC_RESTART_COUNT doesn't seem to be broadcasted to all worker,https://github.com/pytorch/pytorch/issues/108158,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: elastic, oncall: r2p",scan when time allows,2024-08-01T18:23:26Z
issue,125389,Support for Kolmogorov-Arnold Networks (KANs),https://github.com/pytorch/pytorch/issues/125389,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-08-01T11:45:32Z
issue,46809,Including AdaBound in the list of Optimizers.,https://github.com/pytorch/pytorch/issues/46809,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2024-08-01T09:18:59Z
issue,132041,`torch.nn.functional.max_unpool2d`'s behavior is different on cpu and gpu on torch 2.5.0.dev20240708+cu121,https://github.com/pytorch/pytorch/issues/132041,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: python frontend",scan when time allows,2024-07-31T20:57:42Z
issue,132300,CUDA Memory Explosion -- Conv3D,https://github.com/pytorch/pytorch/issues/132300,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cublas,scan when time allows,2024-07-31T20:30:20Z
issue,126856,TorchDynamo ONNX Export does not work as expected with masking (ScatterElements),https://github.com/pytorch/pytorch/issues/126856,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2024-07-31T17:01:46Z
issue,131688,Drop Connect Layer,https://github.com/pytorch/pytorch/issues/131688,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-07-31T15:20:31Z
issue,132045,`torch.scatter`'s behavior is different on cpu and gpu on torch 2.5.0.dev20240708+cu121,https://github.com/pytorch/pytorch/issues/132045,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: python frontend",scan when time allows,2024-07-30T22:05:15Z
issue,131753,PyTorch complains ```A leaf node was passed to _will_engine_execute_node``` when calculating the gradient of a model warpped by ```FlopCounterMode```.,https://github.com/pytorch/pytorch/issues/131753,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: __torch_dispatch__, module: autograd",scan when time allows,2024-07-30T15:46:11Z
issue,70529,BatchNorm on variable-length sequences or batches,https://github.com/pytorch/pytorch/issues/70529,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2024-07-30T07:53:47Z
issue,119123,"`DataLoader` generic argument is dropped in `__iter__`, making it non type-safe",https://github.com/pytorch/pytorch/issues/119123,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dataloader, module: typing",scan when time allows,2024-07-29T23:36:01Z
issue,129459,Add pmap to torch.func for Parallel Function Execution,https://github.com/pytorch/pytorch/issues/129459,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: functorch, oncall: distributed",scan when time allows,2024-07-29T20:06:51Z
issue,124990,[DTensor] Sharding strategy not implemented error should be thrown earlier,https://github.com/pytorch/pytorch/issues/124990,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dtensor, oncall: distributed",scan when time allows,2024-07-29T19:48:07Z
issue,131769,FlopCounterMode requires `tabulate` dependency,https://github.com/pytorch/pytorch/issues/131769,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python frontend,scan when time allows,2024-07-29T19:06:00Z
issue,132034,"Implementation of a Q-RNN layer nn.QRNN(filters,kernel,stride,n_cnn,)",https://github.com/pytorch/pytorch/issues/132034,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: rnn",scan when time allows,2024-07-29T18:59:21Z
issue,132047,Inconsistent behavior of gumbel_softmax on cpu vs. gpu,https://github.com/pytorch/pytorch/issues/132047,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: python frontend",scan when time allows,2024-07-29T16:49:48Z
issue,132052,Make C++ torch::autograd::Function support functorch,https://github.com/pytorch/pytorch/issues/132052,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: functorch,scan when time allows,2024-07-29T16:27:01Z
issue,132013,failed to build v2.3.0 docker image,https://github.com/pytorch/pytorch/issues/132013,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: docker",scan when time allows,2024-07-29T16:23:24Z
issue,132024,Refactor serialization IO infrastructure and support HDFS/HTTP,https://github.com/pytorch/pytorch/issues/132024,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: serialization,scan when time allows,2024-07-29T14:52:29Z
issue,131931,Start processes in _MultiProcessingDataLoaderIter on a ThreadPool,https://github.com/pytorch/pytorch/issues/131931,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dataloader, module: performance",scan when time allows,2024-07-29T04:31:04Z
issue,131967,Layer normalization on Nested Tensor ragged dimension fails when `lengths is not None`,https://github.com/pytorch/pytorch/issues/131967,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nestedtensor,scan when time allows,2024-07-29T04:16:45Z
issue,111806,Revisit security implications of #31875,https://github.com/pytorch/pytorch/issues/111806,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2024-07-29T00:39:06Z
issue,105982,"vmap and rnn/lstm ""accessing '.data' under vmap transform is not allowed""",https://github.com/pytorch/pytorch/issues/105982,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: functorch,scan when time allows,2024-07-28T13:52:30Z
issue,32167,Support in-place pinning of memory,https://github.com/pytorch/pytorch/issues/32167,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: serialization,scan when time allows,2024-07-27T19:05:15Z
issue,131962,Distributed tests failing on Amazon2023 AMI,https://github.com/pytorch/pytorch/issues/131962,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2024-07-26T22:45:40Z
issue,39357,Enable scripting modules with recursive functions,https://github.com/pytorch/pytorch/issues/39357,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2024-07-26T10:00:36Z
issue,115819,Allow `input` tensor to have arbitrary tailing extra dimensions than `index` tensor in `torch.gather()` / `torch.scatter()`,https://github.com/pytorch/pytorch/issues/115819,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: scatter & gather ops,scan when time allows,2024-07-26T02:53:02Z
issue,123555,torch.onnx.errors.UnsupportedOperatorError: Exporting the operator 'quantized::linear_dynamic' to ONNX opset version 17 is not supported. Please feel free to request support or submit a pull request on PyTorch GitHub: https://github.com/pytorch/pytorch/issues,https://github.com/pytorch/pytorch/issues/123555,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2024-07-25T17:17:05Z
issue,131772,Default argument removal for python dispatch fails for pin_memory on factory functions,https://github.com/pytorch/pytorch/issues/131772,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: __torch_dispatch__, module: library",scan when time allows,2024-07-25T16:25:29Z
issue,131755,CUDA error: an illegal memory access was encountered when use TORCH_NCCL_NAN_CHECK.,https://github.com/pytorch/pytorch/issues/131755,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: c10d, oncall: distributed",scan when time allows,2024-07-25T15:34:48Z
issue,131736,MPS gives incorrect result when torch.nn.functional.softplus follows moveaxis,https://github.com/pytorch/pytorch/issues/131736,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: mps",scan when time allows,2024-07-25T14:21:20Z
issue,74935,[RFC]A suggestion of channels last memory format implementation for 3D tensor,https://github.com/pytorch/pytorch/issues/74935,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: memory format,scan when time allows,2024-07-25T13:13:52Z
issue,131695,Keep pytorch/pytorch-canary's main branch in sync with pytorch/pytorch's viable strict,https://github.com/pytorch/pytorch/issues/131695,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2024-07-25T11:52:53Z
issue,88968,M1 runner i-090e1df32b6f48a20 run out of disk space,https://github.com/pytorch/pytorch/issues/88968,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: arm, module: ci, module: macos",scan when time allows,2024-07-25T11:28:06Z
issue,97188,Building LibTorch on Ubuntu with Mac M1,https://github.com/pytorch/pytorch/issues/97188,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: arm, module: build, module: macos, module: torchbind",scan when time allows,2024-07-25T11:27:54Z
issue,117094,LBFGS MPS test_can_load_older_state_dict fails on older MacOS versions,https://github.com/pytorch/pytorch/issues/117094,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: macos, module: mps, module: optimizer, module: tests",scan when time allows,2024-07-25T07:00:04Z
issue,36650,state_dict and load_state_dict methods for DataLoader and Sampler to continue training at specific epoch and batch,https://github.com/pytorch/pytorch/issues/36650,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2024-07-25T06:56:11Z
issue,114962,`CUDAExtension` no longer works with `ccache`,https://github.com/pytorch/pytorch/issues/114962,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp-extensions, module: cuda, module: third_party",scan when time allows,2024-07-25T06:25:20Z
issue,18173,[docs] How to achieve high-order derivation in my .cpp?,https://github.com/pytorch/pytorch/issues/18173,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: cpp, module: docs",scan when time allows,2024-07-24T20:52:11Z
issue,19453,How to load PyTorch model with LSTM using C++ api,https://github.com/pytorch/pytorch/issues/19453,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp,scan when time allows,2024-07-24T20:52:00Z
issue,7457,[feature request] torch.nn.DataParallel should work nicely both for cpu and gpu devices,https://github.com/pytorch/pytorch/issues/7457,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: data parallel,scan when time allows,2024-07-24T20:49:48Z
issue,34375,.detach() behaves differently for dense tensors vs sparse tensors,https://github.com/pytorch/pytorch/issues/34375,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2024-07-24T20:48:14Z
issue,17984,"""unknown builtin op"" error with static library",https://github.com/pytorch/pytorch/issues/17984,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp,scan when time allows,2024-07-24T20:46:31Z
issue,22122,ASSERT FAILED at /pytorch/torch/csrc/autograd/generated/VariableType_0.cpp:12721,https://github.com/pytorch/pytorch/issues/22122,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: assert failure,scan when time allows,2024-07-24T20:46:18Z
issue,131656,Remove CTCLoss limitation of maximum target sequence length 256 (to satisify CuDNN),https://github.com/pytorch/pytorch/issues/131656,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: cudnn",scan when time allows,2024-07-24T18:41:30Z
issue,131655,tensor.share_memory_ causes temporarily increased memory consumption,https://github.com/pytorch/pytorch/issues/131655,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: memory usage",scan when time allows,2024-07-24T18:40:56Z
issue,131631,DataLoader worker exited unexpectedly,https://github.com/pytorch/pytorch/issues/131631,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2024-07-24T18:40:05Z
issue,101185,Pure virtual function call exception on Python interpreter exit when using debug wheel,https://github.com/pytorch/pytorch/issues/101185,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python frontend,scan when time allows,2024-07-24T04:35:12Z
issue,81856,Compatibility with newest MKL,https://github.com/pytorch/pytorch/issues/81856,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: mkl",scan when time allows,2024-07-24T01:28:01Z
issue,131274,The empty 0D or more D tensor with exceeding `dim`(`dim=100` or `dim=-100`) for `cummax()` works,https://github.com/pytorch/pytorch/issues/131274,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2024-07-23T22:13:26Z
issue,117705,Add GPU runners for Dynamo shards,https://github.com/pytorch/pytorch/issues/117705,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2024-07-23T21:59:48Z
issue,113854,Timer: Add Standard Deviation for Single Measurement,https://github.com/pytorch/pytorch/issues/113854,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: benchmark,scan when time allows,2024-07-23T21:45:54Z
issue,131491,`torch._assert_async` ignores `assert_msg`,https://github.com/pytorch/pytorch/issues/131491,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: error checking",scan when time allows,2024-07-23T21:45:03Z
issue,30723,Distributed Package asynchronous send/receive not working as expected (Gloo),https://github.com/pytorch/pytorch/issues/30723,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2024-07-23T21:34:24Z
issue,71203,Keys of a `ModuleDict` cannot have the same name as existing `ModuleDict` class attributes.,https://github.com/pytorch/pytorch/issues/71203,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-07-23T20:54:26Z
issue,101407,Delete old vmap prototype,https://github.com/pytorch/pytorch/issues/101407,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: functorch",scan when time allows,2024-07-23T15:54:09Z
issue,131273,The empty 0D or more D tensor with exceeding `dim`(`dim=100` or `dim=-100`) for `cummin()` works,https://github.com/pytorch/pytorch/issues/131273,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: edge cases, module: error checking",scan when time allows,2024-07-23T06:38:46Z
issue,131318,"RuntimeError: INTERNAL ASSERT FAILED at ""../aten/src/ATen/native/sparse/SparseCsrTensorMath.cpp"":1306",https://github.com/pytorch/pytorch/issues/131318,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2024-07-23T04:21:02Z
issue,131432,Remove all # mypy: ignore-errors,https://github.com/pytorch/pytorch/issues/131432,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: typing,scan when time allows,2024-07-23T04:12:44Z
issue,131349,[ONNX] opset 11 aten::index_put generating invalid add ops with None args,https://github.com/pytorch/pytorch/issues/131349,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2024-07-23T00:34:31Z
issue,131357,Follow ups from Increase (4x+) in build times #130744,https://github.com/pytorch/pytorch/issues/131357,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: devx,scan when time allows,2024-07-22T21:28:54Z
issue,130807,Generalizing cross product to hodge star operator,https://github.com/pytorch/pytorch/issues/130807,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python frontend,scan when time allows,2024-07-22T21:17:50Z
issue,130932,Error using torch.func transforms with DDP (broadcast_buffers=True),https://github.com/pytorch/pytorch/issues/130932,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ddp, module: functorch, oncall: distributed",scan when time allows,2024-07-22T20:47:57Z
issue,131110,Proposal for CUDA-Accelerated Dynamic Time Warping (DTW) Implementation in PyTorch,https://github.com/pytorch/pytorch/issues/131110,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-07-22T16:49:10Z
issue,74442,kron has unnecessary and undocumented dependence on memory layout,https://github.com/pytorch/pytorch/issues/74442,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: memory format,scan when time allows,2024-07-22T15:46:34Z
issue,33864,Memory leak in embedding layer and LSTM,https://github.com/pytorch/pytorch/issues/33864,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: memory usage,scan when time allows,2024-07-22T10:58:56Z
issue,122957,How to export torch.optim.LBFGS using torch.onnx.export,https://github.com/pytorch/pytorch/issues/122957,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: onnx, module: optimizer",scan when time allows,2024-07-22T09:48:29Z
issue,114723,[RFC] Intel GPU Upstreaming,https://github.com/pytorch/pytorch/issues/114723,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",scan when time allows,2024-07-22T07:36:04Z
issue,131113,pytorch dataloader has function that raises error,https://github.com/pytorch/pytorch/issues/131113,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: data, module: dataloader",scan when time allows,2024-07-19T20:53:03Z
issue,131062,Add some more testing around checkpoint over compile,https://github.com/pytorch/pytorch/issues/131062,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: activation checkpointing,scan when time allows,2024-07-19T20:51:18Z
issue,75166,Disable Python mode (torch dispatch mode) inside of mode-induced __torch_dispatch__call,https://github.com/pytorch/pytorch/issues/75166,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: __torch_dispatch__,scan when time allows,2024-07-19T18:01:22Z
issue,131213,Clean up coverage_ignore_functions,https://github.com/pytorch/pytorch/issues/131213,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ci, module: docs",scan when time allows,2024-07-19T17:53:10Z
issue,131148,"The empty 1D or more D tensor for `median()/nanmedian()` with the deepest `dim` gets errors, not getting NaNs",https://github.com/pytorch/pytorch/issues/131148,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: empty tensor, module: error checking",scan when time allows,2024-07-19T15:37:06Z
issue,44544,torch.distributed server back-connects to clients on random ports leading to firewall problems,https://github.com/pytorch/pytorch/issues/44544,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2024-07-19T15:25:32Z
issue,130610,dataloader hangs with fork and pin memory=True,https://github.com/pytorch/pytorch/issues/130610,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2024-07-19T06:45:50Z
issue,131011,xeon.run_cpu will not use numactl when whithin a docker container run with --cpuset-cpus non-0 numa node,https://github.com/pytorch/pytorch/issues/131011,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: performance",scan when time allows,2024-07-19T00:58:47Z
issue,128204,Documentation for DDP-related environment variables,https://github.com/pytorch/pytorch/issues/128204,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2024-07-18T21:16:16Z
issue,131054,Shape-aware type hinting,https://github.com/pytorch/pytorch/issues/131054,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: typing,scan when time allows,2024-07-18T19:45:00Z
issue,124395,Provide TORCH_LIBRARY m.def schema types documentation,https://github.com/pytorch/pytorch/issues/124395,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: library",scan when time allows,2024-07-18T17:06:29Z
issue,121641,torch.isin on GPU vs numpy isin: perf way slower than expected,https://github.com/pytorch/pytorch/issues/121641,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: performance",scan when time allows,2024-07-18T15:38:03Z
issue,58828,[Feature Pitch] Fast extremal eigensolvers,https://github.com/pytorch/pytorch/issues/58828,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: linear algebra, module: sparse",scan when time allows,2024-07-18T13:10:38Z
issue,129553,RVV support in PyTorch,https://github.com/pytorch/pytorch/issues/129553,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: performance",scan when time allows,2024-07-18T06:59:11Z
issue,130792,`with torch.device()` modifies the `as_tensor` intended behavior,https://github.com/pytorch/pytorch/issues/130792,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2024-07-18T00:53:46Z
issue,130985,Runtime error when running huggingface pretrained model with torch-xla,https://github.com/pytorch/pytorch/issues/130985,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: xla,scan when time allows,2024-07-18T00:30:27Z
issue,130847,Where does torch check for autocast op availability?,https://github.com/pytorch/pytorch/issues/130847,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision), module: docs",scan when time allows,2024-07-18T00:30:08Z
issue,130950,Improve detail / Create tutorial for torch.distributions,https://github.com/pytorch/pytorch/issues/130950,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: distributions, module: docs",scan when time allows,2024-07-18T00:25:19Z
issue,95460,torch.profiler.tensorboard_trace_handler Generates an incorrect JSON file,https://github.com/pytorch/pytorch/issues/95460,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: tensorboard, oncall: visualization",scan when time allows,2024-07-17T23:13:38Z
issue,130512,"The argument of `take()`, `take_along_dim()` and `gather()` should be `index` or `indices` for consistency.",https://github.com/pytorch/pytorch/issues/130512,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python frontend,scan when time allows,2024-07-17T21:58:57Z
issue,104697,LSTM built-in dropout not reproducible on GPU,https://github.com/pytorch/pytorch/issues/104697,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cudnn, module: determinism, module: nn, module: random",scan when time allows,2024-07-17T15:08:41Z
issue,128020,Make `torch.autograd.Function` support `vmap`,https://github.com/pytorch/pytorch/issues/128020,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: functorch",scan when time allows,2024-07-17T14:57:25Z
issue,91182,onednn(mkldnn) backend support for quantized operators,https://github.com/pytorch/pytorch/issues/91182,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: arm, oncall: quantization",scan when time allows,2024-07-17T13:27:29Z
issue,37250,Inconsistency between GPU memory usage in torch.cuda.memory_summary and nvidia-smi,https://github.com/pytorch/pytorch/issues/37250,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: memory usage",scan when time allows,2024-07-17T12:52:40Z
issue,50275,torch.transpose is divergent from np.transpose,https://github.com/pytorch/pytorch/issues/50275,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: deprecation, module: numpy",scan when time allows,2024-07-17T08:47:34Z
issue,130863,CMake cannot detect non-default fp16 installation path,https://github.com/pytorch/pytorch/issues/130863,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2024-07-17T07:06:12Z
issue,67011,Importing numpy interacts with `tensor.sum` perf,https://github.com/pytorch/pytorch/issues/67011,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: numpy, module: performance",scan when time allows,2024-07-17T06:38:07Z
issue,130693,The doc of `vstack()` should say `tuple` or `list` of tensors for `tensors` argument,https://github.com/pytorch/pytorch/issues/130693,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python frontend,scan when time allows,2024-07-16T20:36:32Z
issue,130621,[Bug Report] Choosing the network interface to use,https://github.com/pytorch/pytorch/issues/130621,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2024-07-16T02:31:47Z
issue,130768,Migrate gradcheck with check_batched_grad=True to use the new torch.vmap,https://github.com/pytorch/pytorch/issues/130768,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: vmap,scan when time allows,2024-07-16T00:21:10Z
issue,130739,Ineffective Buffer Size Adjustment in Multi-Node NCCL Communication,https://github.com/pytorch/pytorch/issues/130739,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nccl,scan when time allows,2024-07-16T00:19:19Z
issue,130701,fx.wrap() doesn't really work for things in torch/*,https://github.com/pytorch/pytorch/issues/130701,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2024-07-15T23:45:22Z
issue,130656,[torchelastic] torchrun with static backend hangs when used with port '0',https://github.com/pytorch/pytorch/issues/130656,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: elastic,scan when time allows,2024-07-15T23:41:32Z
issue,130342,Rescale to zero torch.nn.CrossEntropyLoss with label_smoothing > 0,https://github.com/pytorch/pytorch/issues/130342,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-07-15T21:05:20Z
issue,113948,Unknown CUDA Architecture Name 9.0a in CUDA_SELECT_NVCC_ARCH_FLAGS (compiling from source),https://github.com/pytorch/pytorch/issues/113948,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cuda",scan when time allows,2024-07-15T19:29:06Z
issue,129477,The ProcessGroupNCCL is not being destructed,https://github.com/pytorch/pytorch/issues/129477,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2024-07-15T19:12:56Z
issue,130636,`dims` of `permute()` without `tuple` or with `list` works agaist the doc,https://github.com/pytorch/pytorch/issues/130636,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2024-07-13T01:44:43Z
issue,130430,Error occurred when executing KolorsSampler:,https://github.com/pytorch/pytorch/issues/130430,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: mps,scan when time allows,2024-07-12T23:29:57Z
issue,130608,Setting `shape` and `dtype` without `dtype=` to `view()` works against error message but an `int` tensor is returned instead of a `float` tensor,https://github.com/pytorch/pytorch/issues/130608,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: error checking, module: python frontend",scan when time allows,2024-07-12T17:47:51Z
issue,130524,Unable to compile pytorch with conda python3.8 and cuda 11.4 in Nvidia AGX Jetson Orin aarch64,https://github.com/pytorch/pytorch/issues/130524,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: arm, module: build, module: cuda",scan when time allows,2024-07-12T17:35:51Z
issue,130529,Support tuple of `int`s in `torch.unsqueeze`,https://github.com/pytorch/pytorch/issues/130529,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python frontend,scan when time allows,2024-07-12T16:07:38Z
issue,130514,C++ API Does Not Expose Lazy NVRTC,https://github.com/pytorch/pytorch/issues/130514,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: cuda",scan when time allows,2024-07-12T16:04:22Z
issue,130525,[Feature request] Extend index_reduce to allow `index` to be of type List[List[int]],https://github.com/pytorch/pytorch/issues/130525,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: advanced indexing, module: python frontend",scan when time allows,2024-07-12T16:03:54Z
issue,130538,"nested_tensor LayerNorm fails with shape (*,*,j1,*) (but works with (*,j1,*,*))",https://github.com/pytorch/pytorch/issues/130538,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nestedtensor,scan when time allows,2024-07-12T16:02:36Z
issue,130448,`n` argument of `randperm()` with a 0D or more D tensor works against the doc,https://github.com/pytorch/pytorch/issues/130448,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: python frontend",scan when time allows,2024-07-12T15:52:39Z
issue,128497,"[bug]When the variable in the dataset is of type cuda, calling the dataloader will result in 0",https://github.com/pytorch/pytorch/issues/128497,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: data, module: dataloader",scan when time allows,2024-07-12T10:03:12Z
issue,129356,Does it still make sense to prevent saving multiple tensors that have a shared storage but different dtypes?,https://github.com/pytorch/pytorch/issues/129356,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: serialization,scan when time allows,2024-07-12T07:43:32Z
issue,113053,[CUDA-12.2] cuSPARSE deprecated support for sparse BSR,https://github.com/pytorch/pytorch/issues/113053,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: sparse",scan when time allows,2024-07-11T22:00:30Z
issue,30702,[FR] multidim squeeze and flatten,https://github.com/pytorch/pytorch/issues/30702,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: viewing and reshaping,scan when time allows,2024-07-11T20:53:42Z
issue,130055,[pipelining] Non-first stage only takes tensors,https://github.com/pytorch/pytorch/issues/130055,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: pipelining,scan when time allows,2024-07-11T16:58:24Z
issue,130464,[dynamo] Investigate unexpected success dynamo-wrapped tests from https://github.com/pytorch/pytorch/pull/129185,https://github.com/pytorch/pytorch/issues/130464,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",scan when time allows,2024-07-11T12:23:26Z
issue,32034,Extend `ConcatDataset` to return dataset index,https://github.com/pytorch/pytorch/issues/32034,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2024-07-10T21:46:56Z
issue,113329,Mark aten.normal as a core aten op,https://github.com/pytorch/pytorch/issues/113329,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: decompositions, oncall: pt2",scan when time allows,2024-07-10T19:20:08Z
issue,129625,Triton segfault with compiled flexattention kernel,https://github.com/pytorch/pytorch/issues/129625,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2024-07-10T18:02:05Z
issue,114353,add support for linalg lstsq in dynamo,https://github.com/pytorch/pytorch/issues/114353,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",scan when time allows,2024-07-10T16:08:51Z
issue,130432,Deferred runtime asserts triggered during backwards should not trigger assert not self.runtime_asserts_frozen,https://github.com/pytorch/pytorch/issues/130432,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dynamic shapes,scan when time allows,2024-07-10T14:33:22Z
issue,130124,Pin memory in subprocess,https://github.com/pytorch/pytorch/issues/130124,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: dataloader",scan when time allows,2024-07-10T06:31:55Z
issue,129329,_foreach_addc_,https://github.com/pytorch/pytorch/issues/129329,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: mta, module: optimizer",scan when time allows,2024-07-10T01:21:26Z
issue,129672,Distributed training with NCCL hangs on init_process_group when debug is DETAIL,https://github.com/pytorch/pytorch/issues/129672,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2024-07-09T18:28:59Z
issue,11372,Suggest: DataLoader add device parameter,https://github.com/pytorch/pytorch/issues/11372,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2024-07-09T14:34:17Z
issue,130254,"Many types for `start`, `end` and `base` argument of `logspace()` work againt what the doc says",https://github.com/pytorch/pytorch/issues/130254,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2024-07-09T13:51:05Z
issue,130128,Installation fails with conflicting dependencies (MarkupSafe>=2.0),https://github.com/pytorch/pytorch/issues/130128,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: binaries,scan when time allows,2024-07-09T13:44:44Z
issue,42791,Build fail on Ubuntu 18.04: caffe2/CMakeFiles/init_test.dir/build.make:106: recipe for target 'bin/init_test' failed,https://github.com/pytorch/pytorch/issues/42791,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2024-07-09T08:08:01Z
issue,130193,Inconsistency `CrossEntropyLoss` weights value between in gpu and in cpu.,https://github.com/pytorch/pytorch/issues/130193,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: nn",scan when time allows,2024-07-09T00:39:15Z
issue,129749,nccl deadlock,https://github.com/pytorch/pytorch/issues/129749,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2024-07-08T20:59:54Z
issue,129442,Volatile GPU-Util keeps 100% when using torch.distributed.isend/irecv,https://github.com/pytorch/pytorch/issues/129442,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2024-07-08T20:19:51Z
issue,121661,Look at gather + gemv performance.,https://github.com/pytorch/pytorch/issues/121661,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: performance, oncall: pt2",scan when time allows,2024-07-08T16:23:09Z
issue,117681,Improved workflow for tracing dynamism through complicated model,https://github.com/pytorch/pytorch/issues/117681,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamic shapes, oncall: pt2",scan when time allows,2024-07-08T16:19:26Z
issue,130078,cuda graph error for torch.isin if the input tensors are too large,https://github.com/pytorch/pytorch/issues/130078,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: cuda graphs",scan when time allows,2024-07-08T14:56:08Z
issue,130217,`unique()` is not intuitively understandable at all.,https://github.com/pytorch/pytorch/issues/130217,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2024-07-08T14:50:43Z
issue,130162,caching_allocator_delete does not free up the memory allocated by caching_allocator_alloc,https://github.com/pytorch/pytorch/issues/130162,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cudacachingallocator,scan when time allows,2024-07-08T14:44:33Z
issue,130096,The time consumption gradually increases when activation checkpoint is iterated and run multiple times,https://github.com/pytorch/pytorch/issues/130096,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: activation checkpointing,scan when time allows,2024-07-08T14:39:24Z
issue,111159,Wrong onnx model from `torch.onnx.export` when using `index_add_` function with duplicate `index` values.,https://github.com/pytorch/pytorch/issues/111159,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: advanced indexing, module: onnx",scan when time allows,2024-07-08T01:24:08Z
issue,127743,RuntimeError using nested tensor in Apple M1 device MPS,https://github.com/pytorch/pytorch/issues/127743,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: mps, module: nestedtensor",scan when time allows,2024-07-06T05:17:53Z
issue,129310,Upgrade dependencies MKL and Intel OpenMP to 2024.2.0,https://github.com/pytorch/pytorch/issues/129310,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: openmp, module: windows",scan when time allows,2024-07-05T21:10:32Z
issue,109213,inductor/test_max_autotune having timeout issues,https://github.com/pytorch/pytorch/issues/109213,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, module: tests, oncall: pt2",scan when time allows,2024-07-05T20:32:07Z
issue,110448,Explore Hybrid (CPU+GPU) Graphs in Scalar parameters,https://github.com/pytorch/pytorch/issues/110448,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",scan when time allows,2024-07-05T19:54:47Z
issue,111033,`CapabilityBasedPartitioner` returns invalid partitions.,https://github.com/pytorch/pytorch/issues/111033,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: xla, oncall: fx",scan when time allows,2024-07-05T19:39:14Z
issue,124270,PyTorch build fail with GCC 13.2.0 due to aten/src/ATen/core/IListRef_inl.h:171:17: error: possibly dangling reference to a temporary [-Werror=dangling-reference],https://github.com/pytorch/pytorch/issues/124270,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2024-07-05T07:51:45Z
issue,126980,Long build compilation time (>5 hours) for CUDA ARM build,https://github.com/pytorch/pytorch/issues/126980,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: arm, module: build, module: cuda",scan when time allows,2024-07-03T21:55:25Z
issue,130058,[pipelining] Expose a transform function to inject additional metadata into model forward,https://github.com/pytorch/pytorch/issues/130058,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: pipelining,scan when time allows,2024-07-03T20:48:24Z
issue,24139,Allow incompatible shapes in load_state_dict(strict=False),https://github.com/pytorch/pytorch/issues/24139,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: serialization,scan when time allows,2024-07-03T17:37:16Z
issue,130021,[Functionalization] The inplace decompose op is meaningless,https://github.com/pytorch/pytorch/issues/130021,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: functionalization, module: xla",scan when time allows,2024-07-03T14:20:55Z
issue,40859,Option to allow loading state dict with mismatching shapes.,https://github.com/pytorch/pytorch/issues/40859,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: serialization",scan when time allows,2024-07-03T10:24:49Z
issue,122568,How do I use exec in a Pytorch Module and train with multiple GPUs?,https://github.com/pytorch/pytorch/issues/122568,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: data parallel, oncall: distributed",scan when time allows,2024-07-03T10:12:51Z
issue,129954,"RuntimeError: NVML_SUCCESS == DriverAPI::get()->nvmlInit_v2_() INTERNAL ASSERT FAILED at ""../c10/cuda/CUDACachingAllocator.cpp"":799",https://github.com/pytorch/pytorch/issues/129954,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: cudacachingallocator",scan when time allows,2024-07-03T00:03:40Z
issue,128735,Flaky test page should include retry runs,https://github.com/pytorch/pytorch/issues/128735,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: infra,scan when time allows,2024-07-02T21:56:26Z
issue,129670,Libtorch does not support SM90,https://github.com/pytorch/pytorch/issues/129670,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2024-07-02T20:01:56Z
issue,10396,"[feature request] Provide a way to redirect shared memory prefix ""/torch_""",https://github.com/pytorch/pytorch/issues/10396,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2024-07-02T17:30:55Z
issue,129784,`@pytorchbot merge -i` not ignoring currently failing checks,https://github.com/pytorch/pytorch/issues/129784,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2024-07-01T21:31:10Z
issue,50010,torch.var and torch.std are not compatible with np.var and np.std,https://github.com/pytorch/pytorch/issues/50010,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: deprecation, module: numpy, module: reductions",scan when time allows,2024-07-01T18:52:10Z
issue,106544,"Branch name in double quotes """"",https://github.com/pytorch/pytorch/issues/106544,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2024-07-01T18:23:24Z
issue,124778,Dense-sparse broadcasted multiplication fails in the backward pass,https://github.com/pytorch/pytorch/issues/124778,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2024-07-01T17:42:26Z
issue,112831,dynamo_export successfully export model but fails at onnx.checker.check_model,https://github.com/pytorch/pytorch/issues/112831,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2024-07-01T15:31:09Z
issue,129293,forward_ad ignores checkpoints,https://github.com/pytorch/pytorch/issues/129293,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: activation checkpointing, module: autograd, module: bc-breaking, module: forward ad",scan when time allows,2024-07-01T05:02:11Z
issue,118499,Autocast on cpu dramatically slow,https://github.com/pytorch/pytorch/issues/118499,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpu,scan when time allows,2024-07-01T03:59:35Z
issue,128053,Handling reasoning about rationals in symbolic shapes,https://github.com/pytorch/pytorch/issues/128053,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dynamic shapes,scan when time allows,2024-06-30T13:00:27Z
issue,129224,"_If pin_memory_thread is alive but dataqueue is empty, it will fall into a dead loop,",https://github.com/pytorch/pytorch/issues/129224,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: data, module: dataloader",scan when time allows,2024-06-28T15:54:03Z
issue,109292,aten::squeeze exported to ONNX as an `If` node,https://github.com/pytorch/pytorch/issues/109292,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2024-06-28T12:59:49Z
issue,123062,Enable UFMT on all files in PyTorch,https://github.com/pytorch/pytorch/issues/123062,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: lint,scan when time allows,2024-06-28T06:19:13Z
issue,126005,CUDNN not detected in official nightly devel image. We need to use `12.4.1` upstream images,https://github.com/pytorch/pytorch/issues/126005,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: cudnn",scan when time allows,2024-06-27T20:37:20Z
issue,102600,Support for activation checkpoint on demand in custom function,https://github.com/pytorch/pytorch/issues/102600,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: activation checkpointing,scan when time allows,2024-06-27T20:28:56Z
issue,126216,Allow Lazy Initialization for PipelineStage,https://github.com/pytorch/pytorch/issues/126216,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2024-06-27T19:18:05Z
issue,92310,[Releng] Windows AMI needs to be pinned for release,https://github.com/pytorch/pytorch/issues/92310,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: releng,scan when time allows,2024-06-27T17:45:26Z
issue,129447,"Pytorch of 2.3.0 version loads bool type mask in triton, encountering error when use tl.where.",https://github.com/pytorch/pytorch/issues/129447,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2024-06-27T15:17:08Z
issue,129564,Support for 5-D output image tensors in `Col2Im`.,https://github.com/pytorch/pytorch/issues/129564,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: python frontend",scan when time allows,2024-06-27T14:43:59Z
issue,116707,AttributeError: `Adam` object has no attribute `param_groups` with Python 3.11,https://github.com/pytorch/pytorch/issues/116707,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: optimizer, module: serialization",scan when time allows,2024-06-27T14:36:35Z
issue,59439,NaN values on torch.nn.functional.conv2d (aarch64),https://github.com/pytorch/pytorch/issues/59439,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: arm, module: convolution, module: jetson, module: nn",scan when time allows,2024-06-27T14:09:53Z
issue,116608,unstable test_GroupNorm_general_cpu in test/test_nn.py,https://github.com/pytorch/pytorch/issues/116608,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-06-27T02:16:49Z
issue,129225,Why need to transpose when collate a sequence data in dataloader?,https://github.com/pytorch/pytorch/issues/129225,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: data, module: dataloader",scan when time allows,2024-06-27T02:02:53Z
issue,126472,[NestedTensor] Graph breaks with SDPA + NT constructor,https://github.com/pytorch/pytorch/issues/126472,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nestedtensor,scan when time allows,2024-06-26T23:02:11Z
issue,129555,Libtorch 2.3.1 requires Glibc_2.28 and can't be used on ubuntu 18.,https://github.com/pytorch/pytorch/issues/129555,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: cpp, module: docs, module: intel",scan when time allows,2024-06-26T21:02:58Z
issue,112956,Optim.step() is significantly SLOW on MPS,https://github.com/pytorch/pytorch/issues/112956,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: mps, module: performance",scan when time allows,2024-06-26T20:58:52Z
issue,129565,ChainedScheduler fail on CosineAnnealingWarmRestarts,https://github.com/pytorch/pytorch/issues/129565,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: lrscheduler, module: optimizer",scan when time allows,2024-06-26T14:47:56Z
issue,129410,"line-length of ruff, black and isort are not the same in pyproject.toml",https://github.com/pytorch/pytorch/issues/129410,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: lint,scan when time allows,2024-06-26T14:45:46Z
issue,129272,RecursionError for MaskedTensor.where,https://github.com/pytorch/pytorch/issues/129272,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: masked operators,scan when time allows,2024-06-26T14:38:17Z
issue,109873,Allow try except check for numpy bfloat16 representation,https://github.com/pytorch/pytorch/issues/109873,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: numpy,scan when time allows,2024-06-26T08:20:58Z
issue,122321,Batch size can't be varied if you export self attention to onnx,https://github.com/pytorch/pytorch/issues/122321,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: onnx, module: windows",scan when time allows,2024-06-26T06:43:27Z
issue,77165,Automate cleanup of header includes,https://github.com/pytorch/pytorch/issues/77165,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: internals,scan when time allows,2024-06-26T00:19:53Z
issue,72263,Removing deprecated `cpp_custom_type_hack`,https://github.com/pytorch/pytorch/issues/72263,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: oncall: jit, oncall: quantization",scan when time allows,2024-06-26T00:15:25Z
issue,46240,"Support ""symmetric"" reflection padding",https://github.com/pytorch/pytorch/issues/46240,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: numpy, module: padding, module: tensorflow",scan when time allows,2024-06-25T18:24:30Z
issue,120240,Cannot get deterministic Mask RCNN without running out of CUDA memory,https://github.com/pytorch/pytorch/issues/120240,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cublas, module: cuda, module: determinism",scan when time allows,2024-06-25T18:14:08Z
issue,71872,Clarify the behavior of DataLoader sampler and batch_sampler parameters,https://github.com/pytorch/pytorch/issues/71872,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dataloader, module: docs",scan when time allows,2024-06-25T16:55:39Z
issue,73479,Strange case of empty non-coalesced sparse tensor,https://github.com/pytorch/pytorch/issues/73479,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2024-06-25T11:52:42Z
issue,101191,custom_op API follow-ups,https://github.com/pytorch/pytorch/issues/101191,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2024-06-24T18:20:30Z
issue,129330,Fuyou Training Framework Integration for PyTorch,https://github.com/pytorch/pytorch/issues/129330,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2024-06-24T18:05:10Z
issue,129386,Support a more flexible way to mark all CI jobs in a similar category as unstable,https://github.com/pytorch/pytorch/issues/129386,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ci, module: devx",scan when time allows,2024-06-24T16:46:55Z
issue,129336,pytorch-nightly export KINETO_USE_DAEMON=1 Cannot initialize CUDA without ATen_cuda library,https://github.com/pytorch/pytorch/issues/129336,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: binaries,scan when time allows,2024-06-24T15:55:33Z
issue,129027,Extract some public APIs from torch::cuda::initModule(module) to torch::initModule(),https://github.com/pytorch/pytorch/issues/129027,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision),scan when time allows,2024-06-24T08:44:10Z
issue,129222,ignore all target for CrossEntropyLoss 2d，return nan but 1d return 0,https://github.com/pytorch/pytorch/issues/129222,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: loss,scan when time allows,2024-06-24T07:48:17Z
issue,128963,RuntimeError: NCCL error: invalid usage when deploy LLM model by vllm. (torch version: 2.3.0+cu118),https://github.com/pytorch/pytorch/issues/128963,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2024-06-24T04:06:47Z
issue,128495,'make html' will print 'duplicate object description' warnings when there are 1~5 CPUs in the running machine,https://github.com/pytorch/pytorch/issues/128495,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2024-06-22T18:13:11Z
issue,112089,"Failed to compile: null in call to `__builtin_memmove(__result, __first, sizeof(_Tp) * _Num);` Debian 12, ppc64le, gcc 12.2",https://github.com/pytorch/pytorch/issues/112089,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2024-06-22T15:28:13Z
issue,128960,[Dynamo x torch_function] magic methods and methods with regular Tensors don't seem to work,https://github.com/pytorch/pytorch/issues/128960,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2024-06-22T00:17:20Z
issue,15167,cudnn not found,https://github.com/pytorch/pytorch/issues/15167,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cudnn",scan when time allows,2024-06-21T23:28:26Z
issue,129191,RuntimeError when using torch.ops.aten._jagged_to_padded_dense_forward with large jagged tensors,https://github.com/pytorch/pytorch/issues/129191,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nestedtensor, module: sparse",scan when time allows,2024-06-21T19:12:11Z
issue,109445,Memory usage steadily increasing when using back propagation with sparse CSR parameter matrices on CPU,https://github.com/pytorch/pytorch/issues/109445,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: memory usage, module: sparse",scan when time allows,2024-06-21T18:04:45Z
issue,129050,Add comment for label_smoothing parameter in torch.nn.CrossEntropyLoss,https://github.com/pytorch/pytorch/issues/129050,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: nn",scan when time allows,2024-06-21T14:36:55Z
issue,129085,Adding betainc,https://github.com/pytorch/pytorch/issues/129085,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: numpy, module: special",scan when time allows,2024-06-21T13:39:38Z
issue,128665,[RFC][pipelining] PipelineStage should let user control send/recv endpoints,https://github.com/pytorch/pytorch/issues/128665,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: pipelining, oncall: distributed",scan when time allows,2024-06-20T17:19:46Z
issue,23854,Fan out calculation broken for group (depthwise) convolution,https://github.com/pytorch/pytorch/issues/23854,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: initialization",scan when time allows,2024-06-20T13:33:28Z
issue,128706,does FSDP support AMSP (a new DP shard strategy),https://github.com/pytorch/pytorch/issues/128706,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2024-06-20T12:19:54Z
issue,73909,SummaryWriter reports encoding error,https://github.com/pytorch/pytorch/issues/73909,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tensorboard,scan when time allows,2024-06-20T08:27:01Z
issue,117130,`torch.cuda.memory_summary()` can give `KeyError`,https://github.com/pytorch/pytorch/issues/117130,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2024-06-19T20:35:55Z
issue,128988,Look up tensor device member inside Tensor is_pinned() implementation instead of accepting an outside input,https://github.com/pytorch/pytorch/issues/128988,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2024-06-19T16:40:45Z
issue,128952,Question about torch.lstsq and torch.linalg.lstsq,https://github.com/pytorch/pytorch/issues/128952,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: linear algebra,scan when time allows,2024-06-19T12:05:43Z
issue,128936,Performance degradation for certain input using Conv2D,https://github.com/pytorch/pytorch/issues/128936,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: cuda, module: performance",scan when time allows,2024-06-19T11:51:52Z
issue,13058,Backward pass over torch.nn.functional.pad is extremely slow with half tensors,https://github.com/pytorch/pytorch/issues/13058,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: half, module: performance",scan when time allows,2024-06-19T11:17:36Z
issue,128821,Questions about TCPStoreLibUV,https://github.com/pytorch/pytorch/issues/128821,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2024-06-19T05:51:53Z
issue,122711,`torch.sparse.sum` does not support boolean and int when summing over dense dimensions,https://github.com/pytorch/pytorch/issues/122711,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2024-06-18T21:38:05Z
issue,102878,[dynamo] Diffusers - Graph break on OrderedDict,https://github.com/pytorch/pytorch/issues/102878,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dynamo,scan when time allows,2024-06-18T18:50:06Z
issue,118227,TorchDynamo mistranslates end of tensor slice,https://github.com/pytorch/pytorch/issues/118227,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: decompositions, oncall: pt2",scan when time allows,2024-06-18T17:05:40Z
issue,128959,caffe2 removal,https://github.com/pytorch/pytorch/issues/128959,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2024-06-18T15:51:46Z
issue,128791,CUDA error in torch.cdist with compute_mode=donot_use_mm_for_euclid_dist,https://github.com/pytorch/pytorch/issues/128791,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: loss",scan when time allows,2024-06-18T14:08:13Z
issue,126360,Dynamo ignores frame using yield,https://github.com/pytorch/pytorch/issues/126360,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",scan when time allows,2024-06-18T05:57:08Z
issue,122092,masked_index_add,https://github.com/pytorch/pytorch/issues/122092,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: cuda graphs, module: performance, module: python frontend",scan when time allows,2024-06-17T23:37:15Z
issue,128696,[DDP] DDP bucket memory release during fwd step,https://github.com/pytorch/pytorch/issues/128696,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2024-06-17T21:01:43Z
issue,128699,NotImplementedError: Operator aten.native_layer_norm_backward.default does not have a sharding strategy registered.,https://github.com/pytorch/pytorch/issues/128699,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2024-06-17T19:36:21Z
issue,124195,torch.save and torch.load is slow. Slower than numpy. Slower even than pickle.,https://github.com/pytorch/pytorch/issues/124195,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: serialization,scan when time allows,2024-06-17T18:00:54Z
issue,128778,False INTERNAL ASSERT FAILED bug whilst training Neural Network,https://github.com/pytorch/pytorch/issues/128778,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: linear algebra,scan when time allows,2024-06-17T15:15:44Z
issue,53785,Support for one-hot of dtypes besides torch.int64,https://github.com/pytorch/pytorch/issues/53785,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-06-17T14:57:31Z
issue,124083,Automatically bind all DispatchKey to Python,https://github.com/pytorch/pytorch/issues/124083,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dispatch,scan when time allows,2024-06-17T14:37:24Z
issue,128703,autograd with `is_grads_batched=True` fails on GroupNorm,https://github.com/pytorch/pytorch/issues/128703,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: functorch, module: norms and normalization, module: vmap",scan when time allows,2024-06-17T14:27:19Z
issue,128600,jacrev and jacfwd yield different results if one uses torch.no_grad blocks in module,https://github.com/pytorch/pytorch/issues/128600,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: docs, module: functorch",scan when time allows,2024-06-17T14:13:39Z
issue,76383,Suggestion to throw a UserWarning when a user forgot .eval() mode during inference,https://github.com/pytorch/pytorch/issues/76383,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: molly-guard, module: nn",scan when time allows,2024-06-16T18:01:58Z
issue,127676,module 'torch.mps' has no attribute 'device',https://github.com/pytorch/pytorch/issues/127676,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: mps,scan when time allows,2024-06-15T00:13:19Z
issue,110074,ImportError: libc10_cuda.so: cannot open shared object file: No such file or directory,https://github.com/pytorch/pytorch/issues/110074,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: build",scan when time allows,2024-06-14T23:45:18Z
issue,128165,`swap_tensors` fail when calling `nn.Module.to` on XLA `DDP` wrapped models.,https://github.com/pytorch/pytorch/issues/128165,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: xla",scan when time allows,2024-06-14T22:19:20Z
issue,128493,Inconsistency `CrossEntropyLoss` vs `BCELoss` regarding logits/probability space,https://github.com/pytorch/pytorch/issues/128493,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-06-14T18:59:06Z
issue,119557,Python 3.12 CPU Build system cannot find MKL libraries,https://github.com/pytorch/pytorch/issues/119557,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: ci, module: mkl",scan when time allows,2024-06-14T17:36:32Z
issue,126286,Adopt inductor-perf-test-nightly.yml to be used for release testing,https://github.com/pytorch/pytorch/issues/126286,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: binaries,scan when time allows,2024-06-14T15:17:37Z
issue,88148,☂️ Issues that trigger crashes due to corner-case API usages,https://github.com/pytorch/pytorch/issues/88148,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: edge cases,scan when time allows,2024-06-14T13:44:39Z
issue,107087,RuntimeError with operations on torch.float8_e5m2 and torch.float_e4m3fn data types,https://github.com/pytorch/pytorch/issues/107087,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: floatx (formerly float8),scan when time allows,2024-06-14T09:59:57Z
issue,11982,[feature request] Publish wheels with debug symbols,https://github.com/pytorch/pytorch/issues/11982,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: binaries,scan when time allows,2024-06-14T07:54:20Z
issue,123238,Fused optimizer kernel for CPU,https://github.com/pytorch/pytorch/issues/123238,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2024-06-14T03:15:34Z
issue,128509,elements of STL get converted to an `IValue` and cannot be converted to an STL.,https://github.com/pytorch/pytorch/issues/128509,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: cpp-extensions",scan when time allows,2024-06-14T01:34:29Z
issue,128641,Backward is not supported by `MaskedTensor`'s `torch.maximum`,https://github.com/pytorch/pytorch/issues/128641,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: masked operators",scan when time allows,2024-06-14T01:31:37Z
issue,128548,[torch.compile] Llama2 failure using dynamic shapes with Torch 2.4 nightly,https://github.com/pytorch/pytorch/issues/128548,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamic shapes, oncall: pt2",scan when time allows,2024-06-13T18:56:04Z
issue,4073,Feature request: Correlation module,https://github.com/pytorch/pytorch/issues/4073,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2024-06-13T17:51:15Z
issue,128601,Retry on CUBLAS_STATUS_ALLOC_FAILED,https://github.com/pytorch/pytorch/issues/128601,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: cudacachingallocator",scan when time allows,2024-06-13T15:49:52Z
issue,124133,ZeroRedundancyOptimizer + AdamW Fused can't load state_dict back,https://github.com/pytorch/pytorch/issues/124133,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: optimizer, oncall: distributed",scan when time allows,2024-06-12T13:45:52Z
issue,45685,"[fx] symbolic trace ""is None"" and ""is not None"" checks",https://github.com/pytorch/pytorch/issues/45685,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2024-06-12T10:53:59Z
issue,31689,DataLoader does not consider default floating point type,https://github.com/pytorch/pytorch/issues/31689,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2024-06-12T10:51:45Z
issue,123113,"Please add instructions for ""uninstall compiled PyTorch from source"" in the readme.",https://github.com/pytorch/pytorch/issues/123113,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2024-06-11T22:02:21Z
issue,93081,[Releng] Optimize PyTorch packaging,https://github.com/pytorch/pytorch/issues/93081,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: releng,scan when time allows,2024-06-11T21:57:16Z
issue,128257,can not create custom gradient when fp8 type is used,https://github.com/pytorch/pytorch/issues/128257,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: floatx (formerly float8),scan when time allows,2024-06-11T20:55:06Z
issue,118570,[Inductor] Kill Mutation Layout,https://github.com/pytorch/pytorch/issues/118570,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: inductor,scan when time allows,2024-06-11T19:46:16Z
issue,29863,n-dimensional non-constant padding functional,https://github.com/pytorch/pytorch/issues/29863,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: padding",scan when time allows,2024-06-11T15:47:50Z
issue,119740,Devcontainer for the nightly devenv,https://github.com/pytorch/pytorch/issues/119740,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2024-06-11T13:26:48Z
issue,127535,Can we have a Y-Split module for torch.nn.Sequential?,https://github.com/pytorch/pytorch/issues/127535,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-06-11T12:19:47Z
issue,50210,"Pytorch 1.7.1 hangs with multi-gpu, while Pytorch 1.6.0 works correctly",https://github.com/pytorch/pytorch/issues/50210,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: data parallel, module: deadlock",scan when time allows,2024-06-11T09:35:31Z
issue,89158,Support disallowing calls to certain instance methods in TorchDynamo,https://github.com/pytorch/pytorch/issues/89158,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dynamo,scan when time allows,2024-06-10T22:24:46Z
issue,128351,`Tensor.is_shared` behavior is incorrect for tensors created with `torch.from_file`,https://github.com/pytorch/pytorch/issues/128351,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: multiprocessing, module: serialization",scan when time allows,2024-06-10T21:53:24Z
issue,105790,torch.sparse.mm() with reduce operator for GPU support and COO matrices,https://github.com/pytorch/pytorch/issues/105790,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: reductions, module: sparse",scan when time allows,2024-06-10T20:51:38Z
issue,100626,Will Deep Implicit Models ever become first class citizens in PyTorch?,https://github.com/pytorch/pytorch/issues/100626,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: functorch",scan when time allows,2024-06-10T20:51:12Z
issue,128296,Proposal to add a CUDA backend for tensor conversion from BSR/BSC tensors to CSR/CSC tensors,https://github.com/pytorch/pytorch/issues/128296,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: sparse",scan when time allows,2024-06-10T20:30:29Z
issue,100241,Interaction of torch.no_grad and torch.autocast context managers with torch.compile,https://github.com/pytorch/pytorch/issues/100241,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: pt2,scan when time allows,2024-06-10T20:15:32Z
issue,128208,Adam is storing step as cpu tensor,https://github.com/pytorch/pytorch/issues/128208,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2024-06-10T16:19:24Z
issue,128314,`docker compose up` fails with `RuntimeError: Unexpected error from cudaGetDeviceCount()`,https://github.com/pytorch/pytorch/issues/128314,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cuda",scan when time allows,2024-06-10T16:18:49Z
issue,68258,Option to configure TorchElastic multiprocessing/api.py PContext.close() timeout,https://github.com/pytorch/pytorch/issues/68258,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2024-06-09T19:50:14Z
issue,104507,[error] while Implementation of pytorch DistributedParallel,https://github.com/pytorch/pytorch/issues/104507,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2024-06-08T18:29:00Z
issue,124906,Many cases in distributed/elastic/multiprocessing/redirects_test.py fails when use pytest,https://github.com/pytorch/pytorch/issues/124906,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: elastic, module: tests, oncall: distributed",scan when time allows,2024-06-07T15:32:52Z
issue,128033,Network module memory is not released in libtorch,https://github.com/pytorch/pytorch/issues/128033,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: cpp, module: memory usage",scan when time allows,2024-06-07T05:23:33Z
issue,52289,DataParallel module fails to handle data with size not divisible by number of GPUs,https://github.com/pytorch/pytorch/issues/52289,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: data parallel,scan when time allows,2024-06-07T02:24:44Z
issue,105460,Specify version,https://github.com/pytorch/pytorch/issues/105460,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2024-06-07T01:38:54Z
issue,128066,reduce_scatter_tensor with strided inputs produces corrupted results,https://github.com/pytorch/pytorch/issues/128066,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2024-06-06T23:32:09Z
issue,128128,Fake tensor support of distributed ops in FX graph tracing,https://github.com/pytorch/pytorch/issues/128128,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2024-06-06T23:24:39Z
issue,127964,[pipelining] Supporting skip connections in 1F1B schedules,https://github.com/pytorch/pytorch/issues/127964,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: pipelining,scan when time allows,2024-06-06T23:19:44Z
issue,116539,Auto bisect for broken periodic CI jobs,https://github.com/pytorch/pytorch/issues/116539,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: devx,scan when time allows,2024-06-06T17:50:43Z
issue,127950,Ensuring reliability of PT2 with TorchRec on PRs,https://github.com/pytorch/pytorch/issues/127950,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ci, oncall: pt2",scan when time allows,2024-06-06T16:57:35Z
issue,128044,Request for in-place FFT,https://github.com/pytorch/pytorch/issues/128044,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fft,scan when time allows,2024-06-06T15:56:42Z
issue,128126,Stateless random numbers,https://github.com/pytorch/pytorch/issues/128126,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: random,scan when time allows,2024-06-06T15:43:07Z
issue,126952,Add integrity check in torch.save,https://github.com/pytorch/pytorch/issues/126952,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: serialization,scan when time allows,2024-06-06T02:22:00Z
issue,117883,Flight Recorder - feature requests,https://github.com/pytorch/pytorch/issues/117883,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2024-06-06T00:06:57Z
issue,122931,Stack smashing detected on nccl initialization,https://github.com/pytorch/pytorch/issues/122931,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: c10d, oncall: distributed",scan when time allows,2024-06-05T22:58:21Z
issue,126008,Torch distribution blocking ports and failing to connect,https://github.com/pytorch/pytorch/issues/126008,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: c10d, module: elastic, oncall: distributed",scan when time allows,2024-06-05T22:50:32Z
issue,127873,[libtorch] `CppCommInterface` subclasses cannot be instantiated,https://github.com/pytorch/pytorch/issues/127873,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ddp, oncall: distributed",scan when time allows,2024-06-05T22:24:45Z
issue,126773,Give`wait_tensor()` a schema that reflects its side effect,https://github.com/pytorch/pytorch/issues/126773,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2024-06-05T22:11:54Z
issue,121207,can not find all the deps when install the native CUDA-toolkit and torch in Redhat-like systems,https://github.com/pytorch/pytorch/issues/121207,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cuda",scan when time allows,2024-06-04T11:36:44Z
issue,127750,aot_function cannot handle index_add,https://github.com/pytorch/pytorch/issues/127750,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: functorch,scan when time allows,2024-06-04T08:18:46Z
issue,107026,a bug about tensor stride,https://github.com/pytorch/pytorch/issues/107026,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: viewing and reshaping,scan when time allows,2024-06-04T05:25:08Z
issue,127788,ghstack and pytorch-bot can race,https://github.com/pytorch/pytorch/issues/127788,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: devx,scan when time allows,2024-06-03T20:16:34Z
issue,127754,profile not record at::meta_xxx time,https://github.com/pytorch/pytorch/issues/127754,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: meta tensors,scan when time allows,2024-06-03T18:00:45Z
issue,127733,torch.nn.Linear does not check the value of the out_features parameter,https://github.com/pytorch/pytorch/issues/127733,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-06-03T17:58:30Z
issue,127737,"`torch.tensor()` can also be used with a list of the `float`, `complex` or `bool` type tensors which have a single element each",https://github.com/pytorch/pytorch/issues/127737,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ux,scan when time allows,2024-06-03T17:56:47Z
issue,127732,torch.nn.ConvTranspose2d/torch.nn.ConvTranspose1d lacks a check for the parameter dilation in the model building process,https://github.com/pytorch/pytorch/issues/127732,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-06-03T17:54:50Z
issue,127734,nn.ConvTranspose2d does not check for parameter combinations resulting in convolution output with dimension 0,https://github.com/pytorch/pytorch/issues/127734,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-06-03T17:54:34Z
issue,127701,The type of exception thrown by torch.nn.Conv2d does not match the actual situation,https://github.com/pytorch/pytorch/issues/127701,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-06-03T17:52:24Z
issue,127606,` to be a quantized tensor. Is this likely due to missing support for quantized `onnx::Pad`,https://github.com/pytorch/pytorch/issues/127606,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2024-06-03T17:42:47Z
issue,32407,Build without MKL is not possible when MKL is installed,https://github.com/pytorch/pytorch/issues/32407,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: mkl",scan when time allows,2024-06-03T15:27:43Z
issue,127194,Should `torch.Size` convert np.ndarrays to lists of ints?,https://github.com/pytorch/pytorch/issues/127194,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: numpy,scan when time allows,2024-06-03T11:24:40Z
issue,119343,Use Dtensor and ColwiseParallel api to do Tensor Parallel for llama2 model but got communication error.,https://github.com/pytorch/pytorch/issues/119343,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dtensor, oncall: distributed",scan when time allows,2024-06-03T09:04:07Z
issue,127541,RuntimeError: derivative for aten::_spdiags is not implemented,https://github.com/pytorch/pytorch/issues/127541,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: sparse",scan when time allows,2024-06-03T07:55:39Z
issue,93188,Why is AvgPool2D taking longer than Conv2D for the same input?,https://github.com/pytorch/pytorch/issues/93188,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: performance",scan when time allows,2024-06-03T06:52:36Z
issue,127072,PyTorch DataLoader improvements for Iterable Dataset,https://github.com/pytorch/pytorch/issues/127072,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2024-06-01T09:33:40Z
issue,107611,Previous version not found,https://github.com/pytorch/pytorch/issues/107611,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: releng,scan when time allows,2024-05-31T18:42:12Z
issue,126755,inplace parameter in dropouts should function as expected regardless of the value of training(or train) paramter,https://github.com/pytorch/pytorch/issues/126755,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-05-31T17:00:26Z
issue,91737,[MPS] Improve the performance of torch.linear(),https://github.com/pytorch/pytorch/issues/91737,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: backend, module: mps",scan when time allows,2024-05-31T09:50:57Z
issue,127530,Run rstcheck on modified docstrings and docs as additional linter,https://github.com/pytorch/pytorch/issues/127530,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: doc infra, module: docs",scan when time allows,2024-05-31T01:59:03Z
issue,127518,THPVariable_Check(list_elem) INTERNAL ASSERT FAILED,https://github.com/pytorch/pytorch/issues/127518,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2024-05-31T00:05:33Z
issue,69364,"[feature request] `quantized::linear_dynamic` on CUDA/eager, and other quantized and low-level int8 operators (matmul, gemm etc) on CUDA + integrate LLM.int8 + integrate ZeroQuant?",https://github.com/pytorch/pytorch/issues/69364,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, oncall: quantization",scan when time allows,2024-05-30T23:05:17Z
issue,96073,Pytorch 2.0 installation tutorial does not work under Macbook,https://github.com/pytorch/pytorch/issues/96073,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: macos",scan when time allows,2024-05-30T19:39:34Z
issue,127555,"Reserving this issue to test something, don't mind me",https://github.com/pytorch/pytorch/issues/127555,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2024-05-30T18:53:49Z
issue,127270,Not supporting higher protobuf versions,https://github.com/pytorch/pytorch/issues/127270,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: protobuf,scan when time allows,2024-05-30T16:57:05Z
issue,127497,torch.compiler docstrings can be derived from _dynamo and _inductor docs,https://github.com/pytorch/pytorch/issues/127497,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, oncall: pt2",scan when time allows,2024-05-30T16:53:59Z
issue,127422,index_select op not implemented on Vulkan backend,https://github.com/pytorch/pytorch/issues/127422,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: embedding, module: vulkan",scan when time allows,2024-05-30T16:14:58Z
issue,127221,lacking checking for ConvTranspose's parameters when running with GPUs,https://github.com/pytorch/pytorch/issues/127221,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: cuda, module: error checking",scan when time allows,2024-05-30T16:08:33Z
issue,127223,[BUG] torch.linalg.lstsq returning wrong result.,https://github.com/pytorch/pytorch/issues/127223,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: linear algebra,scan when time allows,2024-05-30T14:58:17Z
issue,127525,Multiple unhandled exceptions in weights_only unpickler,https://github.com/pytorch/pytorch/issues/127525,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: serialization,scan when time allows,2024-05-30T14:35:54Z
issue,127387,"Backwards pass through Beta distribution rsample gives inf for 4 < alpha - 2**16 < 1040, beta = 3/2",https://github.com/pytorch/pytorch/issues/127387,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distributions,scan when time allows,2024-05-30T13:42:51Z
issue,127375,Different tensor strides can result in surprisingly large discrepancies in Conv2d outputs,https://github.com/pytorch/pytorch/issues/127375,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: tf32",scan when time allows,2024-05-30T13:33:50Z
issue,48389,"Add parameter ""half_pixel_center =False"" to the Bilinear function",https://github.com/pytorch/pytorch/issues/48389,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: interpolation,scan when time allows,2024-05-30T11:46:19Z
issue,29398,[feature request] torch.kthvalue to support a new argument largest,https://github.com/pytorch/pytorch/issues/29398,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sorting and selection,scan when time allows,2024-05-30T01:36:46Z
issue,118507,RuntimeError in torch.istft with center=False: Window Overlap Add Issue,https://github.com/pytorch/pytorch/issues/118507,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fft,scan when time allows,2024-05-29T19:57:46Z
issue,49298,[question] How hard would it be to implement 4-bit precision training?,https://github.com/pytorch/pytorch/issues/49298,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: internals,scan when time allows,2024-05-29T19:02:17Z
issue,41571,"torch.cuda.BoolTensor uses 8 bits per element, not 1 bit as reported by element_size()",https://github.com/pytorch/pytorch/issues/41571,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2024-05-29T19:02:13Z
issue,125467,Inaccurate filename due to test class metaprogramming that doesn't set __file__ and __name__,https://github.com/pytorch/pytorch/issues/125467,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: devx, module: tests",scan when time allows,2024-05-29T16:48:00Z
issue,127196,torch.topk results differ on CPU and CUDA,https://github.com/pytorch/pytorch/issues/127196,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: sorting and selection",scan when time allows,2024-05-29T01:58:35Z
issue,126714,move label_to_label.yml to an issue?,https://github.com/pytorch/pytorch/issues/126714,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: devx,scan when time allows,2024-05-29T00:27:07Z
issue,127352,`flake8: noqa` disables flake8 linter for the whole file and it's not obvious,https://github.com/pytorch/pytorch/issues/127352,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: lint,scan when time allows,2024-05-29T00:05:57Z
issue,80411,Tracker: Slow gradcheck failures possibly indicating incorrect gradients,https://github.com/pytorch/pytorch/issues/80411,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2024-05-28T14:54:29Z
issue,126899,function signature for multiprocessing.spawn is multiprocessing.spawn.spawn,https://github.com/pytorch/pytorch/issues/126899,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2024-05-28T13:46:52Z
issue,127282,torch.fx.symbolic_trace doesn't support many Callable types,https://github.com/pytorch/pytorch/issues/127282,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2024-05-28T12:50:33Z
issue,127189,`store_param_remainders` from Apex DistributedFusedAdam,https://github.com/pytorch/pytorch/issues/127189,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: optimizer, oncall: distributed",scan when time allows,2024-05-28T12:00:24Z
issue,127164,randn generates different output for 4x4 tensor size sliced to match shape of direct 2x4 or 4x2 and compare output,https://github.com/pytorch/pytorch/issues/127164,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: random,scan when time allows,2024-05-28T11:59:57Z
issue,56360,Support for stereo audio data in from torch.utils.tensorboard.SummaryWriter,https://github.com/pytorch/pytorch/issues/56360,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: data, module: tensorboard",scan when time allows,2024-05-27T10:39:57Z
issue,22281,Box constraints for optimizers,https://github.com/pytorch/pytorch/issues/22281,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2024-05-27T09:36:08Z
issue,122068,"NVML_SUCCESS == r INTERNAL ASSERT FAILED at ""../c10/cuda/CUDACachingAllocator.cpp"":830, please report a bug to PyTorch.",https://github.com/pytorch/pytorch/issues/122068,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: cudacachingallocator",scan when time allows,2024-05-27T09:15:42Z
issue,127075,What is the processing principle when the complex64 input tensor contains nan or inf for addition?,https://github.com/pytorch/pytorch/issues/127075,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: complex,scan when time allows,2024-05-27T03:59:52Z
issue,101273,IPEX as TorchDynamo Backend Performance Dashboard,https://github.com/pytorch/pytorch/issues/101273,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: oncall: cpu inductor, oncall: pt2",scan when time allows,2024-05-27T01:33:33Z
issue,127031,Make grid_sample a bit more intuitive to use,https://github.com/pytorch/pytorch/issues/127031,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: nn",scan when time allows,2024-05-24T17:46:44Z
issue,127083,"Setting a `float`, `complex` or `bool` type value to `precision` argument of `set_printoptions()` has problem when creating a `float`, `complex` or `bool` type tensor",https://github.com/pytorch/pytorch/issues/127083,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python frontend,scan when time allows,2024-05-24T17:39:35Z
issue,126947,`dsplit()` with `indices_or_sections=` doesn't work while `dsplit()` without `indices_or_sections=` works,https://github.com/pytorch/pytorch/issues/126947,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python frontend,scan when time allows,2024-05-24T17:05:23Z
issue,127010,"`indices_or_sections` parameter of `tensor_split()` doesn't work while `sections`, `indices` or `tensor_indices_or_sections` parameter works",https://github.com/pytorch/pytorch/issues/127010,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python frontend,scan when time allows,2024-05-24T16:42:01Z
issue,127009,`indices_or_sections` parameter of `dsplit()` doesn't work while `sections` or `indices` parameter works,https://github.com/pytorch/pytorch/issues/127009,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python frontend,scan when time allows,2024-05-24T16:41:56Z
issue,127019,`indices_or_sections` parameter of `vsplit()` doesn't work while `sections` or `indices` parameter works,https://github.com/pytorch/pytorch/issues/127019,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python frontend,scan when time allows,2024-05-24T16:41:51Z
issue,127020,`indices_or_sections` parameter of `hsplit()` doesn't work while `sections` or `indices` parameter works,https://github.com/pytorch/pytorch/issues/127020,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python frontend,scan when time allows,2024-05-24T16:40:14Z
issue,63337,Property based testing like AFL,https://github.com/pytorch/pytorch/issues/63337,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tests,scan when time allows,2024-05-24T09:12:24Z
issue,63971,benchmark.Compare raises: TypeError: object of type 'NoneType' has no len(),https://github.com/pytorch/pytorch/issues/63971,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: benchmark, module: error checking",scan when time allows,2024-05-24T04:55:04Z
issue,77880,Make the appropriate backend `DimensionNode` visible to LTC core,https://github.com/pytorch/pytorch/issues/77880,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2024-05-24T00:10:50Z
issue,22573,Batched symeig and qr are very slow on GPU,https://github.com/pytorch/pytorch/issues/22573,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: linear algebra, module: performance",scan when time allows,2024-05-23T23:02:23Z
issue,126660,UnsupportedOperatorError: aten::scatter_reduce when include_self=False,https://github.com/pytorch/pytorch/issues/126660,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2024-05-23T22:34:51Z
issue,100914,[MPS] Track failures of test_module.py for MPS backend,https://github.com/pytorch/pytorch/issues/100914,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: backend, module: mps",scan when time allows,2024-05-23T20:25:30Z
issue,127001,Add ability to pass `MAP_NORESERVE` to`mmap(2)` call in `torch.serialization.set_default_mmap_options`,https://github.com/pytorch/pytorch/issues/127001,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: serialization,scan when time allows,2024-05-23T19:56:12Z
issue,126544,ncclCommWatchdog always terminates the process and prevents error handling if CUDA context is corrupted,https://github.com/pytorch/pytorch/issues/126544,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: c10d, oncall: distributed",scan when time allows,2024-05-23T17:26:03Z
issue,70540,cumcount / cumulative count,https://github.com/pytorch/pytorch/issues/70540,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2024-05-23T17:24:15Z
issue,52915,Broadcasting behaviour for linear algebra solvers,https://github.com/pytorch/pytorch/issues/52915,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: linear algebra,scan when time allows,2024-05-22T21:27:13Z
issue,124566,Can't download pytorch/ timeouterror,https://github.com/pytorch/pytorch/issues/124566,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: binaries,scan when time allows,2024-05-22T20:47:22Z
issue,126549,Error: Exporting the operator 'aten::searchsorted' to ONNX opset version 17,https://github.com/pytorch/pytorch/issues/126549,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2024-05-21T18:29:48Z
issue,126748,"""torch.geqrf"" performs differently on cpu and gpu",https://github.com/pytorch/pytorch/issues/126748,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: nans and infs",scan when time allows,2024-05-21T18:01:30Z
issue,126758,The unexpected behavior of `min()`,https://github.com/pytorch/pytorch/issues/126758,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: error checking,scan when time allows,2024-05-21T17:53:32Z
issue,126776,One complex number or boolean value of a 1D or more D tensor with `argmin()` and `dim=` works,https://github.com/pytorch/pytorch/issues/126776,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: edge cases,scan when time allows,2024-05-21T17:13:29Z
issue,126777,One complex number or boolean value of a 1D or more D tensor with `argmax()` and `dim=` works,https://github.com/pytorch/pytorch/issues/126777,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: edge cases,scan when time allows,2024-05-21T17:13:03Z
issue,125844,`WeightedRandomSampler` convert weights `dtype=torch.double` internally but `MPS` only support `dtype=torch.float32`,https://github.com/pytorch/pytorch/issues/125844,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dataloader, module: mps",scan when time allows,2024-05-21T15:20:15Z
issue,125234,torch.Library can easily cause segfault on loading/unloading,https://github.com/pytorch/pytorch/issues/125234,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: library,scan when time allows,2024-05-21T13:47:21Z
issue,126708,Inductor Channels-last slowdown,https://github.com/pytorch/pytorch/issues/126708,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",scan when time allows,2024-05-20T22:30:13Z
issue,126659,`topk()` cannot be used with `complex` or `bool` type of a 0D(only `cuda`) or more D tensor but `topk()` can be used with `complex` or `bool` type of a 0D(only `cpu`) tensor,https://github.com/pytorch/pytorch/issues/126659,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: edge cases,scan when time allows,2024-05-20T21:08:03Z
issue,126658,`kthvalue()` cannot be used with `complex` or `bool` type of a 1D or more D tensor but `kthvalue()` can be used with `complex` or `bool` type of a 0D tensor,https://github.com/pytorch/pytorch/issues/126658,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: edge cases,scan when time allows,2024-05-20T16:36:24Z
issue,33520,Reverse Cumulative Sum,https://github.com/pytorch/pytorch/issues/33520,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2024-05-20T15:31:04Z
issue,126404,Add `int` and `str` type to `device` parameter of `torch.tensor()`,https://github.com/pytorch/pytorch/issues/126404,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python frontend,scan when time allows,2024-05-20T15:23:15Z
issue,126386,lintrunner failures on main,https://github.com/pytorch/pytorch/issues/126386,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: lint,scan when time allows,2024-05-20T15:17:49Z
issue,126519,ambiguous overload for TORCH_LIBRARY `def`,https://github.com/pytorch/pytorch/issues/126519,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: internals,scan when time allows,2024-05-20T15:16:02Z
issue,126620,dataloader crashes after several epochs if the trained model contains triton-based operators,https://github.com/pytorch/pytorch/issues/126620,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2024-05-20T14:55:17Z
issue,126638,Test failures with new configs in test_grad_scaling_autocast in test_torch.py,https://github.com/pytorch/pytorch/issues/126638,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2024-05-20T12:34:21Z
issue,86162,torch.nn.functional.one_hot only works for int64,https://github.com/pytorch/pytorch/issues/86162,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-05-19T14:33:04Z
issue,32912,Optional seq_len argument to torch.nn.utils.rnn.pad_sequence,https://github.com/pytorch/pytorch/issues/32912,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: padding,scan when time allows,2024-05-18T13:32:36Z
issue,125893,dynamo_export fails to produce a graph from nn.GLU : no tensor operations were found,https://github.com/pytorch/pytorch/issues/125893,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2024-05-17T18:31:06Z
issue,42665,Optimizer support via Libtorch C++ on Android,https://github.com/pytorch/pytorch/issues/42665,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: optimizer, oncall: mobile",scan when time allows,2024-05-17T09:56:01Z
issue,126253,`dtype` of `type()` doesn't accept a string,https://github.com/pytorch/pytorch/issues/126253,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python frontend,scan when time allows,2024-05-17T09:11:35Z
issue,20551,torch.norm produces incorrect results,https://github.com/pytorch/pytorch/issues/20551,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: norms and normalization, module: numerical-reproducibility",scan when time allows,2024-05-17T08:35:36Z
issue,55207,[RFC] Model Sharding for distributed training,https://github.com/pytorch/pytorch/issues/55207,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2024-05-17T06:54:28Z
issue,105335,Enable SLEEF on ARM,https://github.com/pytorch/pytorch/issues/105335,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: arm, module: build, module: sleef",scan when time allows,2024-05-17T04:51:55Z
issue,125902,How to export onnx with fixed shape output ?,https://github.com/pytorch/pytorch/issues/125902,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2024-05-17T04:35:24Z
issue,126453,Add testing for NJT serialization,https://github.com/pytorch/pytorch/issues/126453,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nestedtensor, module: serialization",scan when time allows,2024-05-16T21:02:12Z
issue,125622,Wrapper subclass serialization should preserve all _make_wrapper_subclass() args,https://github.com/pytorch/pytorch/issues/125622,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: serialization,scan when time allows,2024-05-16T15:48:00Z
issue,117275,deepcopy of `nn.Parameter` does not transfer __dict__,https://github.com/pytorch/pytorch/issues/117275,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-05-16T13:47:43Z
issue,75662,"multiprocessing and torch.tensor, Cannot allocate memory error",https://github.com/pytorch/pytorch/issues/75662,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: multiprocessing,scan when time allows,2024-05-16T11:40:01Z
issue,126115,filtering on CPU slower in pytorch than scipy,https://github.com/pytorch/pytorch/issues/126115,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: performance",scan when time allows,2024-05-16T05:31:57Z
issue,126323,Move pytorch_sphinx_theme out of the container build,https://github.com/pytorch/pytorch/issues/126323,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ci, module: docs",scan when time allows,2024-05-16T00:58:43Z
issue,68944,Pytorch to ONNX model conversion,https://github.com/pytorch/pytorch/issues/68944,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2024-05-16T00:39:21Z
issue,98727,Pytorch member variable not working after converting to onnx format,https://github.com/pytorch/pytorch/issues/98727,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2024-05-16T00:35:34Z
issue,126347,Break the dependency between torch.nn and torch.distributed,https://github.com/pytorch/pytorch/issues/126347,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2024-05-15T22:52:35Z
issue,126244,"`dtype` of `torch.tensor()` also accepts `int()`, `float()` and `bool()` which are python built-in functions.",https://github.com/pytorch/pytorch/issues/126244,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python frontend,scan when time allows,2024-05-15T22:21:19Z
issue,126250,"`dtype` of `to()` also accepts `int()`, `float()` and `bool()` which are python built-in functions.",https://github.com/pytorch/pytorch/issues/126250,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python frontend,scan when time allows,2024-05-15T22:21:11Z
issue,126247,"`type1` and `type2` of `torch.promote_types()` also accept `int()`, `float()` and `bool()` which are python built-in functions.",https://github.com/pytorch/pytorch/issues/126247,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python frontend,scan when time allows,2024-05-15T22:21:02Z
issue,106485,Increasing batch size makes network forward 1000 times slower,https://github.com/pytorch/pytorch/issues/106485,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: cudnn, module: nn",scan when time allows,2024-05-15T22:19:42Z
issue,119368,"Support nn.Module.to(..., copy=True)",https://github.com/pytorch/pytorch/issues/119368,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-05-15T18:15:19Z
issue,82139,Re-enable DynamicQuantModule in iOS simulator tests,https://github.com/pytorch/pytorch/issues/82139,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ci, module: ios",scan when time allows,2024-05-15T14:43:14Z
issue,98006,Create a plan on removing conda dependency from CI/CD,https://github.com/pytorch/pytorch/issues/98006,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: releng,scan when time allows,2024-05-15T14:41:44Z
issue,85499,Execute smoke test for Better Transformer feature,https://github.com/pytorch/pytorch/issues/85499,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2024-05-15T14:32:28Z
issue,87782,Libtorch windows binaries publishing,https://github.com/pytorch/pytorch/issues/87782,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2024-05-15T14:32:17Z
issue,107780,Add caffe2 ideep/onednn tests to OSS CI,https://github.com/pytorch/pytorch/issues/107780,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ci, module: mkldnn, oncall: releng",scan when time allows,2024-05-15T14:31:19Z
issue,126021,Incorrect output from `torch.histc` when min and max are equal.,https://github.com/pytorch/pytorch/issues/126021,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python frontend,scan when time allows,2024-05-15T13:50:25Z
issue,126164,Context manager for determinism,https://github.com/pytorch/pytorch/issues/126164,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: determinism,scan when time allows,2024-05-15T12:31:19Z
issue,109017,torch.sparse_coo_tensor argname quirks + [feature request] `.numpy()`/`from_numpy` method for sparse_coo_tensor/sparse_csr_tensor (or maybe name them as `.scipy()`/`.from_scipy()` or at least under some `torch.utils.*` namespace for zero-copy conversions to<>from scipy sparse arrays,https://github.com/pytorch/pytorch/issues/109017,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: scipy compatibility, module: sparse",scan when time allows,2024-05-15T11:41:40Z
issue,105641,Turn indexing with a scalar tensor into an copy into a view and avoid a D2H synchronization.,https://github.com/pytorch/pytorch/issues/105641,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: advanced indexing, module: bc-breaking, module: numpy",scan when time allows,2024-05-14T21:29:03Z
issue,126006,The arguments `tesnor1` and `tensor2` of `result_type()` don't work but `tensor` and `other` work,https://github.com/pytorch/pytorch/issues/126006,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python frontend,scan when time allows,2024-05-14T18:26:09Z
issue,126014,The unexpected behavior of `can_cast()`,https://github.com/pytorch/pytorch/issues/126014,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python frontend,scan when time allows,2024-05-14T18:25:22Z
issue,126034,A scalar with `tensor` argument of `result_type()` doesn't work,https://github.com/pytorch/pytorch/issues/126034,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python frontend,scan when time allows,2024-05-14T18:21:33Z
issue,126041,A scalar with `input` argument of `remainder()` doesn't work,https://github.com/pytorch/pytorch/issues/126041,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python frontend,scan when time allows,2024-05-14T18:21:02Z
issue,126037,A scalar with `input` argument of `pow()` doesn't work,https://github.com/pytorch/pytorch/issues/126037,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python frontend,scan when time allows,2024-05-14T18:20:51Z
issue,126039,A scalar with `input` argument of `float_power()` doesn't work,https://github.com/pytorch/pytorch/issues/126039,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python frontend,scan when time allows,2024-05-14T18:20:37Z
issue,126190,Centralized Mechanism To Pytorch Query Global Flags,https://github.com/pytorch/pytorch/issues/126190,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: pt2,scan when time allows,2024-05-14T18:08:45Z
issue,126065,[ONNX] `SymbolicValueError` with `onnx::quantized_gather`,https://github.com/pytorch/pytorch/issues/126065,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2024-05-14T17:02:08Z
issue,126000,Inform when weights loading promotes types,https://github.com/pytorch/pytorch/issues/126000,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-05-14T16:54:17Z
issue,125940,[NestedTensor] multiply batch and ragged dimension to get shape of values tensor,https://github.com/pytorch/pytorch/issues/125940,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nestedtensor,scan when time allows,2024-05-14T04:23:34Z
issue,29115,"crash when call dist.new_group(ranks=local_ranks, backend='gloo')",https://github.com/pytorch/pytorch/issues/29115,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2024-05-14T03:23:26Z
issue,87501,Synchronize domain builds to be executed after core build have completed,https://github.com/pytorch/pytorch/issues/87501,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2024-05-14T00:26:25Z
issue,125471,Build failure when using CUDAGraph,https://github.com/pytorch/pytorch/issues/125471,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: cuda, module: docs",scan when time allows,2024-05-13T22:24:06Z
issue,125800,"[BE][DTensor] get rid of @with_comms decorator for tests, and use run_test to init pg",https://github.com/pytorch/pytorch/issues/125800,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2024-05-13T20:35:58Z
issue,125895,Support for third-party use of pipeline,https://github.com/pytorch/pytorch/issues/125895,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: privateuse1, oncall: distributed",scan when time allows,2024-05-13T20:35:26Z
issue,109833,Implement Copy-on-write (COW) tensors,https://github.com/pytorch/pytorch/issues/109833,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: copy on write, module: internals, module: viewing and reshaping",scan when time allows,2024-05-13T14:54:37Z
issue,125837,Support for PyTorch on iOS Following PEP 730,https://github.com/pytorch/pytorch/issues/125837,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: ios",scan when time allows,2024-05-11T04:13:17Z
issue,65140,profiler_legacy `_parse_legacy_records` deduplication logic has false-positive.,https://github.com/pytorch/pytorch/issues/65140,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2024-05-10T20:40:19Z
issue,124974,[Releng] Triton version for minor releases starting with 2.4,https://github.com/pytorch/pytorch/issues/124974,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: releng,scan when time allows,2024-05-10T15:47:43Z
issue,111908,Inconsistent Keyword Arguments behaviors in torch.triangular_solve(),https://github.com/pytorch/pytorch/issues/111908,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: linear algebra, module: python frontend",scan when time allows,2024-05-10T08:53:43Z
issue,7733,Inserting a tensor into a python dict causes strange behavior,https://github.com/pytorch/pytorch/issues/7733,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-05-09T23:06:33Z
issue,125861,`torch.nn.LazyLinear` custom weight initialization,https://github.com/pytorch/pytorch/issues/125861,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-05-09T20:47:20Z
issue,104045,"""TorchDynamo Deeper Dive"" doc is missing some part",https://github.com/pytorch/pytorch/issues/104045,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2024-05-09T17:00:40Z
issue,125531,`LRScheduler` Refactoring,https://github.com/pytorch/pytorch/issues/125531,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: lrscheduler, module: optimizer",scan when time allows,2024-05-09T15:11:59Z
issue,125547,PyTorch model slows down 30x when binding threads with fewer than half the CPU cores.,https://github.com/pytorch/pytorch/issues/125547,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: docs, module: performance",scan when time allows,2024-05-09T15:08:48Z
issue,125502,C++ libtorch x86_64 Darwin deprecation note missing,https://github.com/pytorch/pytorch/issues/125502,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: cpp, module: macos",scan when time allows,2024-05-09T13:07:12Z
issue,125835,"When using torch.onnx.export, the result of torch.scatter_add may be wrong for opset11",https://github.com/pytorch/pytorch/issues/125835,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2024-05-09T12:52:45Z
issue,125539,`triton` package that ships with torch is not compatible with `torch.ops.matmul`,https://github.com/pytorch/pytorch/issues/125539,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cuda",scan when time allows,2024-05-09T12:43:03Z
issue,53095,Reset mask for torch.cumsum?,https://github.com/pytorch/pytorch/issues/53095,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: numpy, module: reductions",scan when time allows,2024-05-09T10:37:02Z
issue,125746,DefaultCPUAllocator: not enough memory in matrix multiplication broadcasting,https://github.com/pytorch/pytorch/issues/125746,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: complex,scan when time allows,2024-05-09T02:00:30Z
issue,125738,[FSDP2] _sharded_param_data is sitll on meta while sharded_param moved to cuda after calling initialize_parameters(),https://github.com/pytorch/pytorch/issues/125738,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fsdp,scan when time allows,2024-05-08T22:55:04Z
issue,125744,Improve oneDNN memory alloction performance for pytorch Windows,https://github.com/pytorch/pytorch/issues/125744,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: performance, module: windows",scan when time allows,2024-05-08T09:57:43Z
issue,123169,Need better documentation for running doctest locally,https://github.com/pytorch/pytorch/issues/123169,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ci, module: docs, module: tests",scan when time allows,2024-05-07T16:25:54Z
issue,125581,fx hook failed,https://github.com/pytorch/pytorch/issues/125581,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2024-05-07T15:54:05Z
issue,50339,Function request: scipy.interpolate.griddata,https://github.com/pytorch/pytorch/issues/50339,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: interpolation, module: numpy",scan when time allows,2024-05-07T09:11:51Z
issue,125051,MPS backend thinks that subnormals are equal to zero,https://github.com/pytorch/pytorch/issues/125051,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: mps",scan when time allows,2024-05-07T05:02:31Z
issue,125626,[BE] Autoclose issues that haven't been updated in X months,https://github.com/pytorch/pytorch/issues/125626,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: devx,scan when time allows,2024-05-06T22:17:08Z
issue,120534,FLOP Counter does not handle kwargs,https://github.com/pytorch/pytorch/issues/120534,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2024-05-06T22:13:21Z
issue,112942,Can't build PyTorch 2.1 from source by GCC 13.2 on M1 MacOS,https://github.com/pytorch/pytorch/issues/112942,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: docs, module: macos",scan when time allows,2024-05-06T20:54:29Z
issue,125620,[BE] DTensor optimizer to utilize OptimizerInfo Constructs,https://github.com/pytorch/pytorch/issues/125620,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2024-05-06T19:18:56Z
issue,62396,"Make interpolation output size compatible with opencv, scikit-image and scipy for floating scale factor",https://github.com/pytorch/pytorch/issues/62396,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: interpolation,scan when time allows,2024-05-06T17:03:43Z
issue,125591,PyTorch gradient checkpointing consumes a lot more memory than JAX,https://github.com/pytorch/pytorch/issues/125591,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: activation checkpointing, module: memory usage",scan when time allows,2024-05-06T15:40:49Z
issue,104981,NotImplementedError: Could not run 'aten::_spdiags' with arguments from the 'CUDA' backend.,https://github.com/pytorch/pytorch/issues/104981,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: sparse",scan when time allows,2024-05-05T17:52:02Z
issue,106141,Sparse COO indices are torch.Int64 -- is this necessary?,https://github.com/pytorch/pytorch/issues/106141,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2024-05-04T16:43:39Z
issue,37554,Add name to Class Parameter(),https://github.com/pytorch/pytorch/issues/37554,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: rpc, oncall: distributed",scan when time allows,2024-05-04T13:40:54Z
issue,102999,Preserve weight_g/weight_v accessors on new weight_norm,https://github.com/pytorch/pytorch/issues/102999,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: nn.utils.parametrize",scan when time allows,2024-05-04T12:09:51Z
issue,35441,"torch::normal only supports (double, double), but at::normal supports (double, double) / (double, Tensor) / (Tensor, double) / (Tensor, Tensor)",https://github.com/pytorch/pytorch/issues/35441,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp,scan when time allows,2024-05-04T06:41:18Z
issue,3473,Feature Request: Distributed send arbitrary objects,https://github.com/pytorch/pytorch/issues/3473,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: pickle, module: serialization, oncall: distributed",scan when time allows,2024-05-04T00:40:02Z
issue,125493,Error calling to_sparse_coo() on any subclass of Tensor,https://github.com/pytorch/pytorch/issues/125493,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2024-05-03T20:02:46Z
issue,125390,[typing] Make arguments to `__getitem__`/`__setitem__`/ etc. positional only.,https://github.com/pytorch/pytorch/issues/125390,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: typing,scan when time allows,2024-05-03T13:11:35Z
issue,124726,Foreach tests should xfail on all dtypes that are not supported,https://github.com/pytorch/pytorch/issues/124726,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: mta, module: tests",scan when time allows,2024-05-03T08:49:47Z
issue,67240,Feature: Add derivative for channel_shuffle,https://github.com/pytorch/pytorch/issues/67240,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: nn",scan when time allows,2024-05-02T10:37:31Z
issue,59168,[CUDA] Add channels_last_3d support for commonly used modules,https://github.com/pytorch/pytorch/issues/59168,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: memory format",scan when time allows,2024-05-02T08:05:46Z
issue,108342,ONNX export constant folding messes up with shared weight deduplication,https://github.com/pytorch/pytorch/issues/108342,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2024-05-02T00:43:34Z
issue,110131,Some ONNX tests have been disabled because of new tensor.split signature,https://github.com/pytorch/pytorch/issues/110131,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2024-05-01T22:02:34Z
issue,116965,Support GridSample 5D,https://github.com/pytorch/pytorch/issues/116965,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2024-05-01T20:48:38Z
issue,70446,Exporting the operator prim_DictConstruct to ONNX opset version 13 is not supported,https://github.com/pytorch/pytorch/issues/70446,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2024-05-01T20:11:36Z
issue,19092,[RFC] Memory format (aka layout aka NHWC) support,https://github.com/pytorch/pytorch/issues/19092,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: internals, module: mkldnn",scan when time allows,2024-05-01T19:39:21Z
issue,21120,[RFC] Adding MKL-DNN Int8 functions to PyTorch/Aten/JIT backend,https://github.com/pytorch/pytorch/issues/21120,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: mkldnn, oncall: jit, oncall: quantization",scan when time allows,2024-05-01T19:39:14Z
issue,105582,RFC: Integrating oneDNN Graph Compiler into Inductor C++/OpenMP Backend for Enhanced Graph Fusion and Performance,https://github.com/pytorch/pytorch/issues/105582,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: cpu inductor, oncall: pt2",scan when time allows,2024-05-01T19:38:57Z
issue,48246,[RFC] XPU device for PyTorch,https://github.com/pytorch/pytorch/issues/48246,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: intel,scan when time allows,2024-05-01T19:37:38Z
issue,94322,RFC: Enabling AVX512 dispatch for compute-intensive ATen ops,https://github.com/pytorch/pytorch/issues/94322,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: intel, module: performance",scan when time allows,2024-05-01T19:37:31Z
issue,114835,[RFC] Building system for SYCL and limited number of SYCL kernels for ATen fallbacks of TorchInductor,https://github.com/pytorch/pytorch/issues/114835,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, module: intel",scan when time allows,2024-05-01T19:36:57Z
issue,125298,Tensorboard add_custom_scalar does not work,https://github.com/pytorch/pytorch/issues/125298,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tensorboard,scan when time allows,2024-05-01T14:05:44Z
issue,125146,Race condition in FileTimerServerTest.test_expired_timers,https://github.com/pytorch/pytorch/issues/125146,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ci, module: testing, module: tests, oncall: r2p",scan when time allows,2024-05-01T02:51:40Z
issue,119442,pytest test/onnx/test_pytorch_onnx_onnxruntime_cuda.py -k test_embedding_bag_2d_per_sample_weights is failing,https://github.com/pytorch/pytorch/issues/119442,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2024-04-30T23:47:28Z
issue,120822,Problem differentiating through a torch.layer_norm() call when exporting to torch.jit.trace,https://github.com/pytorch/pytorch/issues/120822,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2024-04-30T23:15:16Z
issue,124915,Exporting torch slice_scatter to onnx Identity,https://github.com/pytorch/pytorch/issues/124915,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2024-04-30T21:50:44Z
issue,124523,optim.apply_optimizer_in_backward does not account for gradient accumulation,https://github.com/pytorch/pytorch/issues/124523,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: optimizer, oncall: distributed",scan when time allows,2024-04-30T21:28:18Z
issue,125243,omp.h not found on macOS during install from source at github repo,https://github.com/pytorch/pytorch/issues/125243,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: macos",scan when time allows,2024-04-30T20:25:29Z
issue,125239,Improved strategy for dealing with deterministically flaky tests which are order sensitive,https://github.com/pytorch/pytorch/issues/125239,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ci, module: tests",scan when time allows,2024-04-30T19:41:20Z
issue,125183,torch/_refs/__init__.py is not autoreloadable in bento / jupyter notebook,https://github.com/pytorch/pytorch/issues/125183,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: library,scan when time allows,2024-04-30T19:30:41Z
issue,125116,Does torch.nn.Linear check the weights shape before assignment?,https://github.com/pytorch/pytorch/issues/125116,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-04-30T19:29:27Z
issue,124788,torch.nn.checkpoint.checkpoint ignores default device in backward() call,https://github.com/pytorch/pytorch/issues/124788,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: activation checkpointing,scan when time allows,2024-04-30T19:13:24Z
issue,75512,Connection closed by peer when using dist.isend in gloo backend,https://github.com/pytorch/pytorch/issues/75512,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2024-04-30T17:12:19Z
issue,98208,Add a deterministic version of reflection_pad2d_backward_cuda,https://github.com/pytorch/pytorch/issues/98208,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: determinism, module: nn",scan when time allows,2024-04-30T16:16:13Z
issue,68102,Can not guarantee reproducible for `nn.ReflectionPad2d`,https://github.com/pytorch/pytorch/issues/68102,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: numerical-reproducibility,scan when time allows,2024-04-30T16:16:11Z
issue,28233,torch native functions cannot be used with inspect.signature,https://github.com/pytorch/pytorch/issues/28233,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: language binding, module: pybind",scan when time allows,2024-04-30T10:48:18Z
issue,72582,"Avoid using thrust:: directly, use THRUST_NS_QUALIFIER:: instead",https://github.com/pytorch/pytorch/issues/72582,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cuda",scan when time allows,2024-04-30T04:04:32Z
issue,75242,"large model, low memory: need `torch.load` that loads one submodule at a time",https://github.com/pytorch/pytorch/issues/75242,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: serialization,scan when time allows,2024-04-29T22:27:24Z
issue,103354,Add a requirements.txt for windows pip packages,https://github.com/pytorch/pytorch/issues/103354,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: devx,scan when time allows,2024-04-29T18:26:30Z
issue,124934,Tensor's storage changes computation outcome for CPU tensors.,https://github.com/pytorch/pytorch/issues/124934,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpu,scan when time allows,2024-04-29T13:40:51Z
issue,114040,IndexError: map::at with MPI CUDA collectives,https://github.com/pytorch/pytorch/issues/114040,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: multiprocessing",scan when time allows,2024-04-29T11:05:33Z
issue,53534,[FX] Ability to wrap functions in other modules for symbolic tracing,https://github.com/pytorch/pytorch/issues/53534,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2024-04-29T09:01:56Z
issue,104516,vec_test_all_types_xxx with dtype c10::complex<float> and c10::complex<double> has failures on division,https://github.com/pytorch/pytorch/issues/104516,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: complex, module: cpu, module: vectorization",scan when time allows,2024-04-29T00:46:25Z
issue,124998,"When I set the operator weights all to 1, the operator weights change as soon as an input is made to the operator",https://github.com/pytorch/pytorch/issues/124998,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2024-04-27T14:27:07Z
issue,125028,OpInfo testing is inadequate for `nextafter`,https://github.com/pytorch/pytorch/issues/125028,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: testing,scan when time allows,2024-04-26T22:00:41Z
issue,125064,ValueError: weight_norm of 'weight' not found in ParametrizedConvTranspose1d,https://github.com/pytorch/pytorch/issues/125064,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn.utils.parametrize,scan when time allows,2024-04-26T21:58:26Z
issue,124973,Export llama v3 to ONNX,https://github.com/pytorch/pytorch/issues/124973,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2024-04-26T18:02:08Z
issue,68892,fusion in fx graph mode did not take care of direct attribute access,https://github.com/pytorch/pytorch/issues/68892,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2024-04-26T14:20:12Z
issue,124262,CUBLAS_STATUS_EXECUTION_FAILED when calling cublasGemmEx,https://github.com/pytorch/pytorch/issues/124262,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cublas, module: cuda",scan when time allows,2024-04-26T09:17:49Z
issue,124989,[DTensor] Keep track of data-dependent ops and skip _propagate_tensor_meta to go thru fake tensor,https://github.com/pytorch/pytorch/issues/124989,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dtensor, oncall: distributed",scan when time allows,2024-04-26T00:39:03Z
issue,97856,make tensor data const correct,https://github.com/pytorch/pytorch/issues/97856,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: copy on write, module: internals",scan when time allows,2024-04-26T00:00:26Z
issue,102227,lintrunner should fail on badly formatted docstrings,https://github.com/pytorch/pytorch/issues/102227,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: devx, oncall: pt2",scan when time allows,2024-04-25T19:34:14Z
issue,99372,PyTorch 2.0.0 encountered CUDA error: an illegal memory access was encountered,https://github.com/pytorch/pytorch/issues/99372,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: multithreading",scan when time allows,2024-04-25T19:26:23Z
issue,124953,PRs with unbalanced quotes can not be merged,https://github.com/pytorch/pytorch/issues/124953,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: infra,scan when time allows,2024-04-25T19:17:52Z
issue,124956,torch.onnx.export generates an incorrect bias shape for certain conv transpose models,https://github.com/pytorch/pytorch/issues/124956,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: onnx, oncall: quantization",scan when time allows,2024-04-25T19:17:44Z
issue,98724,Conflict between ``torch.func`` transformations and ``torch.jit.trace``,https://github.com/pytorch/pytorch/issues/98724,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: functorch,scan when time allows,2024-04-25T15:55:11Z
issue,88136,Placing LSTM model on bfloat16 on GPU causes error,https://github.com/pytorch/pytorch/issues/88136,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bfloat16, module: cuda, module: rnn",scan when time allows,2024-04-25T07:26:19Z
issue,124838,Forwardpropagation of high-dimensional tensors through the nn.Linear module becomes multiple times slower since pytorch 2.1.0,https://github.com/pytorch/pytorch/issues/124838,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2024-04-25T03:32:14Z
issue,120989,input.is_sparse() INTERNAL ASSERT FAILED,https://github.com/pytorch/pytorch/issues/120989,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2024-04-24T21:06:41Z
issue,51020,arm64-v8a not compiling due to libpytorch_jni.so,https://github.com/pytorch/pytorch/issues/51020,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: android, module: arm, module: build",scan when time allows,2024-04-24T20:08:52Z
issue,116506,[discussion] Route pointwise Conv1d/Conv2d to matmul? (also in eager),https://github.com/pytorch/pytorch/issues/116506,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: cuda, module: mkldnn, module: performance",scan when time allows,2024-04-24T18:39:53Z
issue,124857,Use lintrunner-adapters for our adapters when possible,https://github.com/pytorch/pytorch/issues/124857,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: lint,scan when time allows,2024-04-24T17:27:47Z
issue,124783,Scatter_add limitation when accumulate beyond 2^24 under float32 precision,https://github.com/pytorch/pytorch/issues/124783,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: scatter & gather ops,scan when time allows,2024-04-24T16:16:48Z
issue,123835,Package manager install on Nvidia Grace Hopper does not make cuda available,https://github.com/pytorch/pytorch/issues/123835,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: arm, module: binaries, module: cuda",scan when time allows,2024-04-24T16:04:46Z
issue,124719,"Questions about parameter initialization, especially with torch.bfloat16 precision",https://github.com/pytorch/pytorch/issues/124719,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: initialization, module: nn",scan when time allows,2024-04-24T15:35:10Z
issue,85533,"[functorch] transforms like jacrev, jacfwd, grad, etc don't work with BatchNorm",https://github.com/pytorch/pytorch/issues/85533,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: functorch,scan when time allows,2024-04-24T13:23:33Z
issue,124456,CudaHostAlloc takes a lot of time during training,https://github.com/pytorch/pytorch/issues/124456,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: cudacachingallocator, module: dataloader",scan when time allows,2024-04-24T02:31:02Z
issue,124572,Variant of slice that doesn't perform clamp behavior,https://github.com/pytorch/pytorch/issues/124572,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dynamic shapes,scan when time allows,2024-04-23T15:08:44Z
issue,124222,Bot to recomment on high pri issues if no activity,https://github.com/pytorch/pytorch/issues/124222,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: devx,scan when time allows,2024-04-22T23:24:47Z
issue,124417,ghstack mergeability check seems wrong,https://github.com/pytorch/pytorch/issues/124417,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2024-04-22T18:07:44Z
issue,124630,TORCH_SHOW_DISPATCH_TRACE should give information about PythonDispatcher,https://github.com/pytorch/pytorch/issues/124630,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dispatch,scan when time allows,2024-04-22T17:59:14Z
issue,34279,`torch.norm` is 113x slower than `torch.sqrt(a**2 + b**2)`,https://github.com/pytorch/pytorch/issues/34279,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: linear algebra, module: performance",scan when time allows,2024-04-22T10:56:16Z
issue,123528,Support integer parameters in FullyShardedDataParallel,https://github.com/pytorch/pytorch/issues/123528,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2024-04-22T10:04:25Z
issue,122592,build libtorch in windows with intel oneapi base 2024 error,https://github.com/pytorch/pytorch/issues/122592,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: windows",scan when time allows,2024-04-22T07:33:40Z
issue,113546,"P0: Improve failure trace back when crashed to identify the cause of a crash and the ranks that the crash, output the real traceback at last.",https://github.com/pytorch/pytorch/issues/113546,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: c10d, module: elastic",scan when time allows,2024-04-21T19:07:09Z
issue,110820,"""Expected a 'mps:0' generator device but found 'cpu'"" using shuffle=True on DataLoader",https://github.com/pytorch/pytorch/issues/110820,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dataloader, module: mps",scan when time allows,2024-04-21T13:32:24Z
issue,116515,mobile model has very different inference result,https://github.com/pytorch/pytorch/issues/116515,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: mobile,scan when time allows,2024-04-20T16:12:23Z
issue,124465,"Matmul with broadcast makes copy because of reshape to bmm, and it is slow",https://github.com/pytorch/pytorch/issues/124465,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: performance,scan when time allows,2024-04-19T15:28:30Z
issue,124383,llvm disabled in official builds from 2.1.0 onwards,https://github.com/pytorch/pytorch/issues/124383,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2024-04-19T15:14:37Z
issue,82843,Tensor operation hangs when used with multiprocessing,https://github.com/pytorch/pytorch/issues/82843,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: determinism, module: multiprocessing",scan when time allows,2024-04-19T13:55:21Z
issue,88775,Batched Random Number Generators,https://github.com/pytorch/pytorch/issues/88775,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: random,scan when time allows,2024-04-18T18:04:27Z
issue,124291,vmap over getitem indexing raises a RunTime error,https://github.com/pytorch/pytorch/issues/124291,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: functorch,scan when time allows,2024-04-18T02:53:44Z
issue,111454,[ONNX][dynamo] Failed to export cumsum with dtype=float16,https://github.com/pytorch/pytorch/issues/111454,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: half, module: onnx",scan when time allows,2024-04-17T21:22:47Z
issue,123120,TD distributed CUDA Rollout list,https://github.com/pytorch/pytorch/issues/123120,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2024-04-17T20:36:47Z
issue,124194,Initial state of RNNs: h_0 or hx?,https://github.com/pytorch/pytorch/issues/124194,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: rnn",scan when time allows,2024-04-17T14:20:26Z
issue,34837,[torch.jit.script] Support tensor indexing after an Ellipsis,https://github.com/pytorch/pytorch/issues/34837,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2024-04-17T07:13:52Z
issue,77159,"A somewhat cryptic error message (for newcomers) - ""Cannot re-initialize CUDA in forked subprocess"" - report and suggestion for a possible solution",https://github.com/pytorch/pytorch/issues/77159,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2024-04-17T01:04:16Z
issue,124228,[DeviceMesh] Current cache for submesh result is not thread save,https://github.com/pytorch/pytorch/issues/124228,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2024-04-16T22:43:00Z
issue,99143,No documentation to show how to implement aten::view for custom backend,https://github.com/pytorch/pytorch/issues/99143,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp-extensions, module: docs",scan when time allows,2024-04-16T16:18:30Z
issue,117961,ReplicatePad2d cuda not support bfloat16,https://github.com/pytorch/pytorch/issues/117961,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: bfloat16,scan when time allows,2024-04-16T06:22:58Z
issue,124145,Use online softmax for just softmax by itself,https://github.com/pytorch/pytorch/issues/124145,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: inductor,scan when time allows,2024-04-16T04:44:59Z
issue,110758,"Fused Adamw RuntimeError: params, grads, exp_avgs, and exp_avg_sqs must have same dtype, device, and layout",https://github.com/pytorch/pytorch/issues/110758,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2024-04-16T04:18:47Z
issue,52291,"""LayerNormKernelImpl"" not implemented for 'Half' - CPU",https://github.com/pytorch/pytorch/issues/52291,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: half, module: nn",scan when time allows,2024-04-15T23:21:06Z
issue,123996,MSE loss triggers ZeroTensor immutable error inside `torch.func`,https://github.com/pytorch/pytorch/issues/123996,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: functorch",scan when time allows,2024-04-15T18:34:35Z
issue,123284,We should walk torch._C and require everything have type annotations,https://github.com/pytorch/pytorch/issues/123284,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: internals, module: typing",scan when time allows,2024-04-15T17:19:06Z
issue,124047,dlopen: cannot load any more object with static TLS,https://github.com/pytorch/pytorch/issues/124047,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: binaries,scan when time allows,2024-04-15T15:28:16Z
issue,33859,Possibility to support int4 data type,https://github.com/pytorch/pytorch/issues/33859,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2024-04-15T10:07:15Z
issue,499,Feature Request: Locally Connected Layer,https://github.com/pytorch/pytorch/issues/499,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-04-15T09:24:01Z
issue,115152,scatter_ op after permute op not working correctly on mps device,https://github.com/pytorch/pytorch/issues/115152,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: mps, module: scatter & gather ops",scan when time allows,2024-04-12T02:34:51Z
issue,123765,Accurately typed ValueRanges / convenient OpsHandler dtype inference,https://github.com/pytorch/pytorch/issues/123765,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dynamic shapes,scan when time allows,2024-04-11T17:53:05Z
issue,1249,Dice Loss PR,https://github.com/pytorch/pytorch/issues/1249,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: loss,scan when time allows,2024-04-11T16:28:24Z
issue,35472,Multiprocessing map gets stuck if doing inference on loaded model,https://github.com/pytorch/pytorch/issues/35472,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: multiprocessing,scan when time allows,2024-04-11T09:19:15Z
issue,123421,nonzero with dtensor hang,https://github.com/pytorch/pytorch/issues/123421,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2024-04-11T05:22:30Z
issue,105299,Support ONNX opset 20 to export GELU to one single op,https://github.com/pytorch/pytorch/issues/105299,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2024-04-10T20:59:39Z
issue,116120,bfloat16 on 'grid_sampler_3d_cuda' not implemented,https://github.com/pytorch/pytorch/issues/116120,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: bfloat16,scan when time allows,2024-04-10T18:33:04Z
issue,58267,Pytorch build issues on PowerPC,https://github.com/pytorch/pytorch/issues/58267,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: power",scan when time allows,2024-04-10T15:52:41Z
issue,37488,undefined reference to pthreadpool_compute*,https://github.com/pytorch/pytorch/issues/37488,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: xnnpack",scan when time allows,2024-04-10T12:35:17Z
issue,123533,Memory allocation for Conv2d,https://github.com/pytorch/pytorch/issues/123533,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: cuda",scan when time allows,2024-04-10T11:23:38Z
issue,123506,[RFE] Disallow submodule updates without successful minimal binary builds,https://github.com/pytorch/pytorch/issues/123506,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ci, oncall: releng",scan when time allows,2024-04-10T11:22:22Z
issue,123702,Issue with Vulkan Device for nn.Conv2d Operation,https://github.com/pytorch/pytorch/issues/123702,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: vulkan,scan when time allows,2024-04-10T11:12:08Z
issue,72791,"why ram memory surges while loading model, with change in torch load device from CPU to GPU",https://github.com/pytorch/pytorch/issues/72791,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: memory usage",scan when time allows,2024-04-10T03:02:32Z
issue,8069,Document torch.acos() behavior near -1 and 1,https://github.com/pytorch/pytorch/issues/8069,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2024-04-09T20:56:16Z
issue,123245,torch.overrides.get_overridable_functions list functions that are not overridable via __torch_function__,https://github.com/pytorch/pytorch/issues/123245,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: __torch_function__,scan when time allows,2024-04-09T10:56:15Z
issue,45139,Mention accessor/data_ptr for raw memory access in Libtorch index API document and discuss performance implications,https://github.com/pytorch/pytorch/issues/45139,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: performance",scan when time allows,2024-04-09T04:49:54Z
issue,123604,"Can't initialize NVML, H100 gpus on ubuntu",https://github.com/pytorch/pytorch/issues/123604,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2024-04-09T02:30:33Z
issue,119181,[inductor][cpu]Background_Matting fP32 static default wrapper multiple threads performance regression,https://github.com/pytorch/pytorch/issues/119181,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: cpu inductor,scan when time allows,2024-04-09T00:00:58Z
issue,123500,"Cannot add 2 sparse_csc_tensor directly, raising RuntimeError",https://github.com/pytorch/pytorch/issues/123500,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2024-04-08T22:19:46Z
issue,123466,Undefined behavior when an OpOverload's Library is deleted but the OpOverload continues to be used,https://github.com/pytorch/pytorch/issues/123466,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2024-04-08T22:19:02Z
issue,22400,Sparse allreduce for ProcessGroupNCCL,https://github.com/pytorch/pytorch/issues/22400,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2024-04-07T07:22:00Z
issue,102936,torch.cuda.is_available() returns False on GTX 1650 with cuda 11.7 and torch==2.0.0+cpu,https://github.com/pytorch/pytorch/issues/102936,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2024-04-06T20:19:33Z
issue,113600,Segmentation fault in RPC worker when DataLoader has num_workers > 0,https://github.com/pytorch/pytorch/issues/113600,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dataloader, oncall: distributed",scan when time allows,2024-04-06T07:02:16Z
issue,66707,`layer_norm` needs to be done in fp32 for fp16 inputs,https://github.com/pytorch/pytorch/issues/66707,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: half, module: norms and normalization, module: numerical-stability",scan when time allows,2024-04-06T04:17:18Z
issue,121960,"RuntimeError: ""max_pool2d_channels_last"" not implemented for 'Half'",https://github.com/pytorch/pytorch/issues/121960,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: half, module: memory format, module: mkldnn, module: nn",scan when time allows,2024-04-05T17:15:49Z
issue,118478,Cannot formatted print at::Tensor in gdb,https://github.com/pytorch/pytorch/issues/118478,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp,scan when time allows,2024-04-05T17:14:27Z
issue,55902,Warn users when cross entropy is called after softmax,https://github.com/pytorch/pytorch/issues/55902,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: loss, module: nn",scan when time allows,2024-04-05T10:59:17Z
issue,103089,"[OOM] Unable to convert 30B model to ONNX, using 4x A100's",https://github.com/pytorch/pytorch/issues/103089,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2024-04-04T23:00:14Z
issue,123344,document torch.ops.import_module on the website; also make sure to use this in tutorials,https://github.com/pytorch/pytorch/issues/123344,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2024-04-04T17:09:21Z
issue,123262,mmap error - no such device (19),https://github.com/pytorch/pytorch/issues/123262,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2024-04-04T08:13:51Z
issue,111704,Add more flexibility on print / output console,https://github.com/pytorch/pytorch/issues/111704,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp,scan when time allows,2024-04-04T04:19:32Z
issue,121650,add scatter to functional collectives,https://github.com/pytorch/pytorch/issues/121650,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2024-04-03T23:47:20Z
issue,103805,Error when building with USE_TENSORRT=1,https://github.com/pytorch/pytorch/issues/103805,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2024-04-03T16:25:49Z
issue,123157,Add support for CUBLAS_COMPUTE_16F for GEMM opeartions in cudaBLAS,https://github.com/pytorch/pytorch/issues/123157,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cublas, module: cuda",scan when time allows,2024-04-02T20:26:47Z
issue,120820,Problems differentiating through certain operations when exporting to TorchScript,https://github.com/pytorch/pytorch/issues/120820,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2024-04-02T17:24:22Z
issue,123172,`torch.utils.data.Sampler`: constructor's `data_source` argument should be removed,https://github.com/pytorch/pytorch/issues/123172,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: data, module: deprecation",scan when time allows,2024-04-02T16:02:07Z
issue,120139,"On `Dataset`, `IterableDataset` inheritance",https://github.com/pytorch/pytorch/issues/120139,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: data,scan when time allows,2024-04-02T16:00:57Z
issue,122410,On Lack of Default `__len__` in Python Abstract Base Classes,https://github.com/pytorch/pytorch/issues/122410,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: data, module: docs",scan when time allows,2024-04-02T15:56:14Z
issue,27421,[dataloader] Sampler abstract constructor API minor proposal,https://github.com/pytorch/pytorch/issues/27421,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2024-04-02T15:51:09Z
issue,57278,Attributes consistency in Sampler and DistributedSampler,https://github.com/pytorch/pytorch/issues/57278,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2024-04-02T15:50:28Z
issue,121797,Can’t save SparseTensor for custom device,https://github.com/pytorch/pytorch/issues/121797,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2024-04-02T12:35:18Z
issue,91950,RuntimeError: philox_cuda_state for an unexpected CUDA generator used during capture,https://github.com/pytorch/pytorch/issues/91950,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda graphs, module: random",scan when time allows,2024-04-02T10:01:11Z
issue,115138,ImportError: /lib64/libstdc++.so.6: version `GLIBCXX_3.4.20' not found,https://github.com/pytorch/pytorch/issues/115138,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: binaries,scan when time allows,2024-04-02T09:25:14Z
issue,123091,build error: memory_desc.cpp:716:1: internal compiler error: Segmentation fault,https://github.com/pytorch/pytorch/issues/123091,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: mkldnn, module: third_party",scan when time allows,2024-04-02T09:01:45Z
issue,97990,Relax version dependencies on CUDA pip wheels?,https://github.com/pytorch/pytorch/issues/97990,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: cuda",scan when time allows,2024-04-02T06:31:45Z
issue,123082,Torch.cdist calculation problem,https://github.com/pytorch/pytorch/issues/123082,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2024-04-01T16:23:18Z
issue,10996,Multiprocess Deadlock when using np.transpose and torch.stack,https://github.com/pytorch/pytorch/issues/10996,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: multiprocessing,scan when time allows,2024-04-01T16:20:49Z
issue,123075,LibTorch Release version 2.2.2+cpu loads traced model only in Debug configuration,https://github.com/pytorch/pytorch/issues/123075,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: binaries,scan when time allows,2024-04-01T16:02:07Z
issue,122756,How to reduce memory usage for large matrix calculations？,https://github.com/pytorch/pytorch/issues/122756,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2024-04-01T15:59:16Z
issue,123070,CUDA out of memory still exist after using FSDP,https://github.com/pytorch/pytorch/issues/123070,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2024-04-01T15:57:02Z
issue,91353,Perf reduction due to munmap with dataloader pinning thread ?,https://github.com/pytorch/pytorch/issues/91353,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2024-04-01T11:42:11Z
issue,98002,DataLoader with collate_fn that returns tensors in GPU memory raises warnings when deleted,https://github.com/pytorch/pytorch/issues/98002,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2024-03-30T23:52:16Z
issue,100637,Optimal Batch Size Selection in Torchdynamo Benchmarks for Different GPU Memory Sizes,https://github.com/pytorch/pytorch/issues/100637,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: pt2,scan when time allows,2024-03-29T23:54:13Z
issue,89417,OSError: libcublas.so.11: cannot open shared object file: No such file or directory,https://github.com/pytorch/pytorch/issues/89417,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: binaries,scan when time allows,2024-03-29T13:30:52Z
issue,65156,[feature request] Jagged / padding version of torch.stack / torch.cat + some general nested tensor discussion,https://github.com/pytorch/pytorch/issues/65156,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nestedtensor, module: padding, module: sparse",scan when time allows,2024-03-29T03:32:36Z
issue,67760,Add an LRScheduler interface for torch schedulers.,https://github.com/pytorch/pytorch/issues/67760,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: lrscheduler, module: optimizer",scan when time allows,2024-03-29T02:24:40Z
issue,72106,Update MKL version used with MAGMA,https://github.com/pytorch/pytorch/issues/72106,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ci, module: magma",scan when time allows,2024-03-28T19:12:43Z
issue,122814,torchdim as torch_function subclass exploration,https://github.com/pytorch/pytorch/issues/122814,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: __torch_function__,scan when time allows,2024-03-28T04:41:56Z
issue,122699,Bug of `svd_lowrank` when passing parameter `M`,https://github.com/pytorch/pytorch/issues/122699,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: linear algebra,scan when time allows,2024-03-28T04:32:44Z
issue,76510,"Docs source link broken for all methods defined on Tensor in _tensor.py, e.g. `Tensor.register_hook()`",https://github.com/pytorch/pytorch/issues/76510,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2024-03-27T22:27:59Z
issue,60149,backward compatibility - need a way to find out when a certain API was added/modified/etc.,https://github.com/pytorch/pytorch/issues/60149,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: codegen, module: deprecation, module: doc infra, module: docs",scan when time allows,2024-03-27T19:35:12Z
issue,122803,Enabling `swap_tensors` path by default for all tensor subclasses errors on some transformer tests,https://github.com/pytorch/pytorch/issues/122803,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-03-27T17:44:45Z
issue,106121,Got error when train models with more than one param_group in torch2.0,https://github.com/pytorch/pytorch/issues/106121,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2024-03-27T17:44:42Z
issue,103483,torch._dynamo.exc.Unsupported: Tensor.backward with aten_graph=True,https://github.com/pytorch/pytorch/issues/103483,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: oncall: export, oncall: pt2",scan when time allows,2024-03-27T17:34:13Z
issue,122081,[Feature Request] add a new argument `storage_offset=None` to `empty_strided`,https://github.com/pytorch/pytorch/issues/122081,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python frontend,scan when time allows,2024-03-27T14:25:58Z
issue,91309,iSTFT produces RuntimeError with center=False and Blackman/Bartlett/Hann windows,https://github.com/pytorch/pytorch/issues/91309,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fft,scan when time allows,2024-03-27T10:23:16Z
issue,95708,Dynamo + MacOS: fatal error: 'omp.h' file not found,https://github.com/pytorch/pytorch/issues/95708,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: macos, oncall: cpu inductor, oncall: pt2",scan when time allows,2024-03-27T04:18:17Z
issue,67979,Feature Request: Support prelu_cuda for BFloat16,https://github.com/pytorch/pytorch/issues/67979,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bfloat16, module: bootcamp, module: cuda, module: nn",scan when time allows,2024-03-27T00:18:49Z
issue,122467,RuntimeError: out_ptr == out_accessor[thread_count_nonzero[tid + 1]].data() INTERNAL ASSERT FAILED,https://github.com/pytorch/pytorch/issues/122467,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: assert failure, module: multiprocessing, module: sparse",scan when time allows,2024-03-26T18:13:29Z
issue,122170,torch.cumsum() super slow on 2d input,https://github.com/pytorch/pytorch/issues/122170,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: performance",scan when time allows,2024-03-26T17:03:25Z
issue,122552,Could not run 'aten::empty_strided' with arguments from the 'CUDA' backend.,https://github.com/pytorch/pytorch/issues/122552,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2024-03-26T15:52:52Z
issue,77821,cannot convert to channels last format for conv2d conv3d hybrid model,https://github.com/pytorch/pytorch/issues/77821,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: convolution,scan when time allows,2024-03-25T20:00:52Z
issue,122571,some official docker images has mismatch nccl version with pytorch,https://github.com/pytorch/pytorch/issues/122571,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: docker",scan when time allows,2024-03-25T15:19:09Z
issue,122287,Tolerance setting for test_modules.py -- TestModuleCUDA.test_cpu_cuda_parity,https://github.com/pytorch/pytorch/issues/122287,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: testing",scan when time allows,2024-03-25T08:53:51Z
issue,112575,"RuntimeError: ""grid_sampler_2d_cuda"" not implemented for 'BFloat16'",https://github.com/pytorch/pytorch/issues/112575,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bfloat16, module: cuda",scan when time allows,2024-03-23T09:51:12Z
issue,109850,torch._export has no logging,https://github.com/pytorch/pytorch/issues/109850,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: logging, oncall: export, oncall: pt2",scan when time allows,2024-03-23T03:11:00Z
issue,93719,Use bytecode rewriting to offer a __future__ feature,https://github.com/pytorch/pytorch/issues/93719,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2024-03-22T16:38:18Z
issue,54147,Implement Truly Parallel Ensemble Layers,https://github.com/pytorch/pytorch/issues/54147,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: batching, module: nn",scan when time allows,2024-03-22T13:15:40Z
issue,122359,Onnx export memory leak with qlora+deepspeed stage 2+ORT,https://github.com/pytorch/pytorch/issues/122359,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2024-03-22T02:37:38Z
issue,93482,[inductor] Add more matmul configurations to `TORCHINDUCTOR_MAX_AUTOTUNE=1` mode,https://github.com/pytorch/pytorch/issues/93482,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",scan when time allows,2024-03-22T02:23:20Z
issue,117930,C++ Tensor.clone() changes max threads in OpenMP,https://github.com/pytorch/pytorch/issues/117930,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: multithreading",scan when time allows,2024-03-21T19:55:01Z
issue,95785,Fully quantized model (`torch.quantization.convert`) produces incorrect output compared to analytical solution,https://github.com/pytorch/pytorch/issues/95785,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2024-03-21T01:49:47Z
issue,122107,Can't export an onnx with dynamic axes if it contains backpropagation,https://github.com/pytorch/pytorch/issues/122107,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2024-03-20T18:45:03Z
issue,120879,FSDP: unsharded gradients accessor,https://github.com/pytorch/pytorch/issues/120879,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fsdp,scan when time allows,2024-03-20T17:31:59Z
issue,77537,batch Kronecker product,https://github.com/pytorch/pytorch/issues/77537,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2024-03-20T04:04:54Z
issue,122187,CuDNN version for pre-built package conflicts with the pip installed version (pytorch 2.2.1 + cuda 12.1),https://github.com/pytorch/pytorch/issues/122187,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2024-03-19T22:26:27Z
issue,122186,Deep-copied model cannot be sent through multiprocessing `Pipe`,https://github.com/pytorch/pytorch/issues/122186,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: multiprocessing,scan when time allows,2024-03-19T22:09:28Z
issue,122185,Numerical issues on positive semidefiniteness constraint,https://github.com/pytorch/pytorch/issues/122185,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distributions,scan when time allows,2024-03-19T22:07:40Z
issue,82397,Bilinear interpolation with antialiasing is slow in performance,https://github.com/pytorch/pytorch/issues/82397,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: performance,scan when time allows,2024-03-19T21:06:26Z
issue,121842,gradcheck fails when inputs are dependent on each other,https://github.com/pytorch/pytorch/issues/121842,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2024-03-19T15:02:02Z
issue,32097,logsumexp with subtraction,https://github.com/pytorch/pytorch/issues/32097,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: reductions,scan when time allows,2024-03-19T13:48:01Z
issue,121926,"Call `torch.pow(int_tensor, torch.tensor(-1))` does not raise error like NumPy.",https://github.com/pytorch/pytorch/issues/121926,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: numpy,scan when time allows,2024-03-19T13:14:48Z
issue,48163,nn.functional.interpolate backward in fp16 is extremely slow,https://github.com/pytorch/pytorch/issues/48163,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: half, module: interpolation, module: nn",scan when time allows,2024-03-19T08:42:34Z
issue,121917,"[PythonDispatch] when the value of a argument is the same as its default value, __torch_dispatch fails to capture it.",https://github.com/pytorch/pytorch/issues/121917,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: __torch_dispatch__,scan when time allows,2024-03-19T03:07:14Z
issue,115706,[export][functionalize] Complicated subgraph from functionalizing inplace op on sliced tensors,https://github.com/pytorch/pytorch/issues/115706,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: functionalization,scan when time allows,2024-03-19T03:04:54Z
issue,121719,index_copy change value when use_deterministic (true),https://github.com/pytorch/pytorch/issues/121719,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: advanced indexing, module: boolean tensor, module: cuda, module: determinism",scan when time allows,2024-03-19T02:26:05Z
issue,122017,"Got ""KeyError: 'classname'"" while runing pytorch unit tests",https://github.com/pytorch/pytorch/issues/122017,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: testing,scan when time allows,2024-03-19T01:47:02Z
issue,122080,graph.owning_module does not point to the correct module after dataparallel forwarding.,https://github.com/pytorch/pytorch/issues/122080,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: data parallel,scan when time allows,2024-03-19T01:35:34Z
issue,122021,Wishart: problems with singular samples.,https://github.com/pytorch/pytorch/issues/122021,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distributions,scan when time allows,2024-03-18T21:52:49Z
issue,116478,The Python module installs cmake files and headers under the Python's site-packages directory that can't be used,https://github.com/pytorch/pytorch/issues/116478,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cpp",scan when time allows,2024-03-18T12:10:04Z
issue,6934,Feature request: SSIM/MS-SSIM,https://github.com/pytorch/pytorch/issues/6934,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: vision,scan when time allows,2024-03-17T19:13:37Z
issue,23850,[quantization] jit::class_ for packed weights,https://github.com/pytorch/pytorch/issues/23850,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2024-03-17T19:08:30Z
issue,87559,diagonal of Jacobian matrix,https://github.com/pytorch/pytorch/issues/87559,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2024-03-17T11:45:34Z
issue,122027,PyTorch fails to build with mkl if cmake is installed from pip,https://github.com/pytorch/pytorch/issues/122027,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cuda, module: mkl",scan when time allows,2024-03-16T18:47:13Z
issue,117236,```torch.linalg.lstsq```: Internal assert failed,https://github.com/pytorch/pytorch/issues/117236,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: complex,scan when time allows,2024-03-16T10:39:41Z
issue,117242,```torch.tensordot```: Internal assert failed,https://github.com/pytorch/pytorch/issues/117242,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2024-03-16T08:27:40Z
issue,121737,Unable to install pytorch from source for version 2.0.0,https://github.com/pytorch/pytorch/issues/121737,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2024-03-16T05:43:46Z
issue,1329,[proposed feature] Eve: Improving Stochastic Gradient Descent with Feedback,https://github.com/pytorch/pytorch/issues/1329,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2024-03-16T01:10:04Z
issue,121715,Depthwise conv2d op produces nan values with non-nan input,https://github.com/pytorch/pytorch/issues/121715,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: edge cases, module: nans and infs, module: nn",scan when time allows,2024-03-15T18:46:33Z
issue,12760,[feature request] Spectral norm support in torch.norm or factor out Power Iteration from spectralnormalization in some other place and orthogonalization from PowerSGD hook,https://github.com/pytorch/pytorch/issues/12760,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: norms and normalization,scan when time allows,2024-03-15T18:43:32Z
issue,108432,Wrong result of first run with torch.compile() when backend is using torch.jit.trace() and model has inplace operators,https://github.com/pytorch/pytorch/issues/108432,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2024-03-15T09:23:42Z
issue,106991,Add cutlass as an alternative backend of PT2 Inductor,https://github.com/pytorch/pytorch/issues/106991,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",scan when time allows,2024-03-15T08:50:32Z
issue,67682,Feature request: bfloat16 support for CUDA matmuls,https://github.com/pytorch/pytorch/issues/67682,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bfloat16, module: cuda",scan when time allows,2024-03-14T23:29:28Z
issue,65510,nn.conv1d padding='same',https://github.com/pytorch/pytorch/issues/65510,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: docs",scan when time allows,2024-03-14T13:37:23Z
issue,121789,Precision Inconsistent for `half` type compared with NumPy and TensorFlow,https://github.com/pytorch/pytorch/issues/121789,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: numerical-stability, module: numpy",scan when time allows,2024-03-14T06:52:45Z
issue,105281,Optimize PyTorch C++ part with Profile-Guided Optimization (PGO),https://github.com/pytorch/pytorch/issues/105281,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: internals, module: performance",scan when time allows,2024-03-14T02:35:28Z
issue,105092,[RFC] Proposal to upgrade LLVM version,https://github.com/pytorch/pytorch/issues/105092,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: pt2,scan when time allows,2024-03-14T02:24:41Z
issue,104531,[RFC] Optional Modular Representation for FX Graph from `torch.compile()`,https://github.com/pytorch/pytorch/issues/104531,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: pt2,scan when time allows,2024-03-14T01:51:44Z
issue,121683,linalg.ldl_factor raises an error with zero tensor,https://github.com/pytorch/pytorch/issues/121683,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: linear algebra, module: mkl",scan when time allows,2024-03-13T21:20:26Z
issue,95225,Remove conda virtualenv from the docker image,https://github.com/pytorch/pytorch/issues/95225,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: docker",scan when time allows,2024-03-13T13:14:11Z
issue,77646,Werror=nonnull in dataloader.cpp (part of tests),https://github.com/pytorch/pytorch/issues/77646,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2024-03-13T09:08:08Z
issue,102360,PyTorch hangs at import when used together with TensorFlow,https://github.com/pytorch/pytorch/issues/102360,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2024-03-13T00:09:11Z
issue,71028,Skip LSTM quantization by default in get_default_qconfig_dict and get_default_qat_qconfig_dict,https://github.com/pytorch/pytorch/issues/71028,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2024-03-12T18:52:36Z
issue,80874,C++ extensions inject a bunch of compilation flags,https://github.com/pytorch/pytorch/issues/80874,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: cpp-extensions",scan when time allows,2024-03-12T13:02:46Z
issue,117233,Submit a bug when trying to assign arg 'out=' in api torch._foreach_add_,https://github.com/pytorch/pytorch/issues/117233,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: mta, module: python frontend",scan when time allows,2024-03-12T05:53:05Z
issue,121261,Convolution and norm with vmap jacobian failing,https://github.com/pytorch/pytorch/issues/121261,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: forward ad, module: functorch, module: norms and normalization",scan when time allows,2024-03-11T22:28:31Z
issue,121596,Memory Usage of Conv2d Rapidly Increases when Params >1G,https://github.com/pytorch/pytorch/issues/121596,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: cudnn, module: memory usage",scan when time allows,2024-03-11T18:48:37Z
issue,52439,[proposal] Way to export/display the convolution algorithm found with cudnn.benchmark = True + allow to override cudnn algo choice with conv/matmul module/function algo/hint arguments,https://github.com/pytorch/pytorch/issues/52439,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: cudnn",scan when time allows,2024-03-11T18:23:53Z
issue,121588,logistic sigmoid function outputs NaN indeterminate on complex input,https://github.com/pytorch/pytorch/issues/121588,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: complex, module: cpu, module: nans and infs",scan when time allows,2024-03-11T16:18:00Z
issue,121629,Pad operator produces non-nan value with nan input,https://github.com/pytorch/pytorch/issues/121629,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2024-03-11T13:51:53Z
issue,50820,NCCL_BLOCKING_WAIT=1 makes training extremely slow (but if not set then OOM on one device will hang training),https://github.com/pytorch/pytorch/issues/50820,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2024-03-11T12:10:46Z
issue,114945,torch.multinomial - Unexpected (incorrect) results when replacement=True in version 2.1.1+cpu,https://github.com/pytorch/pytorch/issues/114945,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: random,scan when time allows,2024-03-10T17:01:26Z
issue,75391,Pytorch linalg test failure with cuda 11.6,https://github.com/pytorch/pytorch/issues/75391,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: linear algebra",scan when time allows,2024-03-09T23:35:51Z
issue,121430,Getting a CUDA tensor through a Queue from within the same process causes a segfault,https://github.com/pytorch/pytorch/issues/121430,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: multiprocessing,scan when time allows,2024-03-08T16:49:45Z
issue,121389,`torch.gather` backward performance extremely low on `torch.bfloat16`,https://github.com/pytorch/pytorch/issues/121389,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bfloat16, module: cuda, module: performance, module: scatter & gather ops",scan when time allows,2024-03-08T16:49:23Z
issue,65418,Sending tensors from short-lived multiprocessing process fails on Linux,https://github.com/pytorch/pytorch/issues/65418,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: multiprocessing,scan when time allows,2024-03-08T07:50:04Z
issue,119873,Different numerical results during training with CUDA graphs,https://github.com/pytorch/pytorch/issues/119873,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda graphs, module: numerical-reproducibility",scan when time allows,2024-03-07T16:55:23Z
issue,121131,Add flags for device that does not support fused or foreach ops,https://github.com/pytorch/pytorch/issues/121131,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp-extensions, module: mta",scan when time allows,2024-03-06T17:44:16Z
issue,121308,"installed pycuda, loss in pytorch has bug",https://github.com/pytorch/pytorch/issues/121308,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: protobuf",scan when time allows,2024-03-06T17:42:37Z
issue,121226,input_data.to produce inconsistent results,https://github.com/pytorch/pytorch/issues/121226,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: error checking, module: python frontend",scan when time allows,2024-03-06T17:42:09Z
issue,121271,`ReduceLROnPlateau` does not call `super().__init__()`,https://github.com/pytorch/pytorch/issues/121271,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: lrscheduler,scan when time allows,2024-03-06T15:41:26Z
issue,121203,How to clear GPU memory without restarting kernel when using a PyTorch model,https://github.com/pytorch/pytorch/issues/121203,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2024-03-06T15:21:20Z
issue,120708,ELU quantization parameters (scale + zp) missing from model's state_dict,https://github.com/pytorch/pytorch/issues/120708,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: serialization, oncall: quantization",scan when time allows,2024-03-06T11:24:41Z
issue,108739,"DDP Elastic ""master_addr"" resolution error in environment variables.",https://github.com/pytorch/pytorch/issues/108739,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: elastic, oncall: distributed",scan when time allows,2024-03-06T08:39:54Z
issue,82886,CUDA graph capturing fails for nn.Embedding and large batch sizes,https://github.com/pytorch/pytorch/issues/82886,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: cuda graphs, module: embedding",scan when time allows,2024-03-06T05:11:38Z
issue,89254,"`torch.nn.ReplicationPad2D` Report ""invalid configuration argument"" Error under Compute Sanitizer",https://github.com/pytorch/pytorch/issues/89254,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: error checking, module: nn, module: padding",scan when time allows,2024-03-06T01:51:58Z
issue,121215,"Build torch From Source is failed on Mac, maybe caused by protobuf",https://github.com/pytorch/pytorch/issues/121215,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: third_party",scan when time allows,2024-03-06T00:46:38Z
issue,71631,Wrapping make_graphed_callables with autocast issue,https://github.com/pytorch/pytorch/issues/71631,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision), module: cuda, module: cuda graphs",scan when time allows,2024-03-05T16:59:50Z
issue,121090,Pytorch nightly docker images set PYTORCH_VERSION variable that doesn't match the installed torch version,https://github.com/pytorch/pytorch/issues/121090,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: docker",scan when time allows,2024-03-05T15:13:47Z
issue,115331,"Please increase ""Number of CUDA devices"" recognized by pytorch.",https://github.com/pytorch/pytorch/issues/115331,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2024-03-05T11:46:46Z
issue,103652,inductor: support horizontal reduction with vec_transpose to improve TIMM swin_base_patch4_window7_224 dynamic shape performance,https://github.com/pytorch/pytorch/issues/103652,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: cpu inductor, oncall: pt2",scan when time allows,2024-03-05T07:02:53Z
issue,120930,make_graphed_callables doesn't turn off autocast for backward,https://github.com/pytorch/pytorch/issues/120930,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision), module: cuda graphs",scan when time allows,2024-03-04T20:16:31Z
issue,107433,No checks when running torch.nn.functional.ctc_loss with bogus inputs,https://github.com/pytorch/pytorch/issues/107433,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: loss",scan when time allows,2024-03-04T08:51:46Z
issue,62066,Memory Leak Found in Persistent DataLoader,https://github.com/pytorch/pytorch/issues/62066,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dataloader, module: memory usage",scan when time allows,2024-03-03T23:33:32Z
issue,13505,[feature request] Singular values and spectral norm for convolutional layers,https://github.com/pytorch/pytorch/issues/13505,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: fft, module: linear algebra",scan when time allows,2024-03-03T02:40:22Z
issue,120994,OOM with mixed precision with OPT1.3B Decoder,https://github.com/pytorch/pytorch/issues/120994,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision),scan when time allows,2024-03-01T17:18:29Z
issue,120973,TORCH_CHECK level changes in flatten,https://github.com/pytorch/pytorch/issues/120973,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: error checking, module: viewing and reshaping",scan when time allows,2024-03-01T16:59:44Z
issue,108341,`adaptive_avg_pool2d_backward_cuda` is not deterministic,https://github.com/pytorch/pytorch/issues/108341,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: determinism,scan when time allows,2024-03-01T16:53:25Z
issue,67324,torch.stft - fill_cuda not implemented for ComplexHalf,https://github.com/pytorch/pytorch/issues/67324,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: complex, module: cuda, module: half",scan when time allows,2024-03-01T09:51:58Z
issue,72766,Feature Request: Deterministic MaxPool3d and AvgPool3d,https://github.com/pytorch/pytorch/issues/72766,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: determinism, module: nn, module: pooling",scan when time allows,2024-02-29T16:00:55Z
issue,120847,[C10D] DeviceMesh + _set_default_timeout doesn't work,https://github.com/pytorch/pytorch/issues/120847,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2024-02-28T23:16:23Z
issue,120689,Tracing UpSample get RuntimeError when use_determinitic_algorithms=True,https://github.com/pytorch/pytorch/issues/120689,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: determinism, oncall: jit",scan when time allows,2024-02-27T21:44:27Z
issue,54393,[docs] torch.optim.lr_scheduler,https://github.com/pytorch/pytorch/issues/54393,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: optimizer",scan when time allows,2024-02-27T21:21:32Z
issue,44593,Docs include the verbose arg of `CosineAnnealingWarmRestarts` but the actual docstring in code doesn't,https://github.com/pytorch/pytorch/issues/44593,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: doc infra, module: docs",scan when time allows,2024-02-27T20:42:01Z
issue,120572,FCD and tensor subclasses,https://github.com/pytorch/pytorch/issues/120572,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: first class dims, module: functorch",scan when time allows,2024-02-27T15:33:55Z
issue,82073,CUDACachingAllocator should be cuda memory merge/compact friendly,https://github.com/pytorch/pytorch/issues/82073,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: cudacachingallocator",scan when time allows,2024-02-27T12:40:47Z
issue,120209,`torch.put_along_dim` and `torch.put_along_dim_` like `np.put_along_axis`,https://github.com/pytorch/pytorch/issues/120209,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: numpy, module: python frontend",scan when time allows,2024-02-27T09:48:29Z
issue,110029,Dataloader resetting with num_workers=1 and persistent_workers=True,https://github.com/pytorch/pytorch/issues/110029,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2024-02-27T09:40:33Z
issue,120291,Cannot download older version of libtorch ( libtorch-macos-1.9.0.zip ),https://github.com/pytorch/pytorch/issues/120291,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: binaries,scan when time allows,2024-02-27T00:21:13Z
issue,120466,faster communication,https://github.com/pytorch/pytorch/issues/120466,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2024-02-26T21:50:52Z
issue,117099,torch-1.11 Cross-Compiling for QNX 7.1.0 with aarch64le issue: no matching function for call to 'isnan(c10::BFloat16&)',https://github.com/pytorch/pytorch/issues/117099,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2024-02-26T16:29:29Z
issue,119190,TensorList Python binding triggering C++ TypeError,https://github.com/pytorch/pytorch/issues/119190,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp-extensions,scan when time allows,2024-02-25T03:54:30Z
issue,120463,Make upsample_bicubic2d_out_frame support BFloat16,https://github.com/pytorch/pytorch/issues/120463,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: bfloat16,scan when time allows,2024-02-24T04:53:15Z
issue,120541,[c10d] distributed.breakpoint fallback when pg is not inited,https://github.com/pytorch/pytorch/issues/120541,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2024-02-24T00:11:52Z
issue,120354,Performance degradation for `aten::fill_` and `aten::add_` in backward pass,https://github.com/pytorch/pytorch/issues/120354,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: multiprocessing,scan when time allows,2024-02-23T21:33:58Z
issue,60548,Ensure warnings relate to user code with stacklevel,https://github.com/pytorch/pytorch/issues/60548,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: deprecation,scan when time allows,2024-02-23T18:48:04Z
issue,113370,Incorrect stride when permuting shapes where a zero dimension is present.,https://github.com/pytorch/pytorch/issues/113370,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: edge cases, module: empty tensor",scan when time allows,2024-02-23T18:06:34Z
issue,39656,Broadcasting for torch.cross,https://github.com/pytorch/pytorch/issues/39656,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: numpy, module: python array api",scan when time allows,2024-02-22T23:16:11Z
issue,120396,Reinitialize parameters in TransformerEncoder,https://github.com/pytorch/pytorch/issues/120396,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-02-22T15:10:07Z
issue,120319,The CUDA download should include cusparseLt.dll since xformers now wants to use it in the context of torch,https://github.com/pytorch/pytorch/issues/120319,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: windows",scan when time allows,2024-02-22T15:01:05Z
issue,55865,Don't define __assert_fail on systems using `musl-dev` headers,https://github.com/pytorch/pytorch/issues/55865,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2024-02-21T23:32:49Z
issue,120255,Add documention to the AT_DISPATCH functionality,https://github.com/pytorch/pytorch/issues/120255,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2024-02-21T16:48:43Z
issue,120237,FFT operations on MPS are less accurate than on CPU,https://github.com/pytorch/pytorch/issues/120237,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: fft, module: mps",scan when time allows,2024-02-20T22:08:19Z
issue,120000,Update Protobuf submodule to v3.21.X,https://github.com/pytorch/pytorch/issues/120000,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: protobuf",scan when time allows,2024-02-20T21:18:02Z
issue,11340,Better user experience for using Generator object,https://github.com/pytorch/pytorch/issues/11340,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: docs, module: random",scan when time allows,2024-02-19T01:34:58Z
issue,106408,PyTorchMPS not showing up in Instruments for `torch.mps.profiler`,https://github.com/pytorch/pytorch/issues/106408,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: mps,scan when time allows,2024-02-17T19:52:12Z
issue,118580,[NestedTensor] Support for operations found commonly in point cloud analysis architectures,https://github.com/pytorch/pytorch/issues/118580,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nestedtensor,scan when time allows,2024-02-17T17:53:40Z
issue,44878,"[ONNX] Support Tuple(List(Tensors),...) in input/ouput",https://github.com/pytorch/pytorch/issues/44878,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2024-02-17T00:06:12Z
issue,120104,[memory profiler] keep consistent MB/MiB between y-axis and CUDACachingAllocator size,https://github.com/pytorch/pytorch/issues/120104,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2024-02-16T19:55:40Z
issue,119714,[NestedSubTensor] add communication operators support,https://github.com/pytorch/pytorch/issues/119714,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dtensor, module: nestedtensor, oncall: distributed",scan when time allows,2024-02-16T19:00:20Z
issue,117146,`test_triton_bsr_scatter_mm` fails on V100,https://github.com/pytorch/pytorch/issues/117146,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2024-02-16T17:44:23Z
issue,120085,Collate function in libtorch c++ API,https://github.com/pytorch/pytorch/issues/120085,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: dataloader",scan when time allows,2024-02-16T17:08:10Z
issue,78720,ValueError during `yaml.dump(dtype)`,https://github.com/pytorch/pytorch/issues/78720,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: serialization,scan when time allows,2024-02-16T12:35:17Z
issue,120037,[FSDP][BE] print kwargs in run_subtests,https://github.com/pytorch/pytorch/issues/120037,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2024-02-16T01:14:24Z
issue,119452,"RuntimeError: !block->expandable_segment_ INTERNAL ASSERT FAILED at ""../c10/cuda/CUDACachingAllocator.cpp"":2814, please report a bug to PyTorch.",https://github.com/pytorch/pytorch/issues/119452,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cudacachingallocator,scan when time allows,2024-02-15T22:07:16Z
issue,120002,[c10d] torch.distributed.breakpoint stuck with multi-thread process group,https://github.com/pytorch/pytorch/issues/120002,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2024-02-15T18:26:30Z
issue,24089,torch.nn.functional.grid_sample with 'circular' border conditions,https://github.com/pytorch/pytorch/issues/24089,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-02-15T03:29:12Z
issue,43579,"For the same complex dtype and same value, comparing a PyTorch tensor with a NumPy array results in False",https://github.com/pytorch/pytorch/issues/43579,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: complex,scan when time allows,2024-02-14T20:12:09Z
issue,68961,Deterministic implementation for upsample_bilinear2d_backward_cuda,https://github.com/pytorch/pytorch/issues/68961,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: determinism",scan when time allows,2024-02-14T19:03:59Z
issue,119843,"Docker image pytorch/pytorch:2.1.2-cuda12.1-cudnn8-runtime uses python 3.10, but default for Ubuntu 20.04 is python 3.8 => lots of issues",https://github.com/pytorch/pytorch/issues/119843,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docker,scan when time allows,2024-02-14T14:24:48Z
issue,119281,Support for dynamic DataLoader (DataLoader.__getitem__),https://github.com/pytorch/pytorch/issues/119281,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2024-02-14T11:35:56Z
issue,119711,torch.sparse.spdiags must support integer offsets,https://github.com/pytorch/pytorch/issues/119711,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2024-02-13T17:16:14Z
issue,119624,Submatrix multiplication selected by rows or columns,https://github.com/pytorch/pytorch/issues/119624,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: performance",scan when time allows,2024-02-13T17:12:19Z
issue,119633,CUDA-GCC version mapping for CUDA 11.7+,https://github.com/pytorch/pytorch/issues/119633,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp-extensions, module: cuda",scan when time allows,2024-02-13T17:07:17Z
issue,50689,RandomSampler is very slow with huge dataset,https://github.com/pytorch/pytorch/issues/50689,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2024-02-13T15:23:58Z
issue,73665,Nice way to override string representation of tensor subclasses,https://github.com/pytorch/pytorch/issues/73665,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2024-02-13T13:55:37Z
issue,40972,len of dataloader when using iterable dataset does not reflect batch size,https://github.com/pytorch/pytorch/issues/40972,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dataloader, module: nn",scan when time allows,2024-02-12T23:07:59Z
issue,119431,FSDP convergence issue with CPU offload,https://github.com/pytorch/pytorch/issues/119431,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: fsdp, oncall: distributed",scan when time allows,2024-02-12T19:37:51Z
issue,118891,Add an option for torchrun c10d store to be collocated with rank 0 training node,https://github.com/pytorch/pytorch/issues/118891,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2024-02-12T17:03:24Z
issue,119486,Conversion of 2d and 4d vectors in c++ to torch::tensor doesn't have the right data,https://github.com/pytorch/pytorch/issues/119486,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp,scan when time allows,2024-02-09T18:42:36Z
issue,118918,Expectile Loss?,https://github.com/pytorch/pytorch/issues/118918,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: loss, module: nn",scan when time allows,2024-02-09T18:17:30Z
issue,119524,"Consistency in using (mean, std) vs. (loc, scale) terminology in Normal distributions",https://github.com/pytorch/pytorch/issues/119524,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distributions,scan when time allows,2024-02-09T16:49:06Z
issue,116626,`test_modules.py` does not test for half precision or complex types,https://github.com/pytorch/pytorch/issues/116626,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ci, module: nn",scan when time allows,2024-02-08T20:51:06Z
issue,119473,Document how to override autocast rules properly,https://github.com/pytorch/pytorch/issues/119473,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision),scan when time allows,2024-02-08T20:43:22Z
issue,119387,Lazy Tensor GetBackendDevice is not working at all,https://github.com/pytorch/pytorch/issues/119387,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2024-02-08T18:18:35Z
issue,119462,Python 3.12 CPU TestBinaryUfuncsCPU.test_cumulative_trapezoid_cpu is failing,https://github.com/pytorch/pytorch/issues/119462,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: scipy compatibility, module: tests",scan when time allows,2024-02-08T18:04:03Z
issue,119464,Python 3.12 CPU TestNamedTensor::test_using_unseen_interned/uninterned tests are failing,https://github.com/pytorch/pytorch/issues/119464,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: named tensor, module: tests",scan when time allows,2024-02-08T18:03:14Z
issue,119206,load_inline should always recompile a kernel if it failed,https://github.com/pytorch/pytorch/issues/119206,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp-extensions,scan when time allows,2024-02-08T15:26:52Z
issue,99584,Training Faster R-CNN model with COCO dataset has been consistently unsuccessful.,https://github.com/pytorch/pytorch/issues/99584,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2024-02-08T13:34:58Z
issue,119418,"[c10d][NCCL] better error msg than ""Software caused connection abort""",https://github.com/pytorch/pytorch/issues/119418,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2024-02-07T22:14:38Z
issue,119417,[c10d] accept [] in torch.distributed.all_gather_object,https://github.com/pytorch/pytorch/issues/119417,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2024-02-07T22:09:25Z
issue,108861,RuntimeError: DataLoader worker (pid 11011) is killed by signal: Bus error. It is possible that dataloader's workers are out of shared memory. Please try to raise your shared memory limit,https://github.com/pytorch/pytorch/issues/108861,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: data, module: dataloader",scan when time allows,2024-02-07T18:47:45Z
issue,119377,Asked to report bug by a `UserWarning`,https://github.com/pytorch/pytorch/issues/119377,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2024-02-07T17:15:04Z
issue,93197,Add Support for RockChip NPUs (RKNN(2)),https://github.com/pytorch/pytorch/issues/93197,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: arm,scan when time allows,2024-02-07T12:14:39Z
issue,12146,Misleading step method in lr_scheduler.ReduceLROnPlateau,https://github.com/pytorch/pytorch/issues/12146,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2024-02-07T09:56:15Z
issue,117245,```at::batch_norm```: Internal assert failed,https://github.com/pytorch/pytorch/issues/117245,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: complex, module: cpp, module: nn",scan when time allows,2024-02-07T07:14:24Z
issue,118829,The inference result is wrong when I apply the torch.quantization.fuse_modules api to resnet50 model.,https://github.com/pytorch/pytorch/issues/118829,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2024-02-07T05:21:05Z
issue,119279,MODULE.bazel support,https://github.com/pytorch/pytorch/issues/119279,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bazel, module: build",scan when time allows,2024-02-06T22:40:50Z
issue,117748,[NCCL 2.19.3] Performance reduction when using NVLSTree algorithm when running nccl-tests all-reduce,https://github.com/pytorch/pytorch/issues/117748,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nccl,scan when time allows,2024-02-05T20:13:49Z
issue,118015,Disable tests UX suggestions,https://github.com/pytorch/pytorch/issues/118015,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: devx,scan when time allows,2024-02-05T19:33:44Z
issue,117647,Easier way to add a lot of tests,https://github.com/pytorch/pytorch/issues/117647,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: devx,scan when time allows,2024-02-05T19:23:02Z
issue,118970,"When concatenating empty variable with non-empty variable, I get CUDA error: illegal memory access encounter",https://github.com/pytorch/pytorch/issues/118970,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2024-02-05T18:26:07Z
issue,118634,symbolically traced variables cannot be used as inputs to control flow,https://github.com/pytorch/pytorch/issues/118634,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2024-02-05T16:57:43Z
issue,27504,[JIT] List Comprehensions With Ifs not Supported,https://github.com/pytorch/pytorch/issues/27504,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2024-02-05T14:59:21Z
issue,67092,Sparse matrix multiplication (torch.sparse.mm) NotImplementedError,https://github.com/pytorch/pytorch/issues/67092,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2024-02-05T07:26:17Z
issue,118863,Pytorch with DSA enabled? Are there binaries? Perhaps the nightly?,https://github.com/pytorch/pytorch/issues/118863,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2024-02-03T16:56:24Z
issue,50273,torch.cross is divergent from np.cross,https://github.com/pytorch/pytorch/issues/50273,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: deprecation, module: numpy",scan when time allows,2024-02-03T10:00:20Z
issue,55267,Improve CUDA extension building experience,https://github.com/pytorch/pytorch/issues/55267,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: cuda",scan when time allows,2024-02-02T21:57:39Z
issue,44343,Asynchronous Execution on CPU,https://github.com/pytorch/pytorch/issues/44343,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpu,scan when time allows,2024-02-02T21:43:37Z
issue,118723,"Libtorch torch::load(tensor, ""tensor.pt""); error",https://github.com/pytorch/pytorch/issues/118723,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: serialization, oncall: jit",scan when time allows,2024-02-02T18:41:38Z
issue,118922,torch.export(strict=False) fails with tensor constants,https://github.com/pytorch/pytorch/issues/118922,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: export,scan when time allows,2024-02-02T18:06:00Z
issue,118883,Torch Sparse Sampled AddMM Backwards Memory Usage Blowup,https://github.com/pytorch/pytorch/issues/118883,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: sparse",scan when time allows,2024-02-02T01:02:49Z
issue,118861,"Wrong typing for `argnums` in `torch.func.jacrev`, `torch.func.jacfwd` and possibly `torch.func.vmap`.",https://github.com/pytorch/pytorch/issues/118861,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: functorch, module: typing",scan when time allows,2024-02-01T22:49:01Z
issue,72117,Force PyTorch to clear CUDA cache,https://github.com/pytorch/pytorch/issues/72117,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2024-02-01T22:26:09Z
issue,59074,c++ use pybind11 to import torch free(): invalid pointer,https://github.com/pytorch/pytorch/issues/59074,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: pybind",scan when time allows,2024-02-01T19:20:13Z
issue,110250,Accessing Particular Nightly Builds Don't Work,https://github.com/pytorch/pytorch/issues/110250,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: binaries,scan when time allows,2024-02-01T17:57:32Z
issue,118859,"[W CudaIPCTypes.cpp:95] Producer process tried to deallocate over 1000 memory blocks referred by consumer processes. Deallocation might be significantly slowed down. We assume it will never going to be the case, but if it is, please file but to https://github.com/pytorch/pytorch",https://github.com/pytorch/pytorch/issues/118859,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: multiprocessing",scan when time allows,2024-02-01T13:01:45Z
issue,100419,Subgraph rewriter: Unable to match constant args,https://github.com/pytorch/pytorch/issues/100419,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2024-01-31T21:22:30Z
issue,96396,[dynamo] add hook to modify instructions before/after instructions be generated,https://github.com/pytorch/pytorch/issues/96396,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dynamo,scan when time allows,2024-01-31T20:07:56Z
issue,96386,"[export] ""strict subset of traced input/output"" error when huggingface `ModelOutput` is returned",https://github.com/pytorch/pytorch/issues/96386,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: export,scan when time allows,2024-01-31T20:06:42Z
issue,95021,Custom operations in inductor,https://github.com/pytorch/pytorch/issues/95021,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",scan when time allows,2024-01-31T19:58:00Z
issue,118736,[typing] `torch.Tensor` subclass types (eg `nn.Parameter`) broken by 2.2 release,https://github.com/pytorch/pytorch/issues/118736,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: python frontend, module: typing",scan when time allows,2024-01-31T16:02:23Z
issue,112716,Quantized model's size gets doubled on optimizing and saving it for mobile,https://github.com/pytorch/pytorch/issues/112716,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: oncall: mobile, oncall: quantization",scan when time allows,2024-01-31T07:14:40Z
issue,116567,[MPS] `sort` incorrectly handles 'negative' NaNs,https://github.com/pytorch/pytorch/issues/116567,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: mps,scan when time allows,2024-01-30T21:28:11Z
issue,74842,error in quantization by quantize_fx.prepare_fx,https://github.com/pytorch/pytorch/issues/74842,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2024-01-30T14:37:54Z
issue,46169,Squared 2-norm pdist (as available in SciPy / Faiss),https://github.com/pytorch/pytorch/issues/46169,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: distance functions, module: norms and normalization, module: numpy",scan when time allows,2024-01-30T10:03:59Z
issue,74614,compilation error with PyTorch v1.11 for CPU on ppc64le,https://github.com/pytorch/pytorch/issues/74614,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: power,scan when time allows,2024-01-30T08:23:23Z
issue,118390,Audit all global mutable state in PT2 to see if it should be thread local instead,https://github.com/pytorch/pytorch/issues/118390,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamo, module: inductor, oncall: pt2",scan when time allows,2024-01-30T00:25:05Z
issue,43760,Error : lib/libstdc++.so.6: version `CXXABI_1.3.11 not found,https://github.com/pytorch/pytorch/issues/43760,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: build",scan when time allows,2024-01-30T00:06:31Z
issue,117976,Unify test_torchinductor.copy_tests and make_test_cls_with_patches,https://github.com/pytorch/pytorch/issues/117976,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",scan when time allows,2024-01-29T23:12:08Z
issue,118373,Torch + CUDA installation,https://github.com/pytorch/pytorch/issues/118373,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2024-01-29T20:42:58Z
issue,118517,ValueError when setting tensor type to match MPS tensor type,https://github.com/pytorch/pytorch/issues/118517,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: mps,scan when time allows,2024-01-29T19:51:00Z
issue,118488,RFC: Enable disallow_untyped_defs by default,https://github.com/pytorch/pytorch/issues/118488,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: python frontend, module: typing",scan when time allows,2024-01-29T19:14:51Z
issue,118508,[Bug] Bugs when installing `torch-sparse` and `torch-scatter` in docker.,https://github.com/pytorch/pytorch/issues/118508,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docker,scan when time allows,2024-01-29T19:12:50Z
issue,118461,[DTensor] DTensor fails to execute backward of index_put op.,https://github.com/pytorch/pytorch/issues/118461,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dtensor,scan when time allows,2024-01-29T19:06:48Z
issue,118095,NAN value for truthfulqa_mc2 on full finetuned model TinyLlama,https://github.com/pytorch/pytorch/issues/118095,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fsdp,scan when time allows,2024-01-29T19:03:54Z
issue,115806,Merge all 3 mypy lint jobs,https://github.com/pytorch/pytorch/issues/115806,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: typing,scan when time allows,2024-01-27T22:31:04Z
issue,44156,CUDA memory leak in multi-processing,https://github.com/pytorch/pytorch/issues/44156,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: memory usage, module: performance",scan when time allows,2024-01-27T20:16:06Z
issue,114392,[BUG][pytree] equal `dict`s do not imply equal leaves and equal treespecs,https://github.com/pytorch/pytorch/issues/114392,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: pytree,scan when time allows,2024-01-27T17:33:43Z
issue,118370,Tests aren't cleaning up state in test_optim.py,https://github.com/pytorch/pytorch/issues/118370,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: optimizer, module: tests",scan when time allows,2024-01-26T16:56:07Z
issue,118352,Unify TEST_WITH_TORCHDYNAMO and PYTORCH_TEST_WITH_DYNAMO,https://github.com/pytorch/pytorch/issues/118352,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: testing,scan when time allows,2024-01-26T16:34:23Z
issue,118294,Failures in test_nn depending on test order,https://github.com/pytorch/pytorch/issues/118294,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: testing",scan when time allows,2024-01-26T16:23:43Z
issue,117826,[MPS] Probable incorrect gradient for einsum,https://github.com/pytorch/pytorch/issues/117826,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: mps,scan when time allows,2024-01-25T06:30:47Z
issue,118245,[FSDP] NO_SHARD numerics with cpu offloading and orig params,https://github.com/pytorch/pytorch/issues/118245,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fsdp,scan when time allows,2024-01-24T23:26:44Z
issue,118058,torch.export doesn't handle functools.partial correctly.,https://github.com/pytorch/pytorch/issues/118058,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: export",scan when time allows,2024-01-24T20:27:56Z
issue,117437,_register_state_dict_hook docstring is not accurate?,https://github.com/pytorch/pytorch/issues/117437,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-01-24T19:38:40Z
issue,107534,NNPACK slow down M1/M2 Mac CPU,https://github.com/pytorch/pytorch/issues/107534,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nnpack,scan when time allows,2024-01-24T13:54:53Z
issue,117967,"Manually modified parameter during trainning, gradients are not being correctly updated. Help",https://github.com/pytorch/pytorch/issues/117967,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2024-01-24T10:57:51Z
issue,5580,"[feature request] F.interpolate to support integral data types: bool, int8, int32, int16, int64 ||| support uint8 on CUDA",https://github.com/pytorch/pytorch/issues/5580,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2024-01-24T10:22:49Z
issue,97772,consider bumping `DEFAULT_PROTOCOL`,https://github.com/pytorch/pytorch/issues/97772,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: serialization,scan when time allows,2024-01-24T02:03:42Z
issue,117576,Warning for unusual log sizes in CI,https://github.com/pytorch/pytorch/issues/117576,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2024-01-23T21:37:15Z
issue,116896,keep-going UX is not good,https://github.com/pytorch/pytorch/issues/116896,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: devx,scan when time allows,2024-01-23T21:16:24Z
issue,115659,Clean up AWS credentials from PyTorch CI and replace them with OIDC,https://github.com/pytorch/pytorch/issues/115659,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2024-01-23T20:07:43Z
issue,118093,torch.where only accepts boolean tensor as condition deviating from np.where where you can also use binary input condition of dtype float,https://github.com/pytorch/pytorch/issues/118093,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python array api,scan when time allows,2024-01-23T19:56:24Z
issue,64327,RFC: multi-part `torch.load`/`torch.save` to support huge models and/or low CPU memory,https://github.com/pytorch/pytorch/issues/64327,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: serialization",scan when time allows,2024-01-23T19:22:27Z
issue,83540,I have the same issue as @samgelman on my MacOS.,https://github.com/pytorch/pytorch/issues/83540,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: macos, module: openmp, module: third_party",scan when time allows,2024-01-23T12:24:12Z
issue,61457,using DistributedSampler occur RuntimeError: Expected a 'cuda:0' generator device but found 'cpu',https://github.com/pytorch/pytorch/issues/61457,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dataloader, oncall: distributed",scan when time allows,2024-01-23T12:09:22Z
issue,4392,"Cache CuDNN benchmark selection, turn it on by default, use it across PyTorch runs",https://github.com/pytorch/pytorch/issues/4392,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cudnn,scan when time allows,2024-01-23T07:40:06Z
issue,116052,Add a capturable impl to ASGD single tensor,https://github.com/pytorch/pytorch/issues/116052,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2024-01-22T22:38:00Z
issue,118018,Enable all optimizers (except SparseAdam + LBFGS) to be capturable,https://github.com/pytorch/pytorch/issues/118018,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda graphs, module: optimizer",scan when time allows,2024-01-22T21:33:14Z
issue,117645,Segfault linear_search_fdes when calling c10::get_backtrace,https://github.com/pytorch/pytorch/issues/117645,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: internals,scan when time allows,2024-01-22T20:54:53Z
issue,117981,[Feature] Foreach view,https://github.com/pytorch/pytorch/issues/117981,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: mta, module: performance",scan when time allows,2024-01-22T17:44:40Z
issue,117791,[RFC][Export] Capture module attribute/annotation in `torch.export.unflatten`,https://github.com/pytorch/pytorch/issues/117791,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: export,scan when time allows,2024-01-22T17:26:29Z
issue,117938,Libtorch Don't Support Multi-thread Inference,https://github.com/pytorch/pytorch/issues/117938,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp,scan when time allows,2024-01-22T15:08:02Z
issue,117963,Does Libtorch c++ convolution operation use optimization?,https://github.com/pytorch/pytorch/issues/117963,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: cpp",scan when time allows,2024-01-22T14:09:50Z
issue,105068,[linalg] test_ops.py::test_python_ref_meta__refs_linalg_svd_cpu_complex failing,https://github.com/pytorch/pytorch/issues/105068,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: linear algebra, module: meta tensors",scan when time allows,2024-01-22T13:55:07Z
issue,94966,CSR matrix add_ error with RuntimeError: CUDA error: kernel launch failure when calling cusparseXcsrgeam2Nnz,https://github.com/pytorch/pytorch/issues/94966,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: sparse",scan when time allows,2024-01-21T10:31:36Z
issue,117910,canonical way to normalize torch.device?,https://github.com/pytorch/pytorch/issues/117910,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: python frontend",scan when time allows,2024-01-20T04:29:17Z
issue,117832,amp need a whitelist,https://github.com/pytorch/pytorch/issues/117832,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision),scan when time allows,2024-01-20T04:23:12Z
issue,117728,[DTensor] addmm strategy cannot handle 0-element input tensors,https://github.com/pytorch/pytorch/issues/117728,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dtensor,scan when time allows,2024-01-19T19:33:43Z
issue,76108,RuntimeError: [enforce fail at C:\actions-runner\_work\pytorch\pytorch\builder\windows\pytorch\caffe2\serialize\inline_container.cc:300],https://github.com/pytorch/pytorch/issues/76108,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: serialization,scan when time allows,2024-01-19T12:52:34Z
issue,117429,Memory leak while initializing tensors in child thread,https://github.com/pytorch/pytorch/issues/117429,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: memory usage",scan when time allows,2024-01-18T14:44:06Z
issue,111621,maximum Python version supported is not indicated,https://github.com/pytorch/pytorch/issues/111621,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2024-01-18T10:15:23Z
issue,35603,Performance bug with convolutions with weights and inputs of similar spatial size,https://github.com/pytorch/pytorch/issues/35603,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: cudnn, module: performance",scan when time allows,2024-01-18T09:08:46Z
issue,114632,[DeviceMesh] back DeviceMesh initialization by custom_pg,https://github.com/pytorch/pytorch/issues/114632,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: devicemesh, oncall: distributed",scan when time allows,2024-01-17T22:19:23Z
issue,116569,Python version inconsistency between Linux and Windows in the getting started,https://github.com/pytorch/pytorch/issues/116569,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2024-01-17T17:12:59Z
issue,65313,openSUSE Build: CMake Generate step failed. Build files cannot be regenerated correctly.,https://github.com/pytorch/pytorch/issues/65313,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2024-01-17T07:15:26Z
issue,117601,[FSDP] more actionable error logging in meta init,https://github.com/pytorch/pytorch/issues/117601,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fsdp,scan when time allows,2024-01-17T02:10:35Z
issue,115058,[DTensor] enforcing foreach for optimizer causes jittering,https://github.com/pytorch/pytorch/issues/115058,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dtensor,scan when time allows,2024-01-16T23:37:39Z
issue,117467,Is this an expected memory profile?,https://github.com/pytorch/pytorch/issues/117467,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: memory usage, module: multiprocessing",scan when time allows,2024-01-16T14:50:04Z
issue,117508,Consistency between torch.ao.quantization _LearnableFakeQuantize PTQ toggle and FakeQuantize PTQ toggle names.,https://github.com/pytorch/pytorch/issues/117508,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2024-01-16T14:48:19Z
issue,42594,support LSTM for quantization aware training,https://github.com/pytorch/pytorch/issues/42594,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2024-01-16T02:46:42Z
issue,117246,```at::smooth_l1_loss```: Internal assert failed,https://github.com/pytorch/pytorch/issues/117246,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: boolean tensor, module: loss",scan when time allows,2024-01-16T01:44:42Z
issue,108399,Generalize weight prepacking during quantized model deserialization,https://github.com/pytorch/pytorch/issues/108399,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2024-01-15T06:44:37Z
issue,117424,Various CI configurations are missing a slow equivalent,https://github.com/pytorch/pytorch/issues/117424,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2024-01-13T00:36:46Z
issue,113543,[PTD] Logging Improvements Main Task,https://github.com/pytorch/pytorch/issues/113543,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: c10d, oncall: distributed",scan when time allows,2024-01-12T21:12:52Z
issue,34306,[docs] Missing docs online for conv_tbc,https://github.com/pytorch/pytorch/issues/34306,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2024-01-12T20:29:45Z
issue,116691,torch.distributed doesn't have is_nccl_available() as the documentation suggests,https://github.com/pytorch/pytorch/issues/116691,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2024-01-12T16:01:45Z
issue,109194,"`torch.mean` not supported for `torch.sparse_coo_tensor`, but `torch.sum` is supported (`scipy.sparse.coo_matrix` does support both `mean` and `sum`)",https://github.com/pytorch/pytorch/issues/109194,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2024-01-12T06:54:43Z
issue,40635,[JIT] Recursive compilation doesn't apply for types used in type expressions but not value expressions,https://github.com/pytorch/pytorch/issues/40635,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2024-01-12T03:05:48Z
issue,35108,"zip(list, tuple) throws an non-actionable error message",https://github.com/pytorch/pytorch/issues/35108,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2024-01-12T03:05:35Z
issue,117259,```torch.Tensor.sum```: Internal assert failed,https://github.com/pytorch/pytorch/issues/117259,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2024-01-11T19:57:55Z
issue,117256,```torch.Tensor.round_```: Internal assert failed,https://github.com/pytorch/pytorch/issues/117256,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2024-01-11T19:57:22Z
issue,117251,"After model compilation and run on second gpu card (cuda:1) , 154 MB still goes to the first gpu card (cuda:0).Using pytorch ignite framework",https://github.com/pytorch/pytorch/issues/117251,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: inductor,scan when time allows,2024-01-11T19:56:26Z
issue,117250,```torch.nn.functional.normalize```: Internal assert failed,https://github.com/pytorch/pytorch/issues/117250,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2024-01-11T19:55:13Z
issue,117249,```torch.Tensor.triangular_solve```: Internal assert failed,https://github.com/pytorch/pytorch/issues/117249,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: boolean tensor, module: sparse",scan when time allows,2024-01-11T19:55:03Z
issue,117248,```torch.Tensor.matmul```: Internal assert failed,https://github.com/pytorch/pytorch/issues/117248,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2024-01-11T19:54:49Z
issue,117244,```torch.Tensor.index_copy_```: Internal assert failed,https://github.com/pytorch/pytorch/issues/117244,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2024-01-11T19:54:03Z
issue,117243,```torch.Tensor.cov```: Internal assertion failed,https://github.com/pytorch/pytorch/issues/117243,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2024-01-11T19:53:25Z
issue,117232,```torch.Tensor.renorm```: Internal assert failed,https://github.com/pytorch/pytorch/issues/117232,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: complex,scan when time allows,2024-01-11T19:43:56Z
issue,117247,```torch.nn.Linear``` class: Internal assertion failed,https://github.com/pytorch/pytorch/issues/117247,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2024-01-11T19:11:04Z
issue,116861,test_serialization_offset may fail with small $TMP,https://github.com/pytorch/pytorch/issues/116861,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: serialization, module: tests",scan when time allows,2024-01-10T12:26:28Z
issue,92398,[Inductor] support complex dtypes,https://github.com/pytorch/pytorch/issues/92398,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: complex, module: inductor, module: random, oncall: cpu inductor",scan when time allows,2024-01-10T12:10:17Z
issue,116633,Sparse tensor element-wise multiplication bug when broadcast needed,https://github.com/pytorch/pytorch/issues/116633,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2024-01-10T01:17:14Z
issue,116775,CI on the main branch should run with keep-going,https://github.com/pytorch/pytorch/issues/116775,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: devx,scan when time allows,2024-01-09T21:55:14Z
issue,116870,cuda FP16 default accumulation type differs between TensorIterator vs. torch.mm,https://github.com/pytorch/pytorch/issues/116870,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: numerical-stability, module: performance",scan when time allows,2024-01-09T20:58:37Z
issue,114388,Respect user-specified USE_ROCM/USE_CUDA,https://github.com/pytorch/pytorch/issues/114388,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: rocm",scan when time allows,2024-01-09T19:14:06Z
issue,96643,PyTorch SGEMV is using 1 single core on AMD CPUs (very slow),https://github.com/pytorch/pytorch/issues/96643,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rocm,scan when time allows,2024-01-09T13:46:47Z
issue,94909,[numpy] mean & nanmean should support int dtypes,https://github.com/pytorch/pytorch/issues/94909,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: numpy, module: reductions",scan when time allows,2024-01-09T13:01:40Z
issue,116980,Sharded shape attributes for DTensor,https://github.com/pytorch/pytorch/issues/116980,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2024-01-09T05:15:50Z
issue,110387,Change unsqueeze(0) to preserve memory layout contiguity of original tensor,https://github.com/pytorch/pytorch/issues/110387,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: memory format, module: numerical-reproducibility",scan when time allows,2024-01-09T02:33:15Z
issue,111509,Sparse Tensor Sum Still Does Not Work for PyTorch Geometric,https://github.com/pytorch/pytorch/issues/111509,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2024-01-08T21:23:50Z
issue,116987,[pytree] class based registration api,https://github.com/pytorch/pytorch/issues/116987,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: pytree,scan when time allows,2024-01-08T20:05:12Z
issue,116320,RuntimeError: CUDA error: an illegal memory access was encountered using vmap and model ensembling call for cuda system,https://github.com/pytorch/pytorch/issues/116320,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: cuda, module: functorch, module: vmap",scan when time allows,2024-01-08T18:59:23Z
issue,116018,Inconsistent behavior of torch.einsum on float32 and complex128,https://github.com/pytorch/pytorch/issues/116018,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: complex, module: linear algebra",scan when time allows,2024-01-08T18:07:52Z
issue,116403,pin_memory() for a custom Python object,https://github.com/pytorch/pytorch/issues/116403,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: data,scan when time allows,2024-01-06T09:42:14Z
issue,116543,Multiprocessing using Gunicorn causing CUDA Error,https://github.com/pytorch/pytorch/issues/116543,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: multiprocessing",scan when time allows,2024-01-05T12:49:08Z
issue,112509,Add a function to torch.nested to create nested tensors from a buffer and sizes,https://github.com/pytorch/pytorch/issues/112509,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nestedtensor, module: serialization",scan when time allows,2024-01-05T11:57:35Z
issue,113024,Upsample bilinear 2d decomposition does not match native implementation for uint8,https://github.com/pytorch/pytorch/issues/113024,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: decompositions,scan when time allows,2024-01-05T09:37:02Z
issue,110819,Segmentation fault on aarch64 (Rpi4) using Pytorch 2.1.0 & torchaudio,https://github.com/pytorch/pytorch/issues/110819,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: binaries,scan when time allows,2024-01-05T02:58:49Z
issue,96595,Why doesn't PyTorch install the REAL nvidia cuDNN pip package?,https://github.com/pytorch/pytorch/issues/96595,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: cuda",scan when time allows,2024-01-04T17:27:58Z
issue,114090,Nested tensors fail on Conv2D,https://github.com/pytorch/pytorch/issues/114090,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nestedtensor,scan when time allows,2024-01-04T17:09:13Z
issue,116762,parametrized should fail to generate tests with weird names,https://github.com/pytorch/pytorch/issues/116762,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: testing,scan when time allows,2024-01-04T10:37:26Z
issue,116533,`backward` implicit conversion from tuple to `torch.Tensor` results in an indexing error message,https://github.com/pytorch/pytorch/issues/116533,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: molly-guard, module: python frontend",scan when time allows,2024-01-03T09:54:47Z
issue,116576,Build with MKL and OpenMP produces libtorch_global_deps.so with undefined symbol omp_get_max_active_levels,https://github.com/pytorch/pytorch/issues/116576,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: dependency bug",scan when time allows,2024-01-03T08:30:15Z
issue,116565,`test_autograd_dense_output_addmm` fails if matrix sizes are square,https://github.com/pytorch/pytorch/issues/116565,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2024-01-03T08:25:22Z
issue,78067,Installation on Jetson target board,https://github.com/pytorch/pytorch/issues/78067,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: jetson,scan when time allows,2024-01-03T08:05:20Z
issue,116610,Align `torch.sign` with Array API,https://github.com/pytorch/pytorch/issues/116610,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python array api,scan when time allows,2024-01-02T15:57:41Z
issue,116400,Pytorch Transformer Distributed Training,https://github.com/pytorch/pytorch/issues/116400,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2024-01-02T14:31:25Z
issue,116592,Support float64 step in fused optimizers,https://github.com/pytorch/pytorch/issues/116592,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2024-01-01T20:03:15Z
issue,116558,verbose stack trace for default collate function,https://github.com/pytorch/pytorch/issues/116558,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dataloader, module: error checking",scan when time allows,2024-01-01T00:44:13Z
issue,115257,Compiling source code reported an error,https://github.com/pytorch/pytorch/issues/115257,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: nccl",scan when time allows,2023-12-31T03:05:39Z
issue,116461,Numpy like Masked/where operations,https://github.com/pytorch/pytorch/issues/116461,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: masked operators,scan when time allows,2023-12-31T03:02:05Z
issue,111368,Implement device parameter in Dropout2d,https://github.com/pytorch/pytorch/issues/111368,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2023-12-30T23:48:33Z
issue,115821,[BUG] missing automatic dtype promotion for out-of-place `torch.scatter_*()` functions,https://github.com/pytorch/pytorch/issues/115821,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: scatter & gather ops,scan when time allows,2023-12-30T03:21:59Z
issue,105319,Implementation of torch.sparse.sampled_baddmm,https://github.com/pytorch/pytorch/issues/105319,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2023-12-28T16:38:21Z
issue,116479,Address huge memory usage on large dataset,https://github.com/pytorch/pytorch/issues/116479,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2023-12-28T16:28:32Z
issue,116502,Add BC test for load_state_dict on OptimizerInfos and ModuleInfos,https://github.com/pytorch/pytorch/issues/116502,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: optimizer, module: tests",scan when time allows,2023-12-28T11:06:17Z
issue,114175,Unable to build on CUDA 11.8 due to cutlass incompatibility,https://github.com/pytorch/pytorch/issues/114175,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cuda",scan when time allows,2023-12-27T12:38:10Z
issue,1642,feature request: On-the-fly Operation Batching in Dynamic Computation Graphs,https://github.com/pytorch/pytorch/issues/1642,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2023-12-26T05:49:11Z
issue,115610,[Bug] Big difference between the output of Conv float precision and double precision,https://github.com/pytorch/pytorch/issues/115610,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: cuda, module: cudnn",scan when time allows,2023-12-23T23:58:18Z
issue,116184,Potential cublas handle leaking,https://github.com/pytorch/pytorch/issues/116184,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cublas, module: cuda",scan when time allows,2023-12-23T23:48:18Z
issue,100411,Sparse Matrix nnz Overflow when casting from COO to CSR,https://github.com/pytorch/pytorch/issues/100411,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2023-12-23T14:28:41Z
issue,75500,RuntimeError: cuDNN error: CUDNN_STATUS_EXECUTION_FAILED,https://github.com/pytorch/pytorch/issues/75500,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: cuda, module: cudnn, module: dependency bug",scan when time allows,2023-12-23T12:06:41Z
issue,113378,[pytree] implement key path API,https://github.com/pytorch/pytorch/issues/113378,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: pytree,scan when time allows,2023-12-21T20:00:35Z
issue,116255,PrivateUseOne backend does't support tensor shollow copy,https://github.com/pytorch/pytorch/issues/116255,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: backend,scan when time allows,2023-12-21T16:10:29Z
issue,74420,`torch.histogram` has wrong output dtype and doesn't support integer inputs,https://github.com/pytorch/pytorch/issues/74420,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: numpy, module: sorting and selection",scan when time allows,2023-12-21T09:34:29Z
issue,115309,"[distributed] add a ""small"" transformer model to the test suite for testing usage",https://github.com/pytorch/pytorch/issues/115309,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2023-12-21T02:12:28Z
issue,98600,Wrong illustration in README.md,https://github.com/pytorch/pytorch/issues/98600,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2023-12-21T00:50:31Z
issue,114188,Unsupported operator error: `aten::to_mkldnn` export to ONNX not supported,https://github.com/pytorch/pytorch/issues/114188,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: mkldnn, module: onnx",scan when time allows,2023-12-20T21:54:40Z
issue,116012,Issue with Protoc while building PyTorch for RISC-V,https://github.com/pytorch/pytorch/issues/116012,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2023-12-20T18:58:47Z
issue,116117,WeightedRandomSampler doesnt keep expected ratio betwen samples,https://github.com/pytorch/pytorch/issues/116117,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2023-12-20T18:54:49Z
issue,116101,[DTensor] Support API to shard to parent mesh,https://github.com/pytorch/pytorch/issues/116101,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dtensor,scan when time allows,2023-12-19T23:31:33Z
issue,115820,Question on XNNPACK support for PowerPC,https://github.com/pytorch/pytorch/issues/115820,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: power, module: third_party, module: xnnpack",scan when time allows,2023-12-19T21:28:19Z
issue,115785,arctan2 fp16 error when optimising,https://github.com/pytorch/pytorch/issues/115785,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: half, module: optimizer",scan when time allows,2023-12-19T18:04:57Z
issue,62466,Support gathering string in DataParallel,https://github.com/pytorch/pytorch/issues/62466,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: data parallel,scan when time allows,2023-12-19T11:50:46Z
issue,22169,[RFC] NestedTensor - 0.0.1,https://github.com/pytorch/pytorch/issues/22169,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: batching,scan when time allows,2023-12-19T02:41:42Z
issue,91263,Make quant_min/quant_max required for observer/fake_quant,https://github.com/pytorch/pytorch/issues/91263,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2023-12-19T01:44:53Z
issue,91604,wrong assert message,https://github.com/pytorch/pytorch/issues/91604,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2023-12-19T01:43:43Z
issue,110515,DynamicQuantizedLinear shows incorrect qscheme after applying eager mode dynamic quantization,https://github.com/pytorch/pytorch/issues/110515,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2023-12-19T01:38:20Z
issue,111902,torchscript file can not be loaded if its saved form the export model produced by torch.export.export,https://github.com/pytorch/pytorch/issues/111902,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: export,scan when time allows,2023-12-19T01:32:12Z
issue,115978,Implement KV cached attention with NestedTensor,https://github.com/pytorch/pytorch/issues/115978,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nestedtensor,scan when time allows,2023-12-18T16:21:42Z
issue,78722,AlBert quantization,https://github.com/pytorch/pytorch/issues/78722,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2023-12-18T09:41:25Z
issue,115966,Adam Eager Benchmark Failures,https://github.com/pytorch/pytorch/issues/115966,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: inductor,scan when time allows,2023-12-17T05:42:02Z
issue,23048,Add automatic tuning flags to utils.data.dataloader,https://github.com/pytorch/pytorch/issues/23048,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2023-12-16T01:12:29Z
issue,16266,support for multiple torch.cuda.max_memory_allocated() counters,https://github.com/pytorch/pytorch/issues/16266,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2023-12-15T23:18:45Z
issue,115614,Make test_public_bindings.py a lint,https://github.com/pytorch/pytorch/issues/115614,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python frontend,scan when time allows,2023-12-14T17:27:41Z
issue,112395,"KeyError during model export while using ""newer"" data types",https://github.com/pytorch/pytorch/issues/112395,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: serialization,scan when time allows,2023-12-14T16:36:18Z
issue,115672,Latest PyTorch is not buildable against CUDA-11.2,https://github.com/pytorch/pytorch/issues/115672,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cuda, module: sparse",scan when time allows,2023-12-14T15:42:50Z
issue,90695,`torch.empty` produces incorrect tensors with `layout=sparse_csr|sparse_csc` on the CPU,https://github.com/pytorch/pytorch/issues/90695,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: docs, module: sparse",scan when time allows,2023-12-14T15:05:24Z
issue,43623,[FR] Raise an exception when constructing non-empty index and empty values sparse tensor,https://github.com/pytorch/pytorch/issues/43623,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: sparse",scan when time allows,2023-12-14T14:45:40Z
issue,115735,Document required semantics of allocation functions in `CUDAPluggableAllocator`,https://github.com/pytorch/pytorch/issues/115735,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: cudacachingallocator, module: docs",scan when time allows,2023-12-14T11:08:24Z
issue,77955,NaN tensor values problem for GTX16xx users (no problem on other devices),https://github.com/pytorch/pytorch/issues/77955,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cudnn,scan when time allows,2023-12-14T10:59:39Z
issue,115624,`amax` fails on `masked_tensor` with multiple negative dims,https://github.com/pytorch/pytorch/issues/115624,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: masked operators,scan when time allows,2023-12-13T16:58:41Z
issue,114943,torch.einsum may choose a strategy for which there is not enough memory,https://github.com/pytorch/pytorch/issues/114943,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: memory usage",scan when time allows,2023-12-13T11:15:54Z
issue,70173,Dynamic Tensor Rematerialization (DTR),https://github.com/pytorch/pytorch/issues/70173,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: memory usage,scan when time allows,2023-12-12T22:31:15Z
issue,87085,gradcheck failure with sparse matrix multiplication,https://github.com/pytorch/pytorch/issues/87085,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: sparse",scan when time allows,2023-12-12T20:19:41Z
issue,115418,max_pool2d_with_indices_backward_out_cuda: remove useless code gradInput.zero_();,https://github.com/pytorch/pytorch/issues/115418,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: performance, module: pooling",scan when time allows,2023-12-12T20:17:26Z
issue,93477,[Inductor] [CPU] optimize thread parallel and loop collapse,https://github.com/pytorch/pytorch/issues/93477,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: oncall: cpu inductor, oncall: pt2",scan when time allows,2023-12-12T04:32:55Z
issue,115077,RuntimeError: cuSPARSELT not supported on your machine. When I was calling: torch._cslt_compress(),https://github.com/pytorch/pytorch/issues/115077,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: sparse",scan when time allows,2023-12-11T18:20:16Z
issue,97156,Implement `torch.distributions.Poisson.cdf()`,https://github.com/pytorch/pytorch/issues/97156,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distributions,scan when time allows,2023-12-11T08:43:30Z
issue,115340,RuntimeError: CUDA unknown error even after sudo modprobe -r nvidia_uvm && sudo modprobe nvidia_uvm,https://github.com/pytorch/pytorch/issues/115340,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2023-12-08T21:30:58Z
issue,115434,torch.kthvalue should take a tensor of `k` values.,https://github.com/pytorch/pytorch/issues/115434,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sorting and selection,scan when time allows,2023-12-08T20:30:27Z
issue,115342,torch.func.grad not working with list comprehension over dictionary,https://github.com/pytorch/pytorch/issues/115342,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: functorch,scan when time allows,2023-12-08T20:24:06Z
issue,41119,TORCH_CUDA_ARCH_LIST deprecation warning,https://github.com/pytorch/pytorch/issues/41119,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2023-12-08T20:09:52Z
issue,115422,string representation method for empty masked tensors fails,https://github.com/pytorch/pytorch/issues/115422,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: masked operators,scan when time allows,2023-12-08T20:08:45Z
issue,106780,inf and nan are mapped to quant_min in torch.fake_quantize_per_tensor_affine,https://github.com/pytorch/pytorch/issues/106780,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2023-12-08T19:00:03Z
issue,67597,torch.kthvalue (and maybe torch.quantile) should support argument descending/largest (=True) like topk/sort,https://github.com/pytorch/pytorch/issues/67597,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sorting and selection,scan when time allows,2023-12-08T18:10:53Z
issue,115421,Document PyTorch docker release policies in RELEASE.MD,https://github.com/pytorch/pytorch/issues/115421,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docker, oncall: releng",scan when time allows,2023-12-08T14:59:27Z
issue,8126,PyTorch multiprocessing using single CPU core,https://github.com/pytorch/pytorch/issues/8126,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: multiprocessing,scan when time allows,2023-12-08T13:14:35Z
issue,115164,nansum with complex support,https://github.com/pytorch/pytorch/issues/115164,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: complex, module: masked operators, module: nans and infs",scan when time allows,2023-12-08T09:08:44Z
issue,114871,Sparse block tensors (`torch.sparse`),https://github.com/pytorch/pytorch/issues/114871,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2023-12-08T08:44:16Z
issue,98872,Expand component configurable logging system to C++,https://github.com/pytorch/pytorch/issues/98872,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: logging,scan when time allows,2023-12-07T20:39:25Z
issue,115350,Enable skipped MPS OptimizerInfo tests,https://github.com/pytorch/pytorch/issues/115350,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: mps, module: optimizer",scan when time allows,2023-12-07T18:45:46Z
issue,115301,inconsistency for enableTiming / FlightRecorder / duration_ms,https://github.com/pytorch/pytorch/issues/115301,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nccl,scan when time allows,2023-12-07T15:33:50Z
issue,65663,Model Evaluation Returns Nan Values Sometimes for the Same Input,https://github.com/pytorch/pytorch/issues/65663,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nans and infs,scan when time allows,2023-12-07T09:39:53Z
issue,115202,Auto skip the entire test when @parametrize takes an empty list,https://github.com/pytorch/pytorch/issues/115202,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tests,scan when time allows,2023-12-06T15:13:39Z
issue,47513,ReduceLROnPlateau fails for negative input,https://github.com/pytorch/pytorch/issues/47513,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: optimizer",scan when time allows,2023-12-06T11:23:09Z
issue,112608,`grad_fn` is not defined in `__torch_dispatch__` arguments when lazy device is used,https://github.com/pytorch/pytorch/issues/112608,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2023-12-05T23:08:24Z
issue,114928,`torch.nn.utils.rnn.pack_padded_sequence` heap-buffer-overflow,https://github.com/pytorch/pytorch/issues/114928,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2023-12-05T21:36:25Z
issue,95423,PT2 Computes Multi Device Backward in a Single Thread,https://github.com/pytorch/pytorch/issues/95423,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",scan when time allows,2023-12-05T20:49:54Z
issue,80851,[bug][nvfuser] Applying nvfuser to the model leads to runtime error,https://github.com/pytorch/pytorch/issues/80851,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nvfuser,scan when time allows,2023-12-05T17:40:03Z
issue,115064,torch.nn.PairwiseDistance，The results vary widely from version to version,https://github.com/pytorch/pytorch/issues/115064,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2023-12-05T08:37:07Z
issue,113985,[export] Helper function for specifying dynamic batch size,https://github.com/pytorch/pytorch/issues/113985,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: export,scan when time allows,2023-12-05T00:53:38Z
issue,115034,libtorch can't switch between with and without layers by is_training(),https://github.com/pytorch/pytorch/issues/115034,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp,scan when time allows,2023-12-04T22:42:44Z
issue,115020,[Libtorch/iOS] Build for iOS as dynamic library fails in linker phase.,https://github.com/pytorch/pytorch/issues/115020,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: ios",scan when time allows,2023-12-04T22:27:08Z
issue,43536,"How do I debug ""RuntimeError: trying to initialize the default process group twice!""",https://github.com/pytorch/pytorch/issues/43536,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2023-12-04T14:46:49Z
issue,108445,_foreach_copy_ with scalar second arg,https://github.com/pytorch/pytorch/issues/108445,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: mta, module: optimizer",scan when time allows,2023-12-04T03:18:38Z
issue,114948,conda package _openmp_mutex makes pytorch dataloader slower when set num_workers > 0,https://github.com/pytorch/pytorch/issues/114948,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: binaries,scan when time allows,2023-12-03T14:34:00Z
issue,114951,Unexpected poor performance of C++ extension / wish for a fast `operator[]`,https://github.com/pytorch/pytorch/issues/114951,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: performance",scan when time allows,2023-12-02T21:12:32Z
issue,47918,Function request: Sparse matrix inverse,https://github.com/pytorch/pytorch/issues/47918,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: linear algebra, module: sparse",scan when time allows,2023-12-01T19:27:47Z
issue,111676,[export] self.buffer += 1 raises error,https://github.com/pytorch/pytorch/issues/111676,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: graph breaks, oncall: export",scan when time allows,2023-12-01T19:27:15Z
issue,114935,print statements used in torch/utils/cpp_extension.py,https://github.com/pytorch/pytorch/issues/114935,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp-extensions,scan when time allows,2023-12-01T16:59:26Z
issue,114949,"Nvidia P100, where to disable upcasting? Plus kernel image missing.",https://github.com/pytorch/pytorch/issues/114949,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: cuda",scan when time allows,2023-12-01T16:50:59Z
issue,94164,Pytorch 2.0: Detection models from torchvision don't work with onnx and tensorrt backends,https://github.com/pytorch/pytorch/issues/94164,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: onnx, oncall: pt2",scan when time allows,2023-12-01T15:30:44Z
issue,113408,Early testing stop logic for CUDA error looks wrong for instantiated_test with pytest,https://github.com/pytorch/pytorch/issues/113408,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: testing",scan when time allows,2023-12-01T07:22:07Z
issue,103475,[Inductor] Optimize More Cases of Int32 -> Int64,https://github.com/pytorch/pytorch/issues/103475,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: inductor,scan when time allows,2023-11-30T19:58:59Z
issue,101117,Barriers to using torch.compile directly in PyTorch library code,https://github.com/pytorch/pytorch/issues/101117,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamo, module: inductor, oncall: pt2",scan when time allows,2023-11-30T19:52:59Z
issue,93639,Explore TorchInductor optimization pass to reorder kernel bodies,https://github.com/pytorch/pytorch/issues/93639,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",scan when time allows,2023-11-30T03:39:47Z
issue,114808,Efficient Cholesky and QR updates,https://github.com/pytorch/pytorch/issues/114808,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2023-11-30T02:51:42Z
issue,114805,Unsupported: call_method UserDefinedObjectVariable(_profiler_enabled) __call__ [] {},https://github.com/pytorch/pytorch/issues/114805,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: graph breaks,scan when time allows,2023-11-30T02:51:10Z
issue,95604,"Performance bugs exists in multiple convolution operations(e.g., `Convtranspose2d`) when useing the `groups` argument",https://github.com/pytorch/pytorch/issues/95604,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: cudnn, module: docs",scan when time allows,2023-11-29T22:44:44Z
issue,31316,quantization - Missing operations needed for object detection,https://github.com/pytorch/pytorch/issues/31316,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2023-11-29T17:49:21Z
issue,114093,Found nn.LazyBatchNorm1d(0) has inconsistency bug between GPU and CPU testing,https://github.com/pytorch/pytorch/issues/114093,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2023-11-29T08:48:09Z
issue,113956,iSTFT gives wrong results for some batched input,https://github.com/pytorch/pytorch/issues/113956,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: cuda, module: fft",scan when time allows,2023-11-29T01:17:11Z
issue,114679,Unexpected Results in PyTorch Tensor Operations with Python Scalars,https://github.com/pytorch/pytorch/issues/114679,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: numpy,scan when time allows,2023-11-28T23:23:20Z
issue,114672,torch.fft.ifft crashes for empty input,https://github.com/pytorch/pytorch/issues/114672,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fft,scan when time allows,2023-11-28T23:13:29Z
issue,114581,Flawed testing of onesidedness in istft,https://github.com/pytorch/pytorch/issues/114581,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fft,scan when time allows,2023-11-28T23:10:46Z
issue,114440,There seems to be redundant calling of findOp in findSchemaOrThrow method of Dispatcher.cpp,https://github.com/pytorch/pytorch/issues/114440,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dispatch,scan when time allows,2023-11-28T22:19:12Z
issue,114652,__torch_dispatch__ fails on functionalization when lazy device and `out=` arg is used,https://github.com/pytorch/pytorch/issues/114652,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: __torch_dispatch__, module: functionalization, module: lazy",scan when time allows,2023-11-28T21:12:36Z
issue,106137,"`export(..., pre_dispatch=True)` for model in eval mode still inserts autograd ops",https://github.com/pytorch/pytorch/issues/106137,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: export,scan when time allows,2023-11-28T17:12:28Z
issue,106302,`torch.nn.modules.MultiheadAttention` yields different graph under pre_dispatch tracing,https://github.com/pytorch/pytorch/issues/106302,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: __torch_function__, oncall: export",scan when time allows,2023-11-28T17:12:15Z
issue,109753,fp16 parity issue with traced code on GPU,https://github.com/pytorch/pytorch/issues/109753,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: half, oncall: export, oncall: jit",scan when time allows,2023-11-28T17:11:07Z
issue,112079,The error type `UserError` in dynamo is confusing.,https://github.com/pytorch/pytorch/issues/112079,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: export,scan when time allows,2023-11-28T17:08:37Z
issue,112095,torch.export fails on a model with optional parameter,https://github.com/pytorch/pytorch/issues/112095,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: export,scan when time allows,2023-11-28T17:08:29Z
issue,113507,Revisit test_edge_op_registration on PyTorch CI,https://github.com/pytorch/pytorch/issues/113507,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: export,scan when time allows,2023-11-28T17:07:41Z
issue,113552,2.1.0 export module：Why can't export() capture reverse graphs?,https://github.com/pytorch/pytorch/issues/113552,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: export,scan when time allows,2023-11-28T17:07:34Z
issue,40134,About torch.backends.cudnn.deterministic issue,https://github.com/pytorch/pytorch/issues/40134,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cudnn,scan when time allows,2023-11-28T10:02:42Z
issue,114604,"lax.cond, lax.switch alternatives",https://github.com/pytorch/pytorch/issues/114604,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: functorch,scan when time allows,2023-11-27T19:59:40Z
issue,114350,The 'out'-parameter in torch.matmul() works for 'cuda' device but not for 'cpu',https://github.com/pytorch/pytorch/issues/114350,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: cpu, module: error checking, module: intel",scan when time allows,2023-11-27T18:15:34Z
issue,114216,torch.nn.functional.max_pool2d outputs inf,https://github.com/pytorch/pytorch/issues/114216,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: edge cases, module: error checking, module: nn",scan when time allows,2023-11-27T15:37:39Z
issue,107854,"""file_descriptor"" multiprocessing sharing strategy works incorrectly in dataloading",https://github.com/pytorch/pytorch/issues/107854,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: data, module: dataloader",scan when time allows,2023-11-25T15:35:20Z
issue,98187,how can i load seperate pytorch_model.bin?,https://github.com/pytorch/pytorch/issues/98187,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: serialization,scan when time allows,2023-11-24T15:32:58Z
issue,114080,torch.matrix_exp(x) get inf and nan,https://github.com/pytorch/pytorch/issues/114080,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: cuda",scan when time allows,2023-11-23T00:09:42Z
issue,114044,Nan on torch.corrcoef(x.t()),https://github.com/pytorch/pytorch/issues/114044,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2023-11-22T23:46:50Z
issue,112857,Vendored FindCUDAToolkit.cmake deviates from upstream in splayed installation support,https://github.com/pytorch/pytorch/issues/112857,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2023-11-22T18:16:52Z
issue,114364,Improve usability of CUDA package by adding description of CUDA,https://github.com/pytorch/pytorch/issues/114364,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2023-11-22T16:54:25Z
issue,114356,"[WARNING] could not load None, generating random data instead is too spammy",https://github.com/pytorch/pytorch/issues/114356,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: minifier,scan when time allows,2023-11-22T14:50:54Z
issue,99025,Is there a way to get the full call stack of pytorch from python to C/C++?,https://github.com/pytorch/pytorch/issues/99025,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2023-11-22T08:56:55Z
issue,62616,LazyTensor cuda fallback,https://github.com/pytorch/pytorch/issues/62616,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: lazy,scan when time allows,2023-11-21T20:49:12Z
issue,62619,LazyTensor support for fuser integration investigation,https://github.com/pytorch/pytorch/issues/62619,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: lazy,scan when time allows,2023-11-21T20:48:59Z
issue,20433,Dataloader's memory usage keeps increasing during one single epoch.,https://github.com/pytorch/pytorch/issues/20433,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2023-11-21T18:03:15Z
issue,98459,Fail to pass test HAVE_XXX_REGEX while building pytorch,https://github.com/pytorch/pytorch/issues/98459,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2023-11-21T06:13:53Z
issue,102833,LibTorch-Lite 1.13.0.1 Crash on iOS 12 on app startup,https://github.com/pytorch/pytorch/issues/102833,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ios,scan when time allows,2023-11-21T02:43:27Z
issue,2478,ReduceLROnPlateau with a naive Backtracking,https://github.com/pytorch/pytorch/issues/2478,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2023-11-20T21:45:01Z
issue,114155,Lintrunner on all files fails locally even though it passes in CI,https://github.com/pytorch/pytorch/issues/114155,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: lint,scan when time allows,2023-11-20T20:24:05Z
issue,114131,Inconsistency of state_dict loading across devices,https://github.com/pytorch/pytorch/issues/114131,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2023-11-20T19:15:59Z
issue,113905,Sympy reasoning with true division is broken,https://github.com/pytorch/pytorch/issues/113905,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dependency bug,scan when time allows,2023-11-20T18:48:13Z
issue,58969,Improper use of quantization API for MHA should fail fast,https://github.com/pytorch/pytorch/issues/58969,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2023-11-20T18:21:00Z
issue,114107,Add the possibility to pass a `Generator` to `gumbel_softmax`,https://github.com/pytorch/pytorch/issues/114107,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: random",scan when time allows,2023-11-20T17:41:51Z
issue,84290,Fix convert path for fixed qparam ops (sigmoid and softmax),https://github.com/pytorch/pytorch/issues/84290,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2023-11-20T16:34:59Z
issue,37343,qnnpack's quantized-add gives wrong result,https://github.com/pytorch/pytorch/issues/37343,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: oncall: mobile, oncall: quantization",scan when time allows,2023-11-20T15:26:16Z
issue,28906,Font used in documentation is not always sharp,https://github.com/pytorch/pytorch/issues/28906,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2023-11-20T15:22:02Z
issue,98929,torch.sparse_csr_tensor() stops gradients,https://github.com/pytorch/pytorch/issues/98929,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2023-11-20T13:24:06Z
issue,58155,"""uncorrectable NVLink error"" making the tests fail",https://github.com/pytorch/pytorch/issues/58155,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: nccl, oncall: distributed",scan when time allows,2023-11-20T08:11:52Z
issue,47885,NCCL 2.7.8 errors on PyTorch distributed process group creation,https://github.com/pytorch/pytorch/issues/47885,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2023-11-20T06:57:32Z
issue,45352,torch.nn.functional.one_hot should gracefully skip negative and out-of-range indices,https://github.com/pytorch/pytorch/issues/45352,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2023-11-19T12:20:47Z
issue,101972,No pytorch_android 2.0.x builds,https://github.com/pytorch/pytorch/issues/101972,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: android, oncall: mobile",scan when time allows,2023-11-18T14:32:06Z
issue,90613,`torch.inverse` multi-threading RuntimeError: lazy wrapper should be called at most once,https://github.com/pytorch/pytorch/issues/90613,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: linear algebra, module: multithreading",scan when time allows,2023-11-18T01:18:33Z
issue,90536,Illegal hardware instruction following Real Time Inference on Raspberry Pi 4 tutorial,https://github.com/pytorch/pytorch/issues/90536,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: arm, oncall: quantization",scan when time allows,2023-11-17T23:49:49Z
issue,46749,Quantization - we need a better solution for tracking quantization backend settings in a model,https://github.com/pytorch/pytorch/issues/46749,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2023-11-17T23:02:03Z
issue,32262,models after prepare/prepare_qat contains `qconfig` which prevents pickling,https://github.com/pytorch/pytorch/issues/32262,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2023-11-17T22:48:14Z
issue,36508,Quantized _out functions don't follow same conventions as other out functions in the codebase,https://github.com/pytorch/pytorch/issues/36508,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2023-11-17T22:41:21Z
issue,36091,[quantization] torch.quantized_lstm and torch.quantized_gru not documented,https://github.com/pytorch/pytorch/issues/36091,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, oncall: quantization",scan when time allows,2023-11-17T22:40:23Z
issue,38569,[quantization] Version support for quantization BC tests,https://github.com/pytorch/pytorch/issues/38569,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2023-11-17T22:32:26Z
issue,39058,Failure when loading quantized pre-trained weights partially,https://github.com/pytorch/pytorch/issues/39058,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2023-11-17T22:30:55Z
issue,41149,eager mode quantization should remove qconfig in the (non-leaf module of) quantized model,https://github.com/pytorch/pytorch/issues/41149,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2023-11-17T22:19:27Z
issue,42779,[quant] Quantized AdaptivePool3d is much slower for ChannelsLast3d.,https://github.com/pytorch/pytorch/issues/42779,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: memory format, module: performance, oncall: quantization",scan when time allows,2023-11-17T22:17:29Z
issue,44676,Make quantized::prepack_fp16 op just do prepacking,https://github.com/pytorch/pytorch/issues/44676,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2023-11-17T22:16:52Z
issue,45635,fx graph mode quantization tutorials,https://github.com/pytorch/pytorch/issues/45635,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2023-11-17T22:15:32Z
issue,46604,torch.concat doesn't raise an error in a quantized model,https://github.com/pytorch/pytorch/issues/46604,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2023-11-17T22:14:28Z
issue,47857,QAT with DDP should have documentation,https://github.com/pytorch/pytorch/issues/47857,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2023-11-17T22:10:57Z
issue,48106,quantization - document get_default_qconfig,https://github.com/pytorch/pytorch/issues/48106,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2023-11-17T22:09:21Z
issue,51152,Make it clearer when a tensor was quantized symmetrically,https://github.com/pytorch/pytorch/issues/51152,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2023-11-17T22:07:20Z
issue,53026,Ensure that nn.MHA and functional multi-head attention is supported by FX graph mode quantization,https://github.com/pytorch/pytorch/issues/53026,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: fx, oncall: quantization",scan when time allows,2023-11-17T22:06:09Z
issue,47535,per channel observer to work for weights of groupwise conv transpose,https://github.com/pytorch/pytorch/issues/47535,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2023-11-17T22:03:08Z
issue,54979,test_quantize_fx.py test_resnet_18_dpp test failure,https://github.com/pytorch/pytorch/issues/54979,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: fx, oncall: quantization",scan when time allows,2023-11-17T22:01:38Z
issue,56262,torch.nn.quantized.functional.conv_transpose1d/2d/3d support,https://github.com/pytorch/pytorch/issues/56262,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, oncall: quantization",scan when time allows,2023-11-17T21:59:18Z
issue,61246,"torch.quantization.fx.fuser forget to pass ""additional_fuser_method_mapping"" to fuse method",https://github.com/pytorch/pytorch/issues/61246,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: fx, oncall: quantization",scan when time allows,2023-11-17T21:53:08Z
issue,60210,Quantized model using `boolean_dispatch` not picklable,https://github.com/pytorch/pytorch/issues/60210,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2023-11-17T21:50:53Z
issue,61046,Audit use of __ARM_NEON__ define,https://github.com/pytorch/pytorch/issues/61046,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2023-11-17T21:50:08Z
issue,69158,Fusing module with multiple hooks causes `RuntimeError: OrderedDict mutated during iteration` in fuse_known_modules,https://github.com/pytorch/pytorch/issues/69158,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2023-11-17T21:48:13Z
issue,70480,Dynamic quantified Conv2d have accuracy issue,https://github.com/pytorch/pytorch/issues/70480,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2023-11-17T21:44:23Z
issue,71889,Make scale/zero_point accessible from TorchScript traced module for QFlaotFunctional modules,https://github.com/pytorch/pytorch/issues/71889,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2023-11-17T21:42:24Z
issue,71919,Graph Mode Quantization does not keep NamedTuple Information,https://github.com/pytorch/pytorch/issues/71919,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: fx, oncall: quantization",scan when time allows,2023-11-17T21:41:47Z
issue,73507,Implement SiLU method for the QuantizedCPU backend,https://github.com/pytorch/pytorch/issues/73507,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2023-11-17T21:39:36Z
issue,74393,Quantized cannot inference with cuda,https://github.com/pytorch/pytorch/issues/74393,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, oncall: quantization",scan when time allows,2023-11-17T21:37:52Z
issue,76304,Quantization fails when padding parameter given as string,https://github.com/pytorch/pytorch/issues/76304,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2023-11-17T21:33:25Z
issue,75785,"[ONNX] Expected quantizer->qscheme() == kPerTensorAffine to be true, but got false.",https://github.com/pytorch/pytorch/issues/75785,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2023-11-17T21:32:00Z
issue,74953,LSTM quantization fails if proj_size > 0,https://github.com/pytorch/pytorch/issues/74953,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2023-11-17T21:30:46Z
issue,76755,Quantization in Libtorch,https://github.com/pytorch/pytorch/issues/76755,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2023-11-17T21:28:10Z
issue,83929,ModuleNotFoundError: No module named 'torch.ao.quantization.experimental',https://github.com/pytorch/pytorch/issues/83929,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2023-11-17T21:20:25Z
issue,84064,Better error message for qlinear_prepack,https://github.com/pytorch/pytorch/issues/84064,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2023-11-17T21:14:41Z
issue,84178,Support setting strides on quantized weights of Embedding,https://github.com/pytorch/pytorch/issues/84178,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2023-11-17T21:13:16Z
issue,85745,[Quant] Remove or clarify the meaning of Nones in QConfig/BackendConfig,https://github.com/pytorch/pytorch/issues/85745,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2023-11-17T20:59:20Z
issue,85851,Print a warning when user specifies a qconfig for some node and the qconfig is not supported by BackendConfig,https://github.com/pytorch/pytorch/issues/85851,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2023-11-17T20:59:18Z
issue,86848,.view(dtype) on a quantized tensor throws SegmentationFault,https://github.com/pytorch/pytorch/issues/86848,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2023-11-17T20:54:05Z
issue,87126,Bug in Histogram Observer Implementation,https://github.com/pytorch/pytorch/issues/87126,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2023-11-17T20:45:38Z
issue,87777,Improvements to fuse optimization,https://github.com/pytorch/pytorch/issues/87777,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: fx, oncall: quantization",scan when time allows,2023-11-17T20:29:39Z
issue,88505,quantization: error message when using `convert_fx` on a model on cuda should be better,https://github.com/pytorch/pytorch/issues/88505,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2023-11-17T20:28:01Z
issue,88579,[Quant] Validate FixedQParams observers in eager mode,https://github.com/pytorch/pytorch/issues/88579,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2023-11-17T20:26:19Z
issue,88800,Quantization error between fake-quantized model and quantized model using the new observer,https://github.com/pytorch/pytorch/issues/88800,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2023-11-17T20:21:16Z
issue,89303,code sharing for fundamental ops in quantization,https://github.com/pytorch/pytorch/issues/89303,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2023-11-17T20:11:46Z
issue,89357,Quantizable LSTM has different behavior than LSTM in bidirectional setting,https://github.com/pytorch/pytorch/issues/89357,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: rnn, oncall: quantization",scan when time allows,2023-11-17T20:10:47Z
issue,90288,quantization observers: can we relax the default epsilon value?,https://github.com/pytorch/pytorch/issues/90288,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2023-11-17T20:02:02Z
issue,90424,A few functions in fbgemm_utils.cpp are defined in global namespace,https://github.com/pytorch/pytorch/issues/90424,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, oncall: quantization",scan when time allows,2023-11-17T19:56:01Z
issue,90553,Embedding dynamic quantization is not documented and hard to use,https://github.com/pytorch/pytorch/issues/90553,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2023-11-17T19:52:53Z
issue,113941,"libtorch_cpu.so: Undefined symbol ""__assert_fail""",https://github.com/pytorch/pytorch/issues/113941,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: binaries,scan when time allows,2023-11-17T18:48:23Z
issue,113847,RuntimeError in use torch 2.1.0 cuda 11.8,https://github.com/pytorch/pytorch/issues/113847,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2023-11-17T17:51:35Z
issue,113831,Update _scaled_mm to support addmm with bias.size == out.size,https://github.com/pytorch/pytorch/issues/113831,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: floatx (formerly float8),scan when time allows,2023-11-17T17:34:42Z
issue,113942,Tensorboard list of tensors as input,https://github.com/pytorch/pytorch/issues/113942,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tensorboard,scan when time allows,2023-11-17T17:14:48Z
issue,104195,Torchscript with dynamic quantization produces inconsistent model outputs,https://github.com/pytorch/pytorch/issues/104195,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: oncall: jit, oncall: quantization",scan when time allows,2023-11-17T15:53:55Z
issue,108224,qnnpack quantized model can not be traced,https://github.com/pytorch/pytorch/issues/108224,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: oncall: jit, oncall: quantization",scan when time allows,2023-11-17T15:33:39Z
issue,113748,New feature: Balanced Sampler,https://github.com/pytorch/pytorch/issues/113748,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: data, module: dataloader",scan when time allows,2023-11-16T15:11:19Z
issue,113621,App crashes when I attempt to run it with an iOS xcframework that relies on the LibTorch-Lite CocoaPod,https://github.com/pytorch/pytorch/issues/113621,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: ios",scan when time allows,2023-11-16T02:06:34Z
issue,1489,Optimizer should track parameter names and not id,https://github.com/pytorch/pytorch/issues/1489,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2023-11-15T19:22:16Z
issue,31742,sparse tensor eliminate_zeros,https://github.com/pytorch/pytorch/issues/31742,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2023-11-15T14:44:58Z
issue,11202,[pytorch][feature request] Cosine distance / simialrity between samples of own tensor or two tensors,https://github.com/pytorch/pytorch/issues/11202,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: distance functions, module: nn, module: numpy",scan when time allows,2023-11-15T11:10:12Z
issue,98860,Sharded Grad Scaler Issue Tracker,https://github.com/pytorch/pytorch/issues/98860,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision), module: fsdp, oncall: distributed",scan when time allows,2023-11-15T01:16:48Z
issue,107016,cov onnx error,https://github.com/pytorch/pytorch/issues/107016,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2023-11-15T00:14:39Z
issue,113579,Creating Gaussian Mixture Models with MultivariateNormal,https://github.com/pytorch/pytorch/issues/113579,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: distributions, module: nn",scan when time allows,2023-11-14T18:04:08Z
issue,35642,Caching Support for class Dataset,https://github.com/pytorch/pytorch/issues/35642,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2023-11-14T15:31:59Z
issue,46705,torch.eye not supported for sparse layout,https://github.com/pytorch/pytorch/issues/46705,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2023-11-14T13:10:27Z
issue,113550,CUDA extension error message doesn't look correct,https://github.com/pytorch/pytorch/issues/113550,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp-extensions, module: cuda",scan when time allows,2023-11-14T02:49:07Z
issue,113525,[Feature request] `stft` doesn't have `pad_value` argument,https://github.com/pytorch/pytorch/issues/113525,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fft,scan when time allows,2023-11-14T01:46:21Z
issue,113515,torchrun discarding --rdzv-endpoint when it should not,https://github.com/pytorch/pytorch/issues/113515,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: elastic,scan when time allows,2023-11-14T01:44:18Z
issue,112833,Failed to remove output annotation in Quantizer for PT2 QAT quantization,https://github.com/pytorch/pytorch/issues/112833,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2023-11-14T01:02:31Z
issue,73008,Followup requires for MKL link issue / cannot find -lmkl_core,https://github.com/pytorch/pytorch/issues/73008,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: cpp",scan when time allows,2023-11-13T23:27:19Z
issue,104502,Get errors after compiling and running PyTorch MINIMAL EXAMPLE for c++ Mac M1 with make,https://github.com/pytorch/pytorch/issues/104502,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: arm, module: binaries, module: cpp, module: macos",scan when time allows,2023-11-13T19:20:35Z
issue,73487,PyTorch not recognizing GPU on WSL - installed cudnn and cuda,https://github.com/pytorch/pytorch/issues/73487,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: wsl",scan when time allows,2023-11-13T09:50:41Z
issue,72814,Standardize Naming for Workflows/Jobs,https://github.com/pytorch/pytorch/issues/72814,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2023-11-11T01:42:15Z
issue,109108,Cannot construct `torch.sparse_coo_tensor` (but `scipy.sparse.coo_matrix` works fine): `TypeError: only integer tensors of a single element can be converted to an index`,https://github.com/pytorch/pytorch/issues/109108,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: scipy compatibility, module: sparse",scan when time allows,2023-11-10T23:29:35Z
issue,113449,cuDNN error: CUDNN_STATUS_MAPPING_ERROR on gtx_1080/A10 when conv1d is called,https://github.com/pytorch/pytorch/issues/113449,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: cudnn",scan when time allows,2023-11-10T15:51:42Z
issue,113369,When will Huawei Shengteng atlas be supported,https://github.com/pytorch/pytorch/issues/113369,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: backend,scan when time allows,2023-11-10T15:37:47Z
issue,113382,TorchXLA - owner @JackCaoG,https://github.com/pytorch/pytorch/issues/113382,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: xla,scan when time allows,2023-11-10T14:16:56Z
issue,113368,Gradient of image rotation,https://github.com/pytorch/pytorch/issues/113368,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: forward ad,scan when time allows,2023-11-10T14:12:57Z
issue,113326,`torch._C._cuda_getDeviceCount` inflates system memory usage,https://github.com/pytorch/pytorch/issues/113326,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2023-11-09T15:13:49Z
issue,97760,Test Failure: TestUnaryUfuncsCPU.test_reference_numerics_normal_cos_cpu_float32 on s390x,https://github.com/pytorch/pytorch/issues/97760,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: numpy,scan when time allows,2023-11-08T10:37:09Z
issue,98904,Test failure: TestCommonCPU.test_python_ref__refs_abs_cpu_complex32,https://github.com/pytorch/pytorch/issues/98904,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tests,scan when time allows,2023-11-08T10:33:14Z
issue,98566,`F.interpolate` and `F.grid_sample` - documentation error and bug,https://github.com/pytorch/pytorch/issues/98566,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2023-11-08T09:29:22Z
issue,81868,grid_sample and mode='bilinear' induces errors at discrete pixel locations,https://github.com/pytorch/pytorch/issues/81868,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2023-11-08T09:20:38Z
issue,113007,`make_fx` failing for `_scaled_dot_product_flash_attention` decomposition,https://github.com/pytorch/pytorch/issues/113007,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: decompositions, module: fx",scan when time allows,2023-11-08T04:10:24Z
issue,112849,CSR matrix on MPS,https://github.com/pytorch/pytorch/issues/112849,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: mps, module: sparse",scan when time allows,2023-11-07T19:35:06Z
issue,108023,[FSDP] Ignored modules on meta device seem to be initialized on CUDA device,https://github.com/pytorch/pytorch/issues/108023,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: fsdp, oncall: distributed",scan when time allows,2023-11-07T17:01:52Z
issue,113034,make_fx produces incorrect graph when used under FunctionalTensorMode,https://github.com/pytorch/pytorch/issues/113034,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: fx, oncall: fx",scan when time allows,2023-11-07T04:17:32Z
issue,112876,How to handle CVE vulnerabilities in underlying operating system?,https://github.com/pytorch/pytorch/issues/112876,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docker,scan when time allows,2023-11-06T22:34:04Z
issue,98814,autocast does not work properly on embedding module,https://github.com/pytorch/pytorch/issues/98814,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision),scan when time allows,2023-11-06T19:27:35Z
issue,112847,High dimensional grid sample,https://github.com/pytorch/pytorch/issues/112847,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2023-11-06T15:17:41Z
issue,88943,`torch.vander` Trigger RuntimeError with UndefinedBehaviorSanitizer,https://github.com/pytorch/pytorch/issues/88943,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: edge cases, module: linear algebra, module: sanitizers",scan when time allows,2023-11-06T00:55:45Z
issue,27617,[feature request] [dataloader] Pad variable-sized tensors in default_collate,https://github.com/pytorch/pytorch/issues/27617,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2023-11-03T19:28:11Z
issue,30401,Unexpected difference torch.multiprocessing.manager.queue and torch.multiprocessing.queue,https://github.com/pytorch/pytorch/issues/30401,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: multiprocessing",scan when time allows,2023-11-03T19:15:14Z
issue,112775,Multi-Threaded GraphModule / torch.fx inference raises an exception,https://github.com/pytorch/pytorch/issues/112775,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: fx, module: multiprocessing",scan when time allows,2023-11-03T17:31:32Z
issue,38650,Implement torch.erf for complex dtypes,https://github.com/pytorch/pytorch/issues/38650,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: complex,scan when time allows,2023-11-03T16:39:56Z
issue,41970,torch.random.fork tries to initialize cuda even when no cuda devices are available,https://github.com/pytorch/pytorch/issues/41970,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: random",scan when time allows,2023-11-03T13:39:50Z
issue,88621,"[FSDP] FSDP produces different gradient norms vs DDP, and w/ grad norm clipping creates different training results",https://github.com/pytorch/pytorch/issues/88621,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2023-11-03T13:00:52Z
issue,112714,log_softmax() on CPU and GPU has expected numerical error when used with low-precision bfloat16,https://github.com/pytorch/pytorch/issues/112714,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: cuda, module: numerical-stability",scan when time allows,2023-11-03T02:46:10Z
issue,80259,Cudnn batch norm kernel (batchnorm_bwtr_nhwc_semiPersist) gets blocked by overlapping NCCL all_reduce calls,https://github.com/pytorch/pytorch/issues/80259,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cudnn, module: dependency bug, module: memory format, module: nccl",scan when time allows,2023-11-02T22:36:30Z
issue,112691,when huawei NPU,https://github.com/pytorch/pytorch/issues/112691,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: backend,scan when time allows,2023-11-02T14:34:29Z
issue,112666,dispatcher cannot determine dispatch key on tuple input,https://github.com/pytorch/pytorch/issues/112666,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dispatch,scan when time allows,2023-11-02T14:14:27Z
issue,112715,Wrong with code_coverage/readme.md,https://github.com/pytorch/pytorch/issues/112715,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2023-11-02T14:05:54Z
issue,109112,Unable to install the latest version of PyTorch using mamba.,https://github.com/pytorch/pytorch/issues/109112,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: binaries,scan when time allows,2023-11-02T13:58:32Z
issue,100366,some of the enteries in the previous version of pytorch section are invalid,https://github.com/pytorch/pytorch/issues/100366,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: docs",scan when time allows,2023-11-02T13:58:29Z
issue,103578,"ImportError: undefined symbol: cublasSetWorkspace_v2, version libcublas.so.11",https://github.com/pytorch/pytorch/issues/103578,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: cuda",scan when time allows,2023-11-02T13:58:28Z
issue,99637,Torch hangs at import if tensorflow is imported first,https://github.com/pytorch/pytorch/issues/99637,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: binaries,scan when time allows,2023-11-02T13:58:28Z
issue,98361,pip doesn't install the right version of pytorch when torchtext is involved,https://github.com/pytorch/pytorch/issues/98361,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: binaries,scan when time allows,2023-11-02T13:58:27Z
issue,96060,Linking libtorch with QT5 OpenGL application using llvmpipe mesa opengl crashes,https://github.com/pytorch/pytorch/issues/96060,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: binaries,scan when time allows,2023-11-02T13:58:26Z
issue,92292,"""Get Started"" tells us to use the anaconda installer for PyTorch 3.x - but this should be python 3.x",https://github.com/pytorch/pytorch/issues/92292,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: binaries,scan when time allows,2023-11-02T13:58:26Z
issue,92273,Always install cpu version automatically,https://github.com/pytorch/pytorch/issues/92273,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: binaries,scan when time allows,2023-11-02T13:58:25Z
issue,80857,Compatibility List,https://github.com/pytorch/pytorch/issues/80857,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: docs",scan when time allows,2023-11-02T13:58:22Z
issue,78681,"Unable to install Preview (Nightly) on M1 macOS: ""Symbol not found""",https://github.com/pytorch/pytorch/issues/78681,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: binaries,scan when time allows,2023-11-02T13:58:22Z
issue,75158,Offical libtorch (pytorch c++ frontend) docker image,https://github.com/pytorch/pytorch/issues/75158,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: binaries,scan when time allows,2023-11-02T13:58:21Z
issue,47439,how to use torch.utils.checkpoint + gru with variable length sequence?,https://github.com/pytorch/pytorch/issues/47439,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rnn,scan when time allows,2023-11-02T13:26:34Z
issue,92226,Hijacked package names from nightly repository,https://github.com/pytorch/pytorch/issues/92226,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: binaries,scan when time allows,2023-11-01T21:29:24Z
issue,27970,can't load model on cuda after call cudaDeviceReset functions,https://github.com/pytorch/pytorch/issues/27970,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2023-11-01T07:13:29Z
issue,50198,deterministic implementation for adaptive_avg_pool2d_backward_cuda,https://github.com/pytorch/pytorch/issues/50198,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: determinism, module: pooling",scan when time allows,2023-11-01T05:28:01Z
issue,111931,[opcheck] Faster gradcheck execution,https://github.com/pytorch/pytorch/issues/111931,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: opcheck,scan when time allows,2023-10-31T20:27:32Z
issue,111654,"Static Linking C++, Op not available at runtime",https://github.com/pytorch/pytorch/issues/111654,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cpp, module: vision",scan when time allows,2023-10-31T18:40:10Z
issue,25032,[FYI] NestedTensor Project Progress,https://github.com/pytorch/pytorch/issues/25032,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nestedtensor,scan when time allows,2023-10-31T17:57:12Z
issue,20655,nn.Embedding backwards slow under high row contention,https://github.com/pytorch/pytorch/issues/20655,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: nn, module: performance",scan when time allows,2023-10-31T11:35:00Z
issue,40024,torch::jit::load -> Unhandled exception,https://github.com/pytorch/pytorch/issues/40024,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2023-10-31T10:12:26Z
issue,94233,unsqueeze a single dimension multiple times,https://github.com/pytorch/pytorch/issues/94233,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: viewing and reshaping,scan when time allows,2023-10-30T23:59:21Z
issue,2732,support grid_sample with batch=1 but supprting batch affine parameters,https://github.com/pytorch/pytorch/issues/2732,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: interpolation, module: vision",scan when time allows,2023-10-30T22:32:14Z
issue,104933,torch.func.jvp fails with BERT training,https://github.com/pytorch/pytorch/issues/104933,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: forward ad, module: functorch",scan when time allows,2023-10-30T19:35:16Z
issue,112318,IBM AfroHacks at AfroTech Participation,https://github.com/pytorch/pytorch/issues/112318,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2023-10-30T18:18:16Z
issue,112342,[pytree] Pytree node registration hygeine: deprecate global _register_pytree_node; only allow enabling registered pytree extensions locally,https://github.com/pytorch/pytorch/issues/112342,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: pytree,scan when time allows,2023-10-30T15:27:44Z
issue,112330,Library is included twice QNNPACK,https://github.com/pytorch/pytorch/issues/112330,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, oncall: mobile",scan when time allows,2023-10-29T16:34:54Z
issue,29981,[feature request] Multivariate normal CDF,https://github.com/pytorch/pytorch/issues/29981,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distributions,scan when time allows,2023-10-29T04:41:33Z
issue,111907,Migration from c10::variant to std::variant causes undefined symbols when linking against older pytorch,https://github.com/pytorch/pytorch/issues/111907,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2023-10-28T00:58:47Z
issue,103962,How to unwrap after auto_wrap in FSDP?,https://github.com/pytorch/pytorch/issues/103962,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: fsdp, oncall: distributed",scan when time allows,2023-10-27T15:16:22Z
issue,112251,Optest is unable to xfail on parametrized tests,https://github.com/pytorch/pytorch/issues/112251,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: opcheck,scan when time allows,2023-10-27T14:26:56Z
issue,109440,[FSDP] supports QLora finetuning,https://github.com/pytorch/pytorch/issues/109440,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fsdp,scan when time allows,2023-10-27T14:19:04Z
issue,110737,pytorch consuming all cpu cores 100% on ARM,https://github.com/pytorch/pytorch/issues/110737,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: arm, module: binaries, module: openmp",scan when time allows,2023-10-26T17:55:07Z
issue,112013,RuntimeError for hessian vector product with jvp,https://github.com/pytorch/pytorch/issues/112013,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: functorch,scan when time allows,2023-10-26T02:23:17Z
issue,49851,[Feature Request] SuperLoss (NeurIPS 2020),https://github.com/pytorch/pytorch/issues/49851,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: loss, module: nn",scan when time allows,2023-10-26T02:11:38Z
issue,90544,[Dispatchable Collectives] Follow up tasks,https://github.com/pytorch/pytorch/issues/90544,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2023-10-25T20:17:03Z
issue,91863,torch.Categorical samples indexes with 0 probability when given logits as argument,https://github.com/pytorch/pytorch/issues/91863,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distributions,scan when time allows,2023-10-25T13:50:01Z
issue,111789,Precisely monitor the collective communication tasks,https://github.com/pytorch/pytorch/issues/111789,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: oncall: distributed, oncall: profiler",scan when time allows,2023-10-25T02:50:40Z
issue,111925,Resize warning in two argument torch.logical_* with broadcasting,https://github.com/pytorch/pytorch/issues/111925,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: boolean tensor, module: safe resize",scan when time allows,2023-10-24T17:21:36Z
issue,111930,[opcheck] Way to reduce Hypothesis sampling when running opcheck,https://github.com/pytorch/pytorch/issues/111930,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: opcheck,scan when time allows,2023-10-24T17:03:12Z
issue,111924,[opcheck] Cannot share failures_dict between multiple tests with differing sets of tests they run,https://github.com/pytorch/pytorch/issues/111924,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: opcheck,scan when time allows,2023-10-24T15:37:29Z
issue,61816,Better argument names for torch.atan2 and other math functions,https://github.com/pytorch/pytorch/issues/61816,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: numpy,scan when time allows,2023-10-24T08:17:22Z
issue,87350,OpenCL 3.0 support: support every GPU on earth through rusticl,https://github.com/pytorch/pytorch/issues/87350,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2023-10-24T01:38:43Z
issue,111739,grad is inf/nan when using torch.amp,https://github.com/pytorch/pytorch/issues/111739,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision),scan when time allows,2023-10-23T23:56:51Z
issue,111819,__slots__ + inheriting from torch.Tensor,https://github.com/pytorch/pytorch/issues/111819,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python frontend,scan when time allows,2023-10-23T23:31:50Z
issue,110966,`torch.is_autocast_enabled()` always False on CPU,https://github.com/pytorch/pytorch/issues/110966,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision),scan when time allows,2023-10-23T17:23:24Z
issue,107538,[fx] tracing function with in-place mutation results in unexpected behaviour due to local vars becoming persisted in `GraphModule(nn.Module)`,https://github.com/pytorch/pytorch/issues/107538,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2023-10-22T01:08:11Z
issue,56978,Do padding of weight and activation tensors to match optimized backend implementation,https://github.com/pytorch/pytorch/issues/56978,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2023-10-21T07:25:57Z
issue,111623,Missing `ignored_param` when calling wrapper_cls (FSDP) recursively,https://github.com/pytorch/pytorch/issues/111623,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2023-10-21T06:50:53Z
issue,109791,Heap-buffer-overflow during tensor unpickling,https://github.com/pytorch/pytorch/issues/109791,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: serialization,scan when time allows,2023-10-20T18:24:42Z
issue,111525,Functorch FCD breaks with tensor subclasses,https://github.com/pytorch/pytorch/issues/111525,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: first class dims, module: functorch",scan when time allows,2023-10-20T17:12:28Z
issue,111552,[Bug]: some parameters' grad is None when using FSDP with torch2.1.0,https://github.com/pytorch/pytorch/issues/111552,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: fsdp, oncall: distributed",scan when time allows,2023-10-20T12:52:09Z
issue,96608,No matching distribution found for torch==1.13.1+cu117,https://github.com/pytorch/pytorch/issues/96608,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: binaries,scan when time allows,2023-10-20T07:26:55Z
issue,111566,build: failure when building pytorch with TBB,https://github.com/pytorch/pytorch/issues/111566,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: tbb",scan when time allows,2023-10-20T06:17:35Z
issue,111564,misusing percision value in test_cuda function in torch/testing/_internal/common_nn.py.,https://github.com/pytorch/pytorch/issues/111564,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: testing,scan when time allows,2023-10-20T04:36:12Z
issue,110698,max_pool3d_with_indices_backward_cuda and avg_pool3d_backward_cuda does not have a deterministic implementation,https://github.com/pytorch/pytorch/issues/110698,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: determinism, module: nn",scan when time allows,2023-10-20T02:41:40Z
issue,84193,Attach execution time to each node in an fx trace,https://github.com/pytorch/pytorch/issues/84193,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2023-10-19T20:00:23Z
issue,55363,it is lack of compatibility of CUDA_HOME (or other env),https://github.com/pytorch/pytorch/issues/55363,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp-extensions, module: cuda",scan when time allows,2023-10-19T14:06:48Z
issue,87696,Enable PostLocalSGDOptimizer on CUDA tensors,https://github.com/pytorch/pytorch/issues/87696,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, oncall: distributed",scan when time allows,2023-10-19T12:40:30Z
issue,86537,Running JIT trace for many times leads to OOM,https://github.com/pytorch/pytorch/issues/86537,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: memory usage, oncall: jit",scan when time allows,2023-10-19T11:40:40Z
issue,110959,`model.named_buffers()` fails if module not hashable.,https://github.com/pytorch/pytorch/issues/110959,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2023-10-19T08:56:45Z
issue,111466,yolov5_train,https://github.com/pytorch/pytorch/issues/111466,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: determinism",scan when time allows,2023-10-19T04:04:29Z
issue,47363,FX quantization: we should preserve original model class name through the quantization passes,https://github.com/pytorch/pytorch/issues/47363,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2023-10-18T21:49:10Z
issue,67773,Instability in `test_input_weight_equalization_activation_values` test for random test values.,https://github.com/pytorch/pytorch/issues/67773,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: tests, oncall: quantization",scan when time allows,2023-10-18T20:58:26Z
issue,60008,Dynamic quantized LSTM does not support output projection now.,https://github.com/pytorch/pytorch/issues/60008,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2023-10-18T20:55:41Z
issue,59730,Quantized conv2d with dilation and groups much slower than float32,https://github.com/pytorch/pytorch/issues/59730,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: performance, oncall: quantization",scan when time allows,2023-10-18T20:55:10Z
issue,110516,Torch Nested Issue With Backward Pass In Transpose,https://github.com/pytorch/pytorch/issues/110516,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nestedtensor,scan when time allows,2023-10-18T16:01:03Z
issue,111410,"Use of -Wl,--as-needed in cmake config files can leak into third-party users' code and modify their own private libraries",https://github.com/pytorch/pytorch/issues/111410,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2023-10-18T14:44:39Z
issue,111210,"RuntimeError: Expected is_sm80 || is_sm90 to be true, but got false. (Using Google Colab)",https://github.com/pytorch/pytorch/issues/111210,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2023-10-17T16:56:36Z
issue,102142,Pytorch CXX11 ABI version,https://github.com/pytorch/pytorch/issues/102142,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: abi, module: build",scan when time allows,2023-10-17T11:53:28Z
issue,111363,Simulating lower memory on GPU does not indicate simulated memory in error message,https://github.com/pytorch/pytorch/issues/111363,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: cudacachingallocator, module: memory usage",scan when time allows,2023-10-17T06:36:29Z
issue,110252,cuda/tf32 docs are outdated,https://github.com/pytorch/pytorch/issues/110252,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: docs, module: tf32",scan when time allows,2023-10-16T21:01:40Z
issue,111311,There is a performance drop because we have not yet implemented the batching rule for aten::mkldnn_rnn_layer_backward.,https://github.com/pytorch/pytorch/issues/111311,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: functorch,scan when time allows,2023-10-16T20:09:38Z
issue,111349,CUDA version 12.2 has differential accuracy when executing CPU and GPU,https://github.com/pytorch/pytorch/issues/111349,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2023-10-16T19:56:46Z
issue,111351,_foreach_copy_ supports fast copy between cpu and cuda devices.,https://github.com/pytorch/pytorch/issues/111351,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: mta,scan when time allows,2023-10-16T19:50:28Z
issue,111366,RuntimeError: CUDAPluggableAllocator does not yet support cacheInfo,https://github.com/pytorch/pytorch/issues/111366,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cudacachingallocator,scan when time allows,2023-10-16T19:44:08Z
issue,111186,[ONNX][Exporter] Maintain support for exporter arguments export_params and keep_initializers_as_inputs,https://github.com/pytorch/pytorch/issues/111186,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2023-10-16T17:37:32Z
issue,25646,libtorch forward memory leak,https://github.com/pytorch/pytorch/issues/25646,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: memory usage",scan when time allows,2023-10-16T02:57:10Z
issue,110795,Build process failure with torch_shm_manager,https://github.com/pytorch/pytorch/issues/110795,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2023-10-15T19:23:05Z
issue,103444,Deprecation warning on lr_scheduler.step(num_steps),https://github.com/pytorch/pytorch/issues/103444,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2023-10-13T20:11:00Z
issue,111157,overloads can perhaps be more performant?,https://github.com/pytorch/pytorch/issues/111157,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: codegen, module: dispatch, module: optimizer",scan when time allows,2023-10-13T19:44:53Z
issue,5405,scatter_add_ should support scalar source (including Python scalar),https://github.com/pytorch/pytorch/issues/5405,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: scatter & gather ops,scan when time allows,2023-10-12T18:22:39Z
issue,110363,"nn.BatchNorm2d (track_running_stats = True) causes ""modified by an in-place operation"" error when in torch.nn.parallel.DistributedDataParallel",https://github.com/pytorch/pytorch/issues/110363,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2023-10-12T17:07:19Z
issue,111025,Unprompted UserWarning,https://github.com/pytorch/pytorch/issues/111025,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2023-10-12T13:15:30Z
issue,111024,ncu python conv2d.py runs indefinitely after activating cudnn.benchmark,https://github.com/pytorch/pytorch/issues/111024,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: cudnn",scan when time allows,2023-10-12T13:15:04Z
issue,111126,type promotion test for torch.div variants is broken,https://github.com/pytorch/pytorch/issues/111126,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: testing,scan when time allows,2023-10-12T13:02:20Z
issue,94371,"QAT + torch.autocast does not work with default settings, missing fused fake_quant support for half",https://github.com/pytorch/pytorch/issues/94371,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2023-10-12T03:54:17Z
issue,80067,"FakeTensor: Support torch.tensor([FakeTensor, 0])",https://github.com/pytorch/pytorch/issues/80067,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: meta tensors,scan when time allows,2023-10-12T00:54:59Z
issue,80580,`torch.ops.aten.find` inconsistent with `str.find`,https://github.com/pytorch/pytorch/issues/80580,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: sorting and selection",scan when time allows,2023-10-11T09:17:57Z
issue,105778,https://pytorch.org/docs/stable/backends.html does not describe torch.backends.cpu,https://github.com/pytorch/pytorch/issues/105778,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: docs",scan when time allows,2023-10-10T23:47:07Z
issue,108484,DistributedDataParallel to support __getattr__,https://github.com/pytorch/pytorch/issues/108484,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2023-10-09T21:26:13Z
issue,28329,No in-place version of where(),https://github.com/pytorch/pytorch/issues/28329,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sorting and selection,scan when time allows,2023-10-09T13:25:08Z
issue,107394,[LibTorch/iOS] Unknown custom class type quantized.Conv2dPackedParamsBase. Please ensure it is registered,https://github.com/pytorch/pytorch/issues/107394,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ios, oncall: mobile",scan when time allows,2023-10-09T07:51:12Z
issue,16330,support `unique_indices` option for `unique`,https://github.com/pytorch/pytorch/issues/16330,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: numpy,scan when time allows,2023-10-08T19:02:49Z
issue,110786,.lldbinit formatters only work when building with clang,https://github.com/pytorch/pytorch/issues/110786,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2023-10-07T05:55:14Z
issue,110605,ValueError issued instead of TypeError when tensor is cast to a scalar,https://github.com/pytorch/pytorch/issues/110605,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: error checking, module: numpy",scan when time allows,2023-10-06T20:41:36Z
issue,39310,Make torch.cross `dim` parameter work intuitively,https://github.com/pytorch/pytorch/issues/39310,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bc-breaking, module: numpy",scan when time allows,2023-10-05T22:41:22Z
issue,41027,out= resizing (and restriding) behavior is confusing,https://github.com/pytorch/pytorch/issues/41027,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bc-breaking, module: numpy, module: safe resize",scan when time allows,2023-10-05T22:41:12Z
issue,46291,ProcessGroup::Work API changes,https://github.com/pytorch/pytorch/issues/46291,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bc-breaking, module: c10d",scan when time allows,2023-10-05T22:40:59Z
issue,50292,Implement dunder getattribute instead of dunder getattr for nn.Module,https://github.com/pytorch/pytorch/issues/50292,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bc-breaking, module: nn",scan when time allows,2023-10-05T22:40:41Z
issue,50342,torch.any and torch.all map uint8 -> uint8 but should map uint8 -> bool,https://github.com/pytorch/pytorch/issues/50342,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bc-breaking, module: numpy, module: reductions",scan when time allows,2023-10-05T22:40:31Z
issue,50530,Unable to compile CUDAExtension with Pytorch 1.5 or above,https://github.com/pytorch/pytorch/issues/50530,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bc-breaking, module: cpp-extensions, module: cuda",scan when time allows,2023-10-05T22:40:25Z
issue,61574,[nn] Passing dtype to `_stacklevel` argument in `log_softmax` silently works,https://github.com/pytorch/pytorch/issues/61574,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bc-breaking, module: deprecation, module: functional ux, module: nn",scan when time allows,2023-10-05T22:33:05Z
issue,68464,"Changing the type of dtype argument in softmax, log_softmax, and softmin is possibly BC-breaking.",https://github.com/pytorch/pytorch/issues/68464,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bc-breaking, module: norms and normalization",scan when time allows,2023-10-05T22:32:25Z
issue,70921,`sort` should only return the sorted input,https://github.com/pytorch/pytorch/issues/70921,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bc-breaking, module: python array api, module: sorting and selection",scan when time allows,2023-10-05T22:32:18Z
issue,70924,change supported arguments for parameter `dim` in `squeeze`,https://github.com/pytorch/pytorch/issues/70924,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bc-breaking, module: python array api",scan when time allows,2023-10-05T22:32:11Z
issue,72317,Feature request: a mode to disallow calling prototype or beta features,https://github.com/pytorch/pytorch/issues/72317,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bc-breaking, module: python frontend",scan when time allows,2023-10-05T22:32:06Z
issue,79195,"[bug] fill_, masked_fill_ : fill ops allow lossy downcasting of fill value",https://github.com/pytorch/pytorch/issues/79195,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bc-breaking, module: primtorch",scan when time allows,2023-10-05T22:31:50Z
issue,92331,backward(inputs= does not need to execute grad_fn of the inputs,https://github.com/pytorch/pytorch/issues/92331,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: bc-breaking",scan when time allows,2023-10-05T22:31:34Z
issue,95116,"When using `ceil_mode=True`, `torch.nn.AvgPool1d` could get negative shape.",https://github.com/pytorch/pytorch/issues/95116,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bc-breaking, module: shape checking",scan when time allows,2023-10-05T22:31:26Z
issue,95711,`copy.deepcopy` does not copy gradients of nn.Parameter,https://github.com/pytorch/pytorch/issues/95711,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: bc-breaking, module: nn",scan when time allows,2023-10-05T22:31:21Z
issue,109781,"`torch.embedding`, `weight[indices]`, `torch.index_select` returns random data with indices on meta device",https://github.com/pytorch/pytorch/issues/109781,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bc-breaking, module: embedding, module: meta tensors",scan when time allows,2023-10-05T22:29:17Z
issue,71409,Random Shuffle along Axis,https://github.com/pytorch/pytorch/issues/71409,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: random,scan when time allows,2023-10-05T14:07:24Z
issue,110461,Custom tensor attributes not preserved with registered functions,https://github.com/pytorch/pytorch/issues/110461,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: custom-operators, module: library",scan when time allows,2023-10-05T01:33:32Z
issue,79145,Redundant info are saved when using torch.save to save part of torch.tensor,https://github.com/pytorch/pytorch/issues/79145,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: serialization,scan when time allows,2023-10-04T23:21:52Z
issue,55279,[Feature Pitch] Full-batch optimization toolkit,https://github.com/pytorch/pytorch/issues/55279,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2023-10-04T20:26:57Z
issue,110366,Categorical Simplex constraint throws error for valid values,https://github.com/pytorch/pytorch/issues/110366,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distributions,scan when time allows,2023-10-04T19:52:18Z
issue,110356,"Dropout signature inconsistent between `torch.dropout`, `torch.nn.Dropout` and `torch.nn.functional.dropout`",https://github.com/pytorch/pytorch/issues/110356,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: python frontend",scan when time allows,2023-10-04T19:52:10Z
issue,104296,affine_grid and grid_sample operators merge/accelleration,https://github.com/pytorch/pytorch/issues/104296,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: performance,scan when time allows,2023-10-04T11:04:08Z
issue,85098,install libtorch cxx11 ABI as default in PyTorch pip installation,https://github.com/pytorch/pytorch/issues/85098,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: binaries,scan when time allows,2023-10-03T18:31:26Z
issue,71721,Improve performance of index for quantized ops,https://github.com/pytorch/pytorch/issues/71721,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2023-10-03T00:37:01Z
issue,110249,`torch.func.functional_call` does not work with `__torch_function__ ` Tensor-like objects,https://github.com/pytorch/pytorch/issues/110249,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: __torch_function__, module: functorch",scan when time allows,2023-10-02T20:47:41Z
issue,108277,Transformer performance drop due to slow PyTorch GEMMs,https://github.com/pytorch/pytorch/issues/108277,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: performance",scan when time allows,2023-10-02T13:01:11Z
issue,98939,"torch.dist with minus norm returns tensor(0.), while with -inf can return result",https://github.com/pytorch/pytorch/issues/98939,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distributions,scan when time allows,2023-10-02T07:44:31Z
issue,110238,PyTorch with non-shared build (building a single shared lib) is unsupported,https://github.com/pytorch/pytorch/issues/110238,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: static linking",scan when time allows,2023-09-30T10:36:12Z
issue,103570,File Missing When i build with C++,https://github.com/pytorch/pytorch/issues/103570,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp,scan when time allows,2023-09-29T19:34:56Z
issue,110130,Create a new heuristic TD rule for failures coming from base commit of the pull requests,https://github.com/pytorch/pytorch/issues/110130,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: devx,scan when time allows,2023-09-29T16:33:39Z
issue,110194,cudaMallocAsync cause too much fragmentation.,https://github.com/pytorch/pytorch/issues/110194,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: cudacachingallocator, module: memory usage",scan when time allows,2023-09-28T21:59:32Z
issue,47688,ARM Mac 16-core Neural Engine,https://github.com/pytorch/pytorch/issues/47688,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2023-09-28T20:03:39Z
issue,104188,Add memory managemenet information for Apple silicon mps backend,https://github.com/pytorch/pytorch/issues/104188,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: mps,scan when time allows,2023-09-28T14:47:04Z
issue,110156,Add _worker_end_fn_t to the DataLoader,https://github.com/pytorch/pytorch/issues/110156,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: data, module: dataloader",scan when time allows,2023-09-28T14:06:36Z
issue,109497,Very big differences in output of `torch.lobpcg` (values and run-time) compared to SciPy on a very ill-conditioned Laplacian matrix,https://github.com/pytorch/pytorch/issues/109497,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: linear algebra,scan when time allows,2023-09-28T09:53:54Z
issue,110116,Skip cuda kernel launch with torch.sum when dimension length is 0,https://github.com/pytorch/pytorch/issues/110116,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: performance",scan when time allows,2023-09-28T00:41:20Z
issue,110084,scatter_add: Mixing 0-dim and 1-dim tensors,https://github.com/pytorch/pytorch/issues/110084,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: edge cases, module: python frontend",scan when time allows,2023-09-28T00:40:32Z
issue,106596,[discussion] move-semantics for tensors,https://github.com/pytorch/pytorch/issues/106596,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python frontend,scan when time allows,2023-09-27T18:37:59Z
issue,109462,Inconsistent behavior for in-place operations on coalesced sparse tensors,https://github.com/pytorch/pytorch/issues/109462,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2023-09-27T16:06:38Z
issue,20102,LSTM forget bias must be initialized properly,https://github.com/pytorch/pytorch/issues/20102,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bc-breaking, module: initialization, module: nn, module: rnn",scan when time allows,2023-09-27T12:41:15Z
issue,109383,cuda rng state for 2.0.1 cannot be used for 2.1.0,https://github.com/pytorch/pytorch/issues/109383,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: random",scan when time allows,2023-09-26T21:26:42Z
issue,108174,FlashAttentionV2 will OOM when building on ci/cd with default settings,https://github.com/pytorch/pytorch/issues/108174,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ci, module: cuda",scan when time allows,2023-09-26T20:05:42Z
issue,110014,tan/tanh discrepancies with complex due to jiterator,https://github.com/pytorch/pytorch/issues/110014,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: jiterator",scan when time allows,2023-09-26T11:23:19Z
issue,110032,Race condition on shutdown involving PThreadPool and autograd,https://github.com/pytorch/pytorch/issues/110032,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: multithreading, module: sanitizers",scan when time allows,2023-09-26T10:40:53Z
issue,42080,To have single cuda context across multiple processes,https://github.com/pytorch/pytorch/issues/42080,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2023-09-26T02:16:46Z
issue,109191,Gradients across different ranks are not synchronized when using DDP,https://github.com/pytorch/pytorch/issues/109191,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2023-09-25T16:17:43Z
issue,109938,Batching for is_in,https://github.com/pytorch/pytorch/issues/109938,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: functorch,scan when time allows,2023-09-25T11:12:11Z
issue,109958,How to compile torch 2.0.1 version from source?,https://github.com/pytorch/pytorch/issues/109958,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: releng,scan when time allows,2023-09-25T11:01:11Z
issue,109946,Indexed batch matrix multiplication to support MoEs and FFFs,https://github.com/pytorch/pytorch/issues/109946,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2023-09-25T10:58:40Z
issue,109968,Dtype hard-coded in DataLoader (for python floats).,https://github.com/pytorch/pytorch/issues/109968,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: data,scan when time allows,2023-09-25T00:33:16Z
issue,40471,please add 'tensor.astype(dtype_string)' syntax for numpy interoperability,https://github.com/pytorch/pytorch/issues/40471,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: numpy,scan when time allows,2023-09-24T19:36:59Z
issue,24870,Recommendations for Grid Sample/Affine Grid/Displacement Fields/Optical Flow,https://github.com/pytorch/pytorch/issues/24870,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: interpolation,scan when time allows,2023-09-24T08:02:28Z
issue,108870,"Inconsistent, platform-dependent torch.ones_like behavior on metatensors",https://github.com/pytorch/pytorch/issues/108870,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: meta tensors,scan when time allows,2023-09-23T21:43:49Z
issue,84348,"RuntimeError: ""slow_conv2d_cpu"" not implemented for 'Half'",https://github.com/pytorch/pytorch/issues/84348,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: half,scan when time allows,2023-09-23T14:56:08Z
issue,40989,Difference between allocated and reserved CUDA memory,https://github.com/pytorch/pytorch/issues/40989,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: memory usage,scan when time allows,2023-09-23T09:29:47Z
issue,109870,Wrongly returns nan for vectorized complex numbers division on PPC/ZArch,https://github.com/pytorch/pytorch/issues/109870,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: power,scan when time allows,2023-09-23T01:19:26Z
issue,109923,Import order issue with torch and pybind11 Library Statically Linked to libstdc++,https://github.com/pytorch/pytorch/issues/109923,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: abi, module: static linking",scan when time allows,2023-09-23T01:19:21Z
issue,109934,test test_2d_fsdp_integration_fsdp_nested_param_groups failed,https://github.com/pytorch/pytorch/issues/109934,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2023-09-23T01:16:24Z
issue,109863,[BUG?] Why Allocator use stream to manage Block?,https://github.com/pytorch/pytorch/issues/109863,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2023-09-23T01:11:49Z
issue,103499,CUBLAS_WORKSPACE_CONFIG can not be parsed,https://github.com/pytorch/pytorch/issues/103499,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cublas,scan when time allows,2023-09-22T18:48:15Z
issue,106931,Other overloads of `_foreach_clamp`,https://github.com/pytorch/pytorch/issues/106931,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: mta, module: optimizer",scan when time allows,2023-09-22T18:02:33Z
issue,106951,stride of gradient is not same as the corresponding tensor,https://github.com/pytorch/pytorch/issues/106951,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: optimizer",scan when time allows,2023-09-22T18:02:25Z
issue,106989,`ray` multiprocessing interference by torch import,https://github.com/pytorch/pytorch/issues/106989,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: multiprocessing,scan when time allows,2023-09-21T22:55:25Z
issue,109724,assert_is_valid_input_type is too weak,https://github.com/pytorch/pytorch/issues/109724,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dispatch,scan when time allows,2023-09-21T13:28:06Z
issue,109747,[RFC][TorchElastic] topology info in training apps/ranks,https://github.com/pytorch/pytorch/issues/109747,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: elastic, oncall: distributed",scan when time allows,2023-09-21T13:27:38Z
issue,107389,caffe does not respect CUDNN_LIB_DIR when building from source (cmake),https://github.com/pytorch/pytorch/issues/107389,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: nvfuser",scan when time allows,2023-09-21T02:19:44Z
issue,109460,[BUG][pytree] treespec serialization for locally defined classes and namedtuple types,https://github.com/pytorch/pytorch/issues/109460,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: pytree,scan when time allows,2023-09-20T15:08:18Z
issue,108483,Efficient and robust calculation of diag(sparse @ diag @ sparse),https://github.com/pytorch/pytorch/issues/108483,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2023-09-20T13:07:15Z
issue,103581,Passing dict in datapipe/dataset will have memory leak problem,https://github.com/pytorch/pytorch/issues/103581,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: data,scan when time allows,2023-09-20T04:22:19Z
issue,109586,Max pool with negative integer inputs and channels_last memory layout gives the wrong values,https://github.com/pytorch/pytorch/issues/109586,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2023-09-19T15:16:45Z
issue,108645,uninformative OOM error,https://github.com/pytorch/pytorch/issues/108645,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2023-09-19T07:10:05Z
issue,51803,torch.fx.symbolic_trace fails on torch.arange with input-dependent size,https://github.com/pytorch/pytorch/issues/51803,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2023-09-19T04:22:35Z
issue,109494,Performance degradation on AMD + A800 when computation is small,https://github.com/pytorch/pytorch/issues/109494,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: performance",scan when time allows,2023-09-19T03:49:50Z
issue,109386,Make Fx Generating Incorrect Graph For GPTQ model,https://github.com/pytorch/pytorch/issues/109386,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: fx, module: proxytensor",scan when time allows,2023-09-18T19:47:40Z
issue,73738,torch.ao.quantization.ReuseInputObserver should also reuse the dtype of the input,https://github.com/pytorch/pytorch/issues/73738,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2023-09-18T17:00:32Z
issue,73941,"[Quant] Framework observes weight in convert, changing numerics",https://github.com/pytorch/pytorch/issues/73941,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2023-09-18T16:59:07Z
issue,74112,torch.jit.script does not custom state_dicts,https://github.com/pytorch/pytorch/issues/74112,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: oncall: jit, oncall: quantization",scan when time allows,2023-09-18T16:58:39Z
issue,55945,Quantizable LSTMCell does not work correctly.,https://github.com/pytorch/pytorch/issues/55945,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2023-09-18T16:52:52Z
issue,99225,Libtorch consumes too much memory as 16225,https://github.com/pytorch/pytorch/issues/99225,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: memory usage",scan when time allows,2023-09-18T09:13:43Z
issue,87468,Categorical fails simplex validation after its own normalisation on CUDA,https://github.com/pytorch/pytorch/issues/87468,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distributions,scan when time allows,2023-09-17T03:13:15Z
issue,109009,TypeError: mask must have dtype bool,https://github.com/pytorch/pytorch/issues/109009,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: masked operators,scan when time allows,2023-09-16T19:43:41Z
issue,8242,Different behavior of LSTM and LSTMCell implementation,https://github.com/pytorch/pytorch/issues/8242,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: numerical-reproducibility, module: numerical-stability, module: rnn",scan when time allows,2023-09-16T09:00:40Z
issue,43949,bytes(byte_tensor) gives strange error + [feature request] support memoryview(tensor),https://github.com/pytorch/pytorch/issues/43949,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: advanced indexing, module: error checking",scan when time allows,2023-09-13T18:26:10Z
issue,70926,Feature Request: Implement `torch.sparse.spdiags` and `torch.sparse.diags`,https://github.com/pytorch/pytorch/issues/70926,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2023-09-13T16:06:48Z
issue,72897,Hanging Validation,https://github.com/pytorch/pytorch/issues/72897,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision), module: cuda, module: ddp, module: deadlock",scan when time allows,2023-09-13T04:53:49Z
issue,108976,About FSDP,https://github.com/pytorch/pytorch/issues/108976,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2023-09-13T04:37:03Z
issue,108983,ZeroTensor (and probably neg/conj) doesn't play well with wrapper tensor subclasses,https://github.com/pytorch/pytorch/issues/108983,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: __torch_dispatch__,scan when time allows,2023-09-12T23:59:02Z
issue,50034,Multinomial without replacement produces samples that have zero probability,https://github.com/pytorch/pytorch/issues/50034,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: distributions, module: numpy, module: random, module: ux",scan when time allows,2023-09-12T21:43:05Z
issue,107581,[LibTorch/iOS] Building with METAL support script is freezing,https://github.com/pytorch/pytorch/issues/107581,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ios, oncall: mobile",scan when time allows,2023-09-12T20:47:26Z
issue,107627,ModuleNotFoundError: No module named 'torchgen.code_template',https://github.com/pytorch/pytorch/issues/107627,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: android, module: build, oncall: mobile",scan when time allows,2023-09-12T20:39:08Z
issue,104606,"Failure in optimize_for_mobile when using conv1d(..., padding='same')",https://github.com/pytorch/pytorch/issues/104606,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: mobile,scan when time allows,2023-09-12T20:32:16Z
issue,105494,workaround for using vmap when .item() is being used internally,https://github.com/pytorch/pytorch/issues/105494,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: functorch, module: vmap",scan when time allows,2023-09-12T19:21:14Z
issue,90551,Could not run 'aten::as_strided' with arguments from the 'Metal' backend.,https://github.com/pytorch/pytorch/issues/90551,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ios,scan when time allows,2023-09-12T07:44:05Z
issue,35636,Caffe2 generate_proposals_op_gpu_test crashes on Windows,https://github.com/pytorch/pytorch/issues/35636,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tests,scan when time allows,2023-09-12T01:08:29Z
issue,102966,"how to workaround the error ""don't have an op for vulkan_prepack::create_linear_context"" ?",https://github.com/pytorch/pytorch/issues/102966,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: vulkan",scan when time allows,2023-09-12T00:19:52Z
issue,74739,GradScaler support FP16 gradients?,https://github.com/pytorch/pytorch/issues/74739,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision),scan when time allows,2023-09-12T00:11:37Z
issue,76609,AttributeError: '_thread._local' object has no attribute 'rel_tol' (cannot use TestCase.assertEqual from other threads),https://github.com/pytorch/pytorch/issues/76609,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tests,scan when time allows,2023-09-11T19:28:51Z
issue,108882,Inconsistent any( ) between cuda and cpu - Incorrect complex to bool conversion,https://github.com/pytorch/pytorch/issues/108882,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: complex, module: correctness (silent)",scan when time allows,2023-09-11T18:41:03Z
issue,108862,"A100 runners down: apt-get install nvidia-docker2, Could not get lock /var/lib/dpkg/lock-frontend",https://github.com/pytorch/pytorch/issues/108862,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: infra,scan when time allows,2023-09-11T18:27:27Z
issue,109010,The following will always fail on NixOS,https://github.com/pytorch/pytorch/issues/109010,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cudnn,scan when time allows,2023-09-11T18:25:18Z
issue,108934,PPC64le: vsx_helpers.h errors,https://github.com/pytorch/pytorch/issues/108934,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2023-09-11T18:22:30Z
issue,108888,RuntimeError: Unrecognized tensor type ID: ZeroTensor,https://github.com/pytorch/pytorch/issues/108888,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2023-09-11T17:38:31Z
issue,108152,`Tensor.uniform_` uses illegal argument name `from`,https://github.com/pytorch/pytorch/issues/108152,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: distributions, module: python frontend",scan when time allows,2023-09-11T17:14:21Z
issue,39421,torch::jit::script::Module::to(torch::kDouble) also casts buffers,https://github.com/pytorch/pytorch/issues/39421,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2023-09-11T14:22:35Z
issue,46253,Add short guide for updating `torch.norm` calls to `torch.linalg.norm`,https://github.com/pytorch/pytorch/issues/46253,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2023-09-11T11:42:31Z
issue,106959,"`tensor.repeat` quirks: has no `torch.` variant, no `out=` variant, no inplace variant | `torch.tile` also does not have `out=` variant and uses `dims=` instead of `dim=`",https://github.com/pytorch/pytorch/issues/106959,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: python frontend, module: viewing and reshaping",scan when time allows,2023-09-10T17:29:07Z
issue,87358,Triangular solver for sparse matrices,https://github.com/pytorch/pytorch/issues/87358,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: linear algebra, module: sparse",scan when time allows,2023-09-10T15:27:24Z
issue,24185,[feature request] Subset of eigenvalues/eigenvectors,https://github.com/pytorch/pytorch/issues/24185,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: performance",scan when time allows,2023-09-10T12:22:04Z
issue,28119,Provide function similar to cdist that returns dist_p^p,https://github.com/pytorch/pytorch/issues/28119,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distance functions,scan when time allows,2023-09-10T11:04:25Z
issue,106011,Misleading error message in multilabel_margin_loss when passing incompatible tensor dimensions,https://github.com/pytorch/pytorch/issues/106011,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2023-09-10T05:20:07Z
issue,90920,Support for Transformer Models on Android with Vulkan Backend,https://github.com/pytorch/pytorch/issues/90920,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: vulkan,scan when time allows,2023-09-10T02:56:13Z
issue,108900,[c10d] fix functional collective reduce op naming convention,https://github.com/pytorch/pytorch/issues/108900,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: c10d, oncall: distributed",scan when time allows,2023-09-08T20:44:39Z
issue,94471,[Functionalization] `index_reduce_` op tests with functionalization enabled,https://github.com/pytorch/pytorch/issues/94471,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: functionalization, module: meta tensors",scan when time allows,2023-09-08T17:43:10Z
issue,97210,"INTERNAL ASSERT FAILED at ""../c10/cuda/CUDAGraphsC10Utils.h"":73, please report a bug to PyTorch. Unknown CUDA graph CaptureStatus32729",https://github.com/pytorch/pytorch/issues/97210,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda graphs,scan when time allows,2023-09-08T16:13:53Z
issue,108859,Certain torch functions are not handled by torch func wrapper,https://github.com/pytorch/pytorch/issues/108859,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nestedtensor,scan when time allows,2023-09-08T14:57:49Z
issue,108642,torch.topk returned values and indices are reordered if sorted=False,https://github.com/pytorch/pytorch/issues/108642,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sorting and selection,scan when time allows,2023-09-08T14:53:36Z
issue,108671,avg_pool3d_backward fails on meta with grad_input parameter,https://github.com/pytorch/pytorch/issues/108671,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: meta tensors,scan when time allows,2023-09-08T14:49:49Z
issue,108627,autocast not consistent across different GPUs (A100 and RTX A6000),https://github.com/pytorch/pytorch/issues/108627,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision),scan when time allows,2023-09-08T14:47:21Z
issue,108651,libtorch: runtime error when iterating batch of dataloader,https://github.com/pytorch/pytorch/issues/108651,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp,scan when time allows,2023-09-08T14:47:04Z
issue,89336,"Conv2d error on M1 mac, RuntimeError: NNPACK SpatialConvolution_updateOutput failed",https://github.com/pytorch/pytorch/issues/89336,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: arm, module: convolution, module: macos",scan when time allows,2023-09-08T14:07:52Z
issue,108851,about nccl not work,https://github.com/pytorch/pytorch/issues/108851,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: nccl",scan when time allows,2023-09-08T13:39:53Z
issue,108858,Ubuntu vs Windows: torch.cuda.OutOfMemoryError only happens on Ubuntu,https://github.com/pytorch/pytorch/issues/108858,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: memory usage,scan when time allows,2023-09-08T13:38:26Z
issue,70243,Unable to compile PyTorch when libcudart_static.so is not available,https://github.com/pytorch/pytorch/issues/70243,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cuda",scan when time allows,2023-09-08T06:44:58Z
issue,108407,torch.einsum() computes different results on cpu and cuda on A100 GPU.,https://github.com/pytorch/pytorch/issues/108407,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: linear algebra",scan when time allows,2023-09-08T04:54:48Z
issue,108047,DDP training can not accept subnet address in IPV6,https://github.com/pytorch/pytorch/issues/108047,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2023-09-08T03:04:35Z
issue,81065,[Releng] Improve the tutorials release process,https://github.com/pytorch/pytorch/issues/81065,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: ci",scan when time allows,2023-09-07T21:49:05Z
issue,81883,RuntimeError: CUDA error: no kernel image is available for execution on the device,https://github.com/pytorch/pytorch/issues/81883,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2023-09-07T21:36:09Z
issue,92838,Enable OnDemand for Open Source CI,https://github.com/pytorch/pytorch/issues/92838,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2023-09-07T17:40:41Z
issue,99918,[RFC] DebugMode,https://github.com/pytorch/pytorch/issues/99918,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2023-09-07T17:40:39Z
issue,34660,"Add ""strict"" flag to ignore missing parameters in Optimizer.load_state_dict",https://github.com/pytorch/pytorch/issues/34660,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2023-09-07T14:17:22Z
issue,34611,Eigen version for PyTorch ?,https://github.com/pytorch/pytorch/issues/34611,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2023-09-07T03:21:17Z
issue,81140,libprotobuf version compatibility,https://github.com/pytorch/pytorch/issues/81140,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build warnings,scan when time allows,2023-09-07T02:52:16Z
issue,108496,The CPU version of `torch.cummax` is slow,https://github.com/pytorch/pytorch/issues/108496,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: performance",scan when time allows,2023-09-06T18:34:57Z
issue,51778,Assertion on the existence of RRef user-facing methods' docstrings,https://github.com/pytorch/pytorch/issues/51778,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: rpc, oncall: distributed",scan when time allows,2023-09-06T13:04:22Z
issue,108231,DistributedSampler class: Change total_size into num_samples,https://github.com/pytorch/pytorch/issues/108231,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2023-09-05T19:00:44Z
issue,108569,Call for a deterministic implementation of scatter_add_cuda_kernel,https://github.com/pytorch/pytorch/issues/108569,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: scatter & gather ops,scan when time allows,2023-09-05T15:17:18Z
issue,108494,backend-friendly distributions,https://github.com/pytorch/pytorch/issues/108494,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: distributions",scan when time allows,2023-09-04T21:56:58Z
issue,75440,[quant] PackedLinearWeight::apply_dynamic_impl does not handle ReLUFused template argument,https://github.com/pytorch/pytorch/issues/75440,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2023-09-01T23:15:09Z
issue,108102,_sampled_addmm_kernel cause 'misaligned address' with new triton pin,https://github.com/pytorch/pytorch/issues/108102,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2023-09-01T21:54:11Z
issue,108244,Pytorch versions without the abi3 flag,https://github.com/pytorch/pytorch/issues/108244,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: python frontend",scan when time allows,2023-08-31T21:13:41Z
issue,106124,[feature request] Better argument checks and error messaging for `tensor.repeat`,https://github.com/pytorch/pytorch/issues/106124,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: error checking,scan when time allows,2023-08-31T20:48:25Z
issue,28754,ABI backwards compatibility,https://github.com/pytorch/pytorch/issues/28754,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: abi,scan when time allows,2023-08-31T18:20:12Z
issue,44989,"Support unfold for integral types (long, byte etc) tensors",https://github.com/pytorch/pytorch/issues/44989,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2023-08-30T15:02:39Z
issue,108197,AdaptiveMaxPool documentation is not detailed,https://github.com/pytorch/pytorch/issues/108197,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: nn, module: pooling",scan when time allows,2023-08-30T14:36:36Z
issue,45009,Median / quantile / mode / rank / percentile pooling,https://github.com/pytorch/pytorch/issues/45009,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: performance, module: pooling, module: sorting and selection",scan when time allows,2023-08-30T11:13:01Z
issue,44991,"Unite/unify tensor.unfold with F.unfold and make them more performant (zero-copy or little-copy with stride tricks, as in NumPy)",https://github.com/pytorch/pytorch/issues/44991,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: ux",scan when time allows,2023-08-30T11:10:05Z
issue,108190,[FSDP] incorrect backward prefetch order when using BackwardPrefetch.BACKWARD_POST,https://github.com/pytorch/pytorch/issues/108190,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fsdp,scan when time allows,2023-08-29T23:07:52Z
issue,108046,Enhanced Available Backend Discovery and Selection in PyTorch 2,https://github.com/pytorch/pytorch/issues/108046,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python frontend,scan when time allows,2023-08-29T18:29:13Z
issue,70919,`linspace` should support an `endpoint` parameter,https://github.com/pytorch/pytorch/issues/70919,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: python array api, module: tensor creation",scan when time allows,2023-08-29T16:25:57Z
issue,21956,nn.init.orthogonal_ doesn't work with multiprocessing,https://github.com/pytorch/pytorch/issues/21956,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dependency bug, module: initialization, module: multiprocessing, module: nn",scan when time allows,2023-08-29T14:24:16Z
issue,31942,Support sparse inputs for torch.block_diag,https://github.com/pytorch/pytorch/issues/31942,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: sparse, module: tensor creation",scan when time allows,2023-08-29T10:38:07Z
issue,107575,halo，I continue pretrain llama2-13B model ，but save state_dict is about 50GB file,https://github.com/pytorch/pytorch/issues/107575,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2023-08-29T08:56:28Z
issue,107700,where.self_out doesn't fail gracefully when inputs have different dtypes,https://github.com/pytorch/pytorch/issues/107700,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: advanced indexing, module: edge cases, module: type promotion",scan when time allows,2023-08-28T17:33:28Z
issue,107693,Inconsistencies when casting to integral types,https://github.com/pytorch/pytorch/issues/107693,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: arm, module: int overflow, module: type promotion",scan when time allows,2023-08-28T17:30:12Z
issue,104193,Conversion from strided to batched sparse compressed tensor with a non-constant number of zeros in batches fails,https://github.com/pytorch/pytorch/issues/104193,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2023-08-28T16:23:59Z
issue,108016,Failure in Initiating Pyotch DDP-style code ( Multi-machine multi-card environment),https://github.com/pytorch/pytorch/issues/108016,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2023-08-28T14:59:02Z
issue,106265,RuntimeError: Expected a proper Tensor but got None (or an undefined Tensor in C++) for argument #0 'grad_y',https://github.com/pytorch/pytorch/issues/106265,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: mps",scan when time allows,2023-08-28T14:38:01Z
issue,38090,ImportError: libtorch_cpu.so: cannot open shared object file: No such file or directory,https://github.com/pytorch/pytorch/issues/38090,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2023-08-27T19:22:00Z
issue,104761,"Torch Filename Storage hangs on ""file_system"" sharing strategy after in-place fill",https://github.com/pytorch/pytorch/issues/104761,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: mps, module: multiprocessing",scan when time allows,2023-08-26T15:09:27Z
issue,107961,"aten.lift throws error in dynamo backends -> RuntimeError: !at::functionalization::impl::isFunctionalTensor(self) INTERNAL ASSERT FAILED at ""../aten/src/ATen/FunctionalizeFallbackKernel.cpp"":167",https://github.com/pytorch/pytorch/issues/107961,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: functionalization,scan when time allows,2023-08-25T22:25:49Z
issue,107821,`torch.distributions.Pareto.sample` sometimes gives `inf`,https://github.com/pytorch/pytorch/issues/107821,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: distributions, module: nans and infs",scan when time allows,2023-08-25T19:43:27Z
issue,107864,`C10_HOST_DEVICE` for `std::isnan(c10::complex<T>)`?,https://github.com/pytorch/pytorch/issues/107864,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2023-08-24T23:04:19Z
issue,107855,About the multi-node example not working properly,https://github.com/pytorch/pytorch/issues/107855,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2023-08-24T23:02:05Z
issue,107820,`add_image_with_boxes` method from `torch.utils.tensorboard.writer.SummaryWriter` is broken,https://github.com/pytorch/pytorch/issues/107820,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: tensorboard, oncall: visualization",scan when time allows,2023-08-24T22:59:37Z
issue,34646,"Support creating a CPU tensor from ctypes pointer in Python / from_blob(ptr, shape, strides, dtype)",https://github.com/pytorch/pytorch/issues/34646,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: memory usage, module: numpy",scan when time allows,2023-08-24T20:00:55Z
issue,107701,arange.out produces incorrect output when out tensor has dtype long,https://github.com/pytorch/pytorch/issues/107701,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python frontend,scan when time allows,2023-08-24T17:47:09Z
issue,107678,[Torch.fx] Torch fx failed to trace torch extension library,https://github.com/pytorch/pytorch/issues/107678,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2023-08-24T01:44:42Z
issue,107443,Determinism by using datapipes shuffle,https://github.com/pytorch/pytorch/issues/107443,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dataloader, module: determinism",scan when time allows,2023-08-23T22:37:07Z
issue,107680,torch.nn.functional.cross_entropy different loss when providing one_hot_target and class weights,https://github.com/pytorch/pytorch/issues/107680,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: loss, module: nn",scan when time allows,2023-08-23T20:31:24Z
issue,31423,Add option in LSTM layer to access all cell states of all time steps,https://github.com/pytorch/pytorch/issues/31423,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: rnn",scan when time allows,2023-08-23T14:42:50Z
issue,45897,Unify matrix multiplications operations,https://github.com/pytorch/pytorch/issues/45897,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: linear algebra, module: sparse",scan when time allows,2023-08-23T10:26:36Z
issue,84510,[NVFuser] RuntimeError: ref_id_it != replayed_concrete_ids_.vector().end() INTERNAL ASSERT FAILED,https://github.com/pytorch/pytorch/issues/84510,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: assert failure, module: nvfuser",scan when time allows,2023-08-23T07:19:16Z
issue,70701,AdaptiveAvgPool1d - RuntimeError: shmem_size <= sharedMemPerBlockINTERNAL ASSERT FAILED,https://github.com/pytorch/pytorch/issues/70701,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2023-08-23T03:48:18Z
issue,107528,Inconsistent results when running torch.arctanh,https://github.com/pytorch/pytorch/issues/107528,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2023-08-23T00:01:59Z
issue,107555,Integer multiplication overflow when running torch.nn.MaxUnpool3d,https://github.com/pytorch/pytorch/issues/107555,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: int overflow,scan when time allows,2023-08-23T00:01:45Z
issue,107751,conv cudnn support integers,https://github.com/pytorch/pytorch/issues/107751,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cudnn,scan when time allows,2023-08-23T00:01:26Z
issue,107703,"Hardtanh docs are inaccurate/incomplete, since hardtanh behaves like clamp",https://github.com/pytorch/pytorch/issues/107703,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2023-08-22T23:58:20Z
issue,107702,Inconsistencies when handling scalars that are out of the range relative to the input tensor's dtype,https://github.com/pytorch/pytorch/issues/107702,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: int overflow,scan when time allows,2023-08-22T23:58:08Z
issue,107695,New variables in torch._ops.py pollute the torch.ops namespace,https://github.com/pytorch/pytorch/issues/107695,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: library,scan when time allows,2023-08-22T23:55:50Z
issue,107590,Select on a coalesced COO tensor returns COO tensor with coalesce flag set to False.,https://github.com/pytorch/pytorch/issues/107590,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2023-08-22T23:46:27Z
issue,107556,Integer multiplication overflow when running torch.nn.AdaptiveAvgPool2d,https://github.com/pytorch/pytorch/issues/107556,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: int overflow,scan when time allows,2023-08-22T23:43:40Z
issue,107554,Integer multiplication overflow when running torch.diagflat,https://github.com/pytorch/pytorch/issues/107554,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: int overflow,scan when time allows,2023-08-22T23:43:32Z
issue,107552,Storage size calculation overflowed when running torch.nn.functional.interpolate,https://github.com/pytorch/pytorch/issues/107552,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: int overflow,scan when time allows,2023-08-22T23:43:13Z
issue,107550,Integer multiplication overflow when running torch.eye,https://github.com/pytorch/pytorch/issues/107550,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: int overflow,scan when time allows,2023-08-22T23:42:14Z
issue,107548,Integer calculation overflow when running torch.nn.functional.adaptive_avg_pool2d,https://github.com/pytorch/pytorch/issues/107548,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: int overflow,scan when time allows,2023-08-22T23:42:05Z
issue,107546,Integer overflow when running torch.nn.functional.upsample_bilinear,https://github.com/pytorch/pytorch/issues/107546,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: int overflow,scan when time allows,2023-08-22T23:41:55Z
issue,107545,Integer overflow when running torch.nn.functional.upsample,https://github.com/pytorch/pytorch/issues/107545,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: int overflow,scan when time allows,2023-08-22T23:41:45Z
issue,107544,Integer overflow when running torch.nn.ReplicationPad3d,https://github.com/pytorch/pytorch/issues/107544,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: int overflow,scan when time allows,2023-08-22T23:38:54Z
issue,107543,Integer overflow when running torch.nn.AdaptiveAvgPool2d,https://github.com/pytorch/pytorch/issues/107543,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: int overflow,scan when time allows,2023-08-22T23:38:46Z
issue,107541,Integer overflow when running torch.nn.MaxUnpool2d,https://github.com/pytorch/pytorch/issues/107541,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: int overflow,scan when time allows,2023-08-22T23:38:34Z
issue,107536,Appending new logs to existing tbevent files when using tensorboard,https://github.com/pytorch/pytorch/issues/107536,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tensorboard,scan when time allows,2023-08-22T23:36:49Z
issue,107494,Mismatch in type of error raised when reducing along empty slice between eager and primtorch,https://github.com/pytorch/pytorch/issues/107494,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: primtorch,scan when time allows,2023-08-22T23:31:20Z
issue,63188,ROCm build documentation lacks needed dependencies,https://github.com/pytorch/pytorch/issues/63188,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: docs, module: rocm",scan when time allows,2023-08-22T21:29:17Z
issue,77576,`stateless.functional_call` doesn't work with `nn.DataParallel`,https://github.com/pytorch/pytorch/issues/77576,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: data parallel, module: nn",scan when time allows,2023-08-22T19:20:41Z
issue,51351,Run time error,https://github.com/pytorch/pytorch/issues/51351,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cublas, module: cuda, module: dependency bug, module: wsl",scan when time allows,2023-08-22T13:39:50Z
issue,51210,"Complex->Real cast is a warning, calling `real` or `imag` on non-complex tensors is an Error.",https://github.com/pytorch/pytorch/issues/51210,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: complex, module: ux",scan when time allows,2023-08-22T02:00:56Z
issue,45833,Support torch.mean for BoolTensors and other integer tensor inputs (without manual upcasting and hopefully without hidden upcasting),https://github.com/pytorch/pytorch/issues/45833,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: numpy, module: reductions, module: type promotion",scan when time allows,2023-08-21T22:11:07Z
issue,107412,"max_pool1d, max_pool2d, max_pool3d Integers for cpu and cuda",https://github.com/pytorch/pytorch/issues/107412,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: pooling,scan when time allows,2023-08-21T21:43:53Z
issue,107514,conv2d wrong results on 3090/3090ti,https://github.com/pytorch/pytorch/issues/107514,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: half,scan when time allows,2023-08-21T21:21:56Z
issue,103530,Deduplicate the operands passed into torch.cond after dynamo tracing.,https://github.com/pytorch/pytorch/issues/103530,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2023-08-21T20:08:47Z
issue,51280,"[ux] Proposal to have t() === transpose(-1, -2), since batches are very frequent",https://github.com/pytorch/pytorch/issues/51280,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: numpy, module: ux",scan when time allows,2023-08-21T19:12:21Z
issue,107539,Integer overflow when running torch.nn.functional.max_unpool2d,https://github.com/pytorch/pytorch/issues/107539,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: int overflow,scan when time allows,2023-08-21T16:34:48Z
issue,84052,Index out of bounds Error with PerChannel Quantization,https://github.com/pytorch/pytorch/issues/84052,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2023-08-21T15:26:24Z
issue,40759,Quantization support for F.Softmax,https://github.com/pytorch/pytorch/issues/40759,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2023-08-21T14:27:27Z
issue,69900,Port MarginRankingLoss to TensorIterator,https://github.com/pytorch/pytorch/issues/69900,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: loss, module: nn, module: performance",scan when time allows,2023-08-21T13:28:15Z
issue,70238,`TensorIterator`: refactor `build_ternary_op` to match binary versions,https://github.com/pytorch/pytorch/issues/70238,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tensoriterator,scan when time allows,2023-08-21T13:28:12Z
issue,72134,Add support for `complex` `mean` to the `normal` operator,https://github.com/pytorch/pytorch/issues/72134,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: complex, module: distributions",scan when time allows,2023-08-21T13:28:03Z
issue,74952,[feature request] Make `index_select` parallel.,https://github.com/pytorch/pytorch/issues/74952,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: advanced indexing, module: performance",scan when time allows,2023-08-21T13:27:52Z
issue,105105,Remaining functions without meta registrations,https://github.com/pytorch/pytorch/issues/105105,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: meta tensors,scan when time allows,2023-08-21T13:27:28Z
issue,106622,FFT Samples Inputs with More than Three Dimensions,https://github.com/pytorch/pytorch/issues/106622,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: fft, module: tests",scan when time allows,2023-08-21T10:50:37Z
issue,106584,Lacking commutativity of `tensor.expand` and `tensor.flatten`,https://github.com/pytorch/pytorch/issues/106584,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: viewing and reshaping,scan when time allows,2023-08-20T21:13:48Z
issue,92818,Segfault when running torch.atan2,https://github.com/pytorch/pytorch/issues/92818,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: complex,scan when time allows,2023-08-20T13:03:33Z
issue,87794,CUDA error: operation not permitted when stream is capturing (2 GPUs),https://github.com/pytorch/pytorch/issues/87794,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: multi-gpu,scan when time allows,2023-08-20T12:11:45Z
issue,35872,torch.autograd.set_detect_anomaly(True) does not exist in C++?,https://github.com/pytorch/pytorch/issues/35872,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp,scan when time allows,2023-08-20T05:23:59Z
issue,107392,Overly strict type hints for `torch.utils.data.random_split`,https://github.com/pytorch/pytorch/issues/107392,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dataloader, module: typing",scan when time allows,2023-08-20T04:59:21Z
issue,107374,DataParallel scatter method split tensor wrong,https://github.com/pytorch/pytorch/issues/107374,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: data parallel",scan when time allows,2023-08-18T12:54:01Z
issue,28244,ProcessGroupMPI reports incorrect world size,https://github.com/pytorch/pytorch/issues/28244,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2023-08-18T02:42:40Z
issue,107387,Incorrect type hint for `torch.library.Library.define`,https://github.com/pytorch/pytorch/issues/107387,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: library, module: typing",scan when time allows,2023-08-17T22:16:29Z
issue,107321,"using Union[str, Tensor] as an argument to a torch.jit.script function",https://github.com/pytorch/pytorch/issues/107321,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2023-08-17T15:39:43Z
issue,65528,I cannot use x.to(GPU) or x.cuda(GPU),https://github.com/pytorch/pytorch/issues/65528,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: cuda graphs",scan when time allows,2023-08-17T14:40:37Z
issue,106467,nn.CrossEntropyLoss with invalid target generates corrups memory eventualy leading to CUDA error: an illegal memory access,https://github.com/pytorch/pytorch/issues/106467,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: loss, module: nn",scan when time allows,2023-08-17T14:30:16Z
issue,104506,TImeout in NCCL doesn't work,https://github.com/pytorch/pytorch/issues/104506,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2023-08-17T08:40:32Z
issue,107253,Conda configuration shouldn't pollute $PATH variable,https://github.com/pytorch/pytorch/issues/107253,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ci, oncall: releng",scan when time allows,2023-08-17T03:09:18Z
issue,107217,Documenting `IterableDataset`'s needing `StopIteration` for finite data,https://github.com/pytorch/pytorch/issues/107217,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: data,scan when time allows,2023-08-17T02:50:31Z
issue,104147,PyTorch2.0 ROCM LayerNorm HIP error: invalid configuration,https://github.com/pytorch/pytorch/issues/104147,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rocm,scan when time allows,2023-08-16T16:01:38Z
issue,104776,torch.nn.CrossEntropyLoss: class weighting changes label_smoothing,https://github.com/pytorch/pytorch/issues/104776,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: loss, module: nn",scan when time allows,2023-08-16T15:40:18Z
issue,107125,combining `vmap` with NN containing `MaxPool2d' leads to discrepancies in output,https://github.com/pytorch/pytorch/issues/107125,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: functorch,scan when time allows,2023-08-16T14:39:50Z
issue,50276,torch.meshgrid is divergent from np.meshgrid,https://github.com/pytorch/pytorch/issues/50276,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: deprecation, module: numpy",scan when time allows,2023-08-16T07:18:06Z
issue,68429,istft gradcheck fails on ROCm,https://github.com/pytorch/pytorch/issues/68429,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: fft, module: rocm",scan when time allows,2023-08-15T21:51:03Z
issue,107201,The difference between channels last backward and channels first backward of AvgPool2d on CUDA is too large,https://github.com/pytorch/pytorch/issues/107201,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: memory format, module: nn",scan when time allows,2023-08-15T21:18:53Z
issue,103482,Document CI retry rules,https://github.com/pytorch/pytorch/issues/103482,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: devx,scan when time allows,2023-08-15T19:44:31Z
issue,100158,`torch.sparse_csc_tensor` matrix multiplication produces MKL error SPARSE_STATUS_ALLOC_FAILED when density is too high,https://github.com/pytorch/pytorch/issues/100158,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2023-08-15T14:29:41Z
issue,107102,Error when using sparse_coo tensor with optimizer,https://github.com/pytorch/pytorch/issues/107102,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2023-08-14T20:04:54Z
issue,46904,Authentication for RPC,https://github.com/pytorch/pytorch/issues/46904,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: rpc, module: tensorpipe, oncall: distributed",scan when time allows,2023-08-12T03:27:41Z
issue,81667,[Tracker] AO migration of quantization from `torch.nn` to `torch.ao.nn`,https://github.com/pytorch/pytorch/issues/81667,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2023-08-11T18:49:42Z
issue,107011,max_pool3d_with_indices_backward_cuda does not have a deterministic implementation,https://github.com/pytorch/pytorch/issues/107011,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: determinism, module: pooling",scan when time allows,2023-08-11T13:47:33Z
issue,26889,Statically checked tensor shapes,https://github.com/pytorch/pytorch/issues/26889,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: internals,scan when time allows,2023-08-11T13:16:22Z
issue,77053,Incorrect documentation in ``gumble_softmax`` function.,https://github.com/pytorch/pytorch/issues/77053,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: nn",scan when time allows,2023-08-11T09:19:30Z
issue,79709,ONEDNN testing is not done properly in quantization codebase,https://github.com/pytorch/pytorch/issues/79709,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2023-08-10T20:08:56Z
issue,79307,`prepare_qat_fx` docstring doesn't run,https://github.com/pytorch/pytorch/issues/79307,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: fx, oncall: quantization",scan when time allows,2023-08-10T20:02:56Z
issue,81287,Runtime error in Libtorch cpp project (Didn't find engine for operation quantized::conv2d_prepack NoQEngine),https://github.com/pytorch/pytorch/issues/81287,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2023-08-10T20:02:24Z
issue,79739,quantization: misleading backend config for linear_dynamic_fp16,https://github.com/pytorch/pytorch/issues/79739,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2023-08-10T20:02:02Z
issue,81381,quantization: QConfigMapping should be easy to print,https://github.com/pytorch/pytorch/issues/81381,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2023-08-10T19:59:12Z
issue,83672,quantization: unexpected casting of tensor min and max to int in histogram observer,https://github.com/pytorch/pytorch/issues/83672,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2023-08-10T19:57:37Z
issue,84148,[Quant] Reference get_default_qconfig_mapping in docs/tutorials,https://github.com/pytorch/pytorch/issues/84148,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2023-08-10T19:51:15Z
issue,34772,"Unsupported ONNX op (Upsample 3D, bicubic) contrary to documentation",https://github.com/pytorch/pytorch/issues/34772,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: vision",scan when time allows,2023-08-10T18:08:16Z
issue,83773,[Nested Tensor] view + inplace for autograd.,https://github.com/pytorch/pytorch/issues/83773,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: nestedtensor",scan when time allows,2023-08-10T16:22:00Z
issue,70239,`pytorch` hangs during interaction with `ray` package,https://github.com/pytorch/pytorch/issues/70239,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: deadlock",scan when time allows,2023-08-10T12:21:34Z
issue,18095,torch.flip is inconsistent with np.flip and also uses `dims` arg instead of `dim`,https://github.com/pytorch/pytorch/issues/18095,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: numpy,scan when time allows,2023-08-10T09:05:04Z
issue,22331,Integer division by Zero giving large number results instead of NaN/inf on Windows,https://github.com/pytorch/pytorch/issues/22331,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpu,scan when time allows,2023-08-09T19:41:13Z
issue,35881,Error when loading jit traced FasterRCNN model in C++,https://github.com/pytorch/pytorch/issues/35881,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2023-08-09T19:38:21Z
issue,27815,Improve the error message when trying to install in a 32-bit Python environment,https://github.com/pytorch/pytorch/issues/27815,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2023-08-09T19:36:57Z
issue,67955,Torch 1.9.1.post3 - Error in magma_getdevice_arch: MAGMA not initialized (call magma_init() first) or bad device,https://github.com/pytorch/pytorch/issues/67955,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2023-08-09T18:47:02Z
issue,99297,Slicing and indexing support negative steps,https://github.com/pytorch/pytorch/issues/99297,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: advanced indexing,scan when time allows,2023-08-09T16:51:12Z
issue,106828,Use expect tests for error inputs,https://github.com/pytorch/pytorch/issues/106828,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: testing,scan when time allows,2023-08-09T03:21:20Z
issue,95946,"Build from source, Undefined symbol: c10::detail::maybe_wrap_dim_slow(long long, long long, bool)",https://github.com/pytorch/pytorch/issues/95946,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ios, oncall: mobile",scan when time allows,2023-08-08T23:49:19Z
issue,101999,[FSDP] Ensure full precision checkpoints,https://github.com/pytorch/pytorch/issues/101999,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: fsdp, oncall: distributed",scan when time allows,2023-08-08T22:04:52Z
issue,106549,Can't build PyTorch 1.13.1 with Vulkan support,https://github.com/pytorch/pytorch/issues/106549,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: vulkan,scan when time allows,2023-08-08T21:27:13Z
issue,106237,ghstack + mergebot race condition,https://github.com/pytorch/pytorch/issues/106237,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2023-08-08T20:27:22Z
issue,106073,Potential lack of CI testing on older NVIDIA GPU,https://github.com/pytorch/pytorch/issues/106073,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2023-08-08T20:20:33Z
issue,106427,`torch.nn.utils.clip_grad_norm_()` causes H2D sync with foreach ops.,https://github.com/pytorch/pytorch/issues/106427,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: mta, module: nn",scan when time allows,2023-08-08T14:47:48Z
issue,54209,[docs] Incomplete examples for torch.Generator,https://github.com/pytorch/pytorch/issues/54209,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2023-08-08T14:37:26Z
issue,41681,TypeError: cannot create 'generator' instances,https://github.com/pytorch/pytorch/issues/41681,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: data parallel,scan when time allows,2023-08-08T11:09:55Z
issue,106713,Dev Container Support for PyTorch,https://github.com/pytorch/pytorch/issues/106713,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2023-08-07T18:50:36Z
issue,106718,Add AMD image to the .devcontainer spec,https://github.com/pytorch/pytorch/issues/106718,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2023-08-07T18:50:13Z
issue,106717,Provide .devcontainer PyTorch - MPS environment,https://github.com/pytorch/pytorch/issues/106717,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2023-08-07T18:48:05Z
issue,66573,tools/amd_build/build_amd.py should fail if any file fails to write,https://github.com/pytorch/pytorch/issues/66573,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: rocm",scan when time allows,2023-08-07T15:53:30Z
issue,106662,Will torch.sparse.mm support multiplying two boolean matrices?,https://github.com/pytorch/pytorch/issues/106662,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: boolean tensor, module: sparse",scan when time allows,2023-08-06T23:20:05Z
issue,106579,"Boolean valued images loaded from disk, when converted to torch int/float tensor, the True valued pixels gets converted to 255 instead of 1",https://github.com/pytorch/pytorch/issues/106579,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: boolean tensor,scan when time allows,2023-08-05T18:32:43Z
issue,91936,Fails to build on ppc64le with clang,https://github.com/pytorch/pytorch/issues/91936,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: vectorization",scan when time allows,2023-08-05T14:18:26Z
issue,106520,Distributed torch.linalg.eigh (and other functions) on cuda using cuSOLVERMG,https://github.com/pytorch/pytorch/issues/106520,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: linear algebra",scan when time allows,2023-08-05T08:28:44Z
issue,106604,Relu6 not able to process nan values,https://github.com/pytorch/pytorch/issues/106604,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nans and infs,scan when time allows,2023-08-05T05:53:17Z
issue,106606,More Performant CachingHostAllocator for Pinned Memory Allocation,https://github.com/pytorch/pytorch/issues/106606,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2023-08-05T03:53:21Z
issue,106563,NotImplementedError: Could not run 'aten::multinomial' with arguments from the 'Meta' backend.,https://github.com/pytorch/pytorch/issues/106563,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: random,scan when time allows,2023-08-05T03:47:01Z
issue,106540,Dataset with Queue issue,https://github.com/pytorch/pytorch/issues/106540,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: data, module: dataloader",scan when time allows,2023-08-05T03:41:51Z
issue,106533,CUDA device support does not register allocator to c10::GetAllocator(...),https://github.com/pytorch/pytorch/issues/106533,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: cuda",scan when time allows,2023-08-05T03:40:33Z
issue,106545,`softmax` to handle dimensions comprised of `-inf`,https://github.com/pytorch/pytorch/issues/106545,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2023-08-04T20:07:06Z
issue,55457,[feature request] TupleSequential module or an option for Sequential for unpacking tuples (i.e. support varargs),https://github.com/pytorch/pytorch/issues/55457,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2023-08-04T11:00:59Z
issue,33298,[feature request] torch.expand to not require unsqueeze and match -1 to existing dimensions if tensor.shape.count(-1) == tensor.ndim,https://github.com/pytorch/pytorch/issues/33298,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2023-08-04T00:44:51Z
issue,42959,[feature request] Faster specialized int16->float32 conversions to match speed with NumPy,https://github.com/pytorch/pytorch/issues/42959,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: performance",scan when time allows,2023-08-03T23:13:38Z
issue,82316,Add more Vulkan operations,https://github.com/pytorch/pytorch/issues/82316,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: vulkan,scan when time allows,2023-08-03T22:14:29Z
issue,106566,DTensor Sharding prop cache stats,https://github.com/pytorch/pytorch/issues/106566,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dtensor, oncall: distributed",scan when time allows,2023-08-03T19:07:35Z
issue,54216,Add inplace variant for torch.minimum / torch.maximum,https://github.com/pytorch/pytorch/issues/54216,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2023-08-03T18:26:08Z
issue,103352,"[feature request] Native method for iterating Python items of tensors: `iteritems()` and a new `tensor.item(i, j, k, ...)` method",https://github.com/pytorch/pytorch/issues/103352,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python frontend,scan when time allows,2023-08-03T14:55:46Z
issue,106469,Extreme slowdown of torch.mm for certain sizes and strides with bfloat16,https://github.com/pytorch/pytorch/issues/106469,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bfloat16, module: cuda",scan when time allows,2023-08-03T02:36:49Z
issue,5740,[feature request] scalar input to scatter_add_ like scatter_,https://github.com/pytorch/pytorch/issues/5740,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: scatter & gather ops,scan when time allows,2023-08-03T00:31:00Z
issue,84044,"hipErrorNoBinaryForGpu, but reversed",https://github.com/pytorch/pytorch/issues/84044,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rocm,scan when time allows,2023-08-02T20:54:49Z
issue,106382,Installing torchvision for CPU leads to unwanted upgrade of torch + pip would not install nightly as considers that release is the latest (?),https://github.com/pytorch/pytorch/issues/106382,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: binaries,scan when time allows,2023-08-02T17:07:15Z
issue,106450,Check for output_padding <= stride/dilation in ConvTranspose1d,https://github.com/pytorch/pytorch/issues/106450,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: padding",scan when time allows,2023-08-02T13:59:47Z
issue,106173,`torch.ops.aten.split.Tensor._schema` return alias annotations are wrong,https://github.com/pytorch/pytorch/issues/106173,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: viewing and reshaping,scan when time allows,2023-08-02T13:46:20Z
issue,106375,nll_loss reference shouldn't be registered as a decomposition.,https://github.com/pytorch/pytorch/issues/106375,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: decompositions, module: primtorch",scan when time allows,2023-08-02T13:45:58Z
issue,51011,DataLoader is slow in spawned processes,https://github.com/pytorch/pytorch/issues/51011,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dataloader, module: multiprocessing, module: performance",scan when time allows,2023-08-02T07:57:46Z
issue,22541,make[2]: *** No rule to make target 'libtorch/lib/libc10.so',https://github.com/pytorch/pytorch/issues/22541,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cpp",scan when time allows,2023-08-02T06:37:36Z
issue,59545,max-sum operation,https://github.com/pytorch/pytorch/issues/59545,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: linear algebra, module: reductions",scan when time allows,2023-08-01T23:36:09Z
issue,98481,Bring CudaPluggableAllocator to feature parity with the Native Allocator,https://github.com/pytorch/pytorch/issues/98481,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: cudacachingallocator, module: internals",scan when time allows,2023-08-01T16:02:20Z
issue,106339,"ReduceLROnPlateau increases learning rate exponentially, causing training to diverge",https://github.com/pytorch/pytorch/issues/106339,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: lrscheduler,scan when time allows,2023-08-01T15:54:57Z
issue,106359,backwards compatibility about class _LRScheduler,https://github.com/pytorch/pytorch/issues/106359,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: lrscheduler,scan when time allows,2023-08-01T15:54:08Z
issue,106297,Can't build with non-static protobuf,https://github.com/pytorch/pytorch/issues/106297,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2023-08-01T07:12:03Z
issue,90289,quantization qconfig: can we set per-channel quant as default for qnnpack?,https://github.com/pytorch/pytorch/issues/90289,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2023-07-31T22:13:33Z
issue,104623,I propose a new overview section in the documentation,https://github.com/pytorch/pytorch/issues/104623,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2023-07-31T15:20:04Z
issue,98406,Contribute to the privateuse1 backend.,https://github.com/pytorch/pytorch/issues/98406,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: backend, module: internals",scan when time allows,2023-07-31T14:34:34Z
issue,40373,[discussion] Expressing tensor dimension semantics / constraints through typing / constraints blocks. Constraints block could be scripted/traced and help for tracing/script execution and codegen,https://github.com/pytorch/pytorch/issues/40373,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: internals,scan when time allows,2023-07-30T20:33:47Z
issue,105213,[DTensor] Module parallelized with ColwiseParallel should return a sharded tensor,https://github.com/pytorch/pytorch/issues/105213,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dtensor,scan when time allows,2023-07-29T00:20:44Z
issue,106206,Build failure due to C++ version mismatch,https://github.com/pytorch/pytorch/issues/106206,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2023-07-28T21:45:58Z
issue,58833,Foreach Functions Tracking Issue,https://github.com/pytorch/pytorch/issues/58833,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: mta, module: performance",scan when time allows,2023-07-27T21:38:22Z
issue,105904,Libtorch linking Error:undefined reference,https://github.com/pytorch/pytorch/issues/105904,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2023-07-27T14:37:39Z
issue,105726,There is a big precision error between A100 and 3090 when using torch.matmul with fp16 precision,https://github.com/pytorch/pytorch/issues/105726,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: half",scan when time allows,2023-07-27T07:57:27Z
issue,52256,[docs] Bad docs SEO,https://github.com/pytorch/pytorch/issues/52256,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: doc infra, module: docs",scan when time allows,2023-07-27T05:54:08Z
issue,105799,inconsistent signature for dataloader in docs/source/data.rst,https://github.com/pytorch/pytorch/issues/105799,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2023-07-26T23:33:31Z
issue,104479,FSDP Optimizer Overlap - follow ups,https://github.com/pytorch/pytorch/issues/104479,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: fsdp, oncall: distributed",scan when time allows,2023-07-26T12:33:13Z
issue,104872,errors in CONTRIBUTING.md,https://github.com/pytorch/pytorch/issues/104872,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2023-07-26T04:43:38Z
issue,105916,Missing coalesced flag from `torch.autograd.Function.backward`,https://github.com/pytorch/pytorch/issues/105916,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2023-07-25T20:20:36Z
issue,99176,Add Debug builds for python with pydebug,https://github.com/pytorch/pytorch/issues/99176,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: devx,scan when time allows,2023-07-25T20:20:16Z
issue,35641,"Dimension reducing variants of bitwise operations (bitwise_or, bitwise_and, bitwise_xor)",https://github.com/pytorch/pytorch/issues/35641,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: reductions,scan when time allows,2023-07-25T20:18:51Z
issue,105954,Ensure PRs are rebased on top of a recent commit (CI check),https://github.com/pytorch/pytorch/issues/105954,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: devx,scan when time allows,2023-07-25T20:16:03Z
issue,105846,Strange backward behavior with sparse tensors,https://github.com/pytorch/pytorch/issues/105846,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: sparse",scan when time allows,2023-07-25T11:01:34Z
issue,105860,Programmation error enabling unlegal memory access on gpu,https://github.com/pytorch/pytorch/issues/105860,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: error checking, module: mps",scan when time allows,2023-07-24T21:30:27Z
issue,98947,Add `torch.cat` support for torch native sparse tensors. (Need for PyG),https://github.com/pytorch/pytorch/issues/98947,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2023-07-24T20:13:02Z
issue,105682,`torch.autocast(bfloat16)` runs bwd matmuls in fp16,https://github.com/pytorch/pytorch/issues/105682,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision),scan when time allows,2023-07-24T16:56:01Z
issue,104389,distributed hooks want to support custom device,https://github.com/pytorch/pytorch/issues/104389,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2023-07-22T04:11:52Z
issue,79299,PyTorch gets stuck when using an NVLink/A6000 and more than two GPUs,https://github.com/pytorch/pytorch/issues/79299,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ddp, oncall: distributed",scan when time allows,2023-07-22T00:48:21Z
issue,105697,Support symmetry in einsum,https://github.com/pytorch/pytorch/issues/105697,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: linear algebra,scan when time allows,2023-07-21T08:53:43Z
issue,101370,SparseAdam: working with dense parameters but sparse gradients - usecase,https://github.com/pytorch/pytorch/issues/101370,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2023-07-20T21:53:38Z
issue,105629,F.pad will accept 0 and negative values as parameter,https://github.com/pytorch/pytorch/issues/105629,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: error checking, module: nn, module: padding",scan when time allows,2023-07-20T15:14:42Z
issue,105548,torch.sparse.sampled_addmm doesn't compute gradients for 3D tensors,https://github.com/pytorch/pytorch/issues/105548,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2023-07-20T14:49:37Z
issue,61222,The syncbatchnorm can not be employed in the MODEL that the gradients including in the loss function,https://github.com/pytorch/pytorch/issues/61222,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: derivatives, module: norms and normalization, oncall: distributed",scan when time allows,2023-07-20T14:30:34Z
issue,105457,Top level Glossary for users (not contributers),https://github.com/pytorch/pytorch/issues/105457,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2023-07-20T13:16:30Z
issue,105529,Efficient BMM for sparse-dense tensors,https://github.com/pytorch/pytorch/issues/105529,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2023-07-20T12:17:31Z
issue,51199,[numpy] torch.ceil and torch.floor don't support integer inputs while Numpy does,https://github.com/pytorch/pytorch/issues/51199,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: numpy,scan when time allows,2023-07-20T07:27:19Z
issue,90760,Large slow down by not calling `torch.set_num_threads`,https://github.com/pytorch/pytorch/issues/90760,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: linear algebra, module: performance",scan when time allows,2023-07-20T03:49:57Z
issue,103829,RPC Framework support custom backend,https://github.com/pytorch/pytorch/issues/103829,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2023-07-20T02:54:16Z
issue,103313,Fast kernels for low rank matrix multiplication,https://github.com/pytorch/pytorch/issues/103313,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: linear algebra,scan when time allows,2023-07-19T18:14:15Z
issue,105329,Softmax doesn't support sparse tensors with the CSR layout,https://github.com/pytorch/pytorch/issues/105329,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2023-07-19T11:12:32Z
issue,91989,Torch's affinity setting lead to openvino using only one core.,https://github.com/pytorch/pytorch/issues/91989,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: intel,scan when time allows,2023-07-19T03:42:58Z
issue,105445,Silent Error of torch.fx.symbolic_trace when forward hooks are registered,https://github.com/pytorch/pytorch/issues/105445,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2023-07-18T21:27:52Z
issue,105499,[FSDP] Revisit mixed-precision casting logic,https://github.com/pytorch/pytorch/issues/105499,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fsdp,scan when time allows,2023-07-18T20:52:18Z
issue,105220,Significant time difference of calculating Jacobian matrix using jacrev and oracle functions,https://github.com/pytorch/pytorch/issues/105220,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: functorch",scan when time allows,2023-07-18T19:54:19Z
issue,105304,"Backward pass with sparse parameters results in error ""Sparse division requires a scalar or zero-dim dense tensor divisor""",https://github.com/pytorch/pytorch/issues/105304,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: loss, module: optimizer, module: sparse",scan when time allows,2023-07-18T15:38:14Z
issue,104510,"Using the latest version of Torch, when the code executes tcpstore, there is no response",https://github.com/pytorch/pytorch/issues/104510,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2023-07-18T12:52:47Z
issue,104867,rfftn and irfftn operations in pt2 return different results compared to v1.12.1,https://github.com/pytorch/pytorch/issues/104867,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: fft, module: third_party",scan when time allows,2023-07-18T09:02:07Z
issue,89490,build: cmake: ability to disable -Werror* (-Werror considered harmful),https://github.com/pytorch/pytorch/issues/89490,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2023-07-18T04:54:32Z
issue,25104,[FR] Dropout modules/functions should take in generator=,https://github.com/pytorch/pytorch/issues/25104,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: random",scan when time allows,2023-07-17T22:01:00Z
issue,32305,[FR] histc should (optionally) return a long tensor,https://github.com/pytorch/pytorch/issues/32305,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sorting and selection,scan when time allows,2023-07-17T21:26:37Z
issue,105062,[feature request] make the input k in rot90 a list of int to rotate tensors individually in a batch,https://github.com/pytorch/pytorch/issues/105062,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python frontend,scan when time allows,2023-07-17T21:06:31Z
issue,105326,Can't vmap over torch.tensor constructor,https://github.com/pytorch/pytorch/issues/105326,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: functorch,scan when time allows,2023-07-17T15:47:54Z
issue,103422,Possible memory leak when using Torch and Torchvision in conjunction with XGBoost,https://github.com/pytorch/pytorch/issues/103422,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: memory usage, module: vision",scan when time allows,2023-07-17T14:53:48Z
issue,102085,BatchNorm can't be symbolically traced with torch.fx as a standalone module,https://github.com/pytorch/pytorch/issues/102085,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: functorch,scan when time allows,2023-07-17T08:17:53Z
issue,31460,DataParallel doesn't properly handle kwargs,https://github.com/pytorch/pytorch/issues/31460,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: data parallel, module: nn",scan when time allows,2023-07-17T01:51:08Z
issue,30711,Sorting in embedding_dense_backward_cuda takes very long time,https://github.com/pytorch/pytorch/issues/30711,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: embedding, module: performance",scan when time allows,2023-07-15T21:19:21Z
issue,104845,Implement `diag` method for sparse COO tensors,https://github.com/pytorch/pytorch/issues/104845,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2023-07-15T10:43:42Z
issue,104711,New Loss Function Add In Pytorch,https://github.com/pytorch/pytorch/issues/104711,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: loss,scan when time allows,2023-07-15T00:28:03Z
issue,104925,Unnecessary record_stream call for backend:cudaMallocAsync,https://github.com/pytorch/pytorch/issues/104925,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: logging",scan when time allows,2023-07-14T21:25:00Z
issue,49016,Discrete Cosine Transform,https://github.com/pytorch/pytorch/issues/49016,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2023-07-14T16:50:17Z
issue,104962,Add a diagram showing the code structure to CONTRIBUTING.md,https://github.com/pytorch/pytorch/issues/104962,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2023-07-14T15:24:55Z
issue,105060,extra information messages for mac in setup.py would help.,https://github.com/pytorch/pytorch/issues/105060,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: docs",scan when time allows,2023-07-14T03:16:36Z
issue,104466,`torch.view_as_real(tensor)` should return `nn.identity(tensor)` if its not complex instead of raising an error,https://github.com/pytorch/pytorch/issues/104466,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: complex,scan when time allows,2023-07-11T17:59:21Z
issue,102740,parameterizations.orthogonal does not work as intended with nn.GRU or nn.LSTM,https://github.com/pytorch/pytorch/issues/102740,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn.utils.parametrize, module: rnn",scan when time allows,2023-07-11T12:57:11Z
issue,104823,Unrelated error messages with torch.nn.AdaptiveAvgPool3d,https://github.com/pytorch/pytorch/issues/104823,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: pooling",scan when time allows,2023-07-11T08:16:51Z
issue,104935,torch version compare,https://github.com/pytorch/pytorch/issues/104935,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python frontend,scan when time allows,2023-07-11T04:17:38Z
issue,104789,[BE] Evaluate and improve eager for-loop optimizer memory perf,https://github.com/pytorch/pytorch/issues/104789,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2023-07-11T01:18:08Z
issue,104602,"F.adaptive_avg_pool3d(input, 1) returns infinity in half precision",https://github.com/pytorch/pytorch/issues/104602,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: nn, module: numerical-stability",scan when time allows,2023-07-10T23:35:43Z
issue,104154,Numbers bigger than the range should be inf while the implementation just keeps the original.,https://github.com/pytorch/pytorch/issues/104154,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: nans and infs",scan when time allows,2023-07-10T23:35:33Z
issue,104875,torch/testing/_comparison.py: If you are a user and see this message during normal operation please file an issue,https://github.com/pytorch/pytorch/issues/104875,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: testing,scan when time allows,2023-07-10T23:34:53Z
issue,104853,torch.norm inconsistency?,https://github.com/pytorch/pytorch/issues/104853,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: norms and normalization, module: numerical-stability",scan when time allows,2023-07-10T23:26:33Z
issue,52262,tolist called on torch scalar does not return a list => proposal to support new arg force = True to provide a work around,https://github.com/pytorch/pytorch/issues/52262,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: numpy",scan when time allows,2023-07-10T21:37:02Z
issue,1362,Feature Request: noise contrastive estimation/negative sampling,https://github.com/pytorch/pytorch/issues/1362,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2023-07-10T17:01:21Z
issue,103879,Can ``torch.vmap`` add ``grad_fn``= SelectBackward when maping over some dimension of the inputs?,https://github.com/pytorch/pytorch/issues/103879,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: functorch,scan when time allows,2023-07-10T13:40:39Z
issue,87449,pytorch could not build from source with cudnn 8.0.5,https://github.com/pytorch/pytorch/issues/87449,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cuda, module: cudnn",scan when time allows,2023-07-10T02:15:41Z
issue,96085,Add support for `__collate__` attrib on dataset elements in `default_collate`,https://github.com/pytorch/pytorch/issues/96085,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2023-07-09T10:32:37Z
issue,104808,PyTorch built with CuDNN-8.8.1 crashes if CuDNN-8.9.2 is installed on the system,https://github.com/pytorch/pytorch/issues/104808,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ci, module: cudnn",scan when time allows,2023-07-07T22:24:16Z
issue,104788,Use `isinstance` instead of `type` when checking for `torch.nn.Parameter`,https://github.com/pytorch/pytorch/issues/104788,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2023-07-07T20:32:09Z
issue,90394,FX graph mode quant: backendconfig configuration missing for torch.nn.GRU,https://github.com/pytorch/pytorch/issues/90394,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2023-07-07T16:22:14Z
issue,104768,Support for `eval` in functional_call,https://github.com/pytorch/pytorch/issues/104768,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: functional ux, module: nn",scan when time allows,2023-07-07T16:07:50Z
issue,100051,Issue of HistogramObserver to handle abnormal value,https://github.com/pytorch/pytorch/issues/100051,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2023-07-07T16:02:34Z
issue,94586,`permute` for named tensors,https://github.com/pytorch/pytorch/issues/94586,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: named tensor,scan when time allows,2023-07-07T12:30:07Z
issue,104704,generate_vmap_rule=True sometimes gives batched grad_output,https://github.com/pytorch/pytorch/issues/104704,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: functorch,scan when time allows,2023-07-06T22:30:35Z
issue,104620,`torch.distributed.rpc.backend_registry.register_backend` fails to update `BackendType` enum,https://github.com/pytorch/pytorch/issues/104620,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: rpc, oncall: distributed",scan when time allows,2023-07-06T20:25:53Z
issue,104698,Incorrect Error Message Ordering for nn.AdaptiveAvgPool2d with Incorrect output_size,https://github.com/pytorch/pytorch/issues/104698,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2023-07-06T18:06:52Z
issue,103258,"Warn / deprecate / remove ProcessGroupNCCL._group_start(), _group_end() APIs",https://github.com/pytorch/pytorch/issues/103258,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2023-07-05T20:42:26Z
issue,104593,Bug in Conv/BN fuser with torch.fx,https://github.com/pytorch/pytorch/issues/104593,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: fx, oncall: fx",scan when time allows,2023-07-05T17:27:15Z
issue,94804,Build Error: no matching function for call to ‘dnnl::graph::stream::stream(<brace-enclosed initializer list>)’,https://github.com/pytorch/pytorch/issues/94804,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: mkldnn",scan when time allows,2023-07-05T13:02:48Z
issue,94869,aten::cudnn_convolution chooses different conv implementation given the same inputs.,https://github.com/pytorch/pytorch/issues/94869,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: cuda, module: cudnn",scan when time allows,2023-07-04T20:30:34Z
issue,103397,LayerNorm freeze processes using torch multiprocessing,https://github.com/pytorch/pytorch/issues/103397,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: multiprocessing,scan when time allows,2023-07-04T20:08:52Z
issue,104315,Torch randperm with device mps does not sample exactly uniformly from all possible permutations,https://github.com/pytorch/pytorch/issues/104315,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: mps, module: random",scan when time allows,2023-07-04T15:41:04Z
issue,72737,Pattern Matching with Tensors,https://github.com/pytorch/pytorch/issues/72737,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2023-07-04T05:34:04Z
issue,102269,Import of torch breaks standard multiprocessing,https://github.com/pytorch/pytorch/issues/102269,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dependency bug, module: multiprocessing, module: openmp",scan when time allows,2023-07-03T08:18:25Z
issue,102772,Support In-place Triangular Matrix Multiplication,https://github.com/pytorch/pytorch/issues/102772,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: linear algebra,scan when time allows,2023-07-02T20:16:58Z
issue,79337,Conda install from pytorch-nightly channel does not install the expected version on macOS,https://github.com/pytorch/pytorch/issues/79337,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: macos",scan when time allows,2023-07-01T09:37:40Z
issue,104472,Investigate numerical stability of forward-mode AD of some foreach functions,https://github.com/pytorch/pytorch/issues/104472,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: forward ad,scan when time allows,2023-06-30T19:56:14Z
issue,86694,Feature request: Deterministic test input generation,https://github.com/pytorch/pytorch/issues/86694,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tests,scan when time allows,2023-06-30T17:14:22Z
issue,104107,Tensor to_sparse fails on large matrices,https://github.com/pytorch/pytorch/issues/104107,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: sparse",scan when time allows,2023-06-30T14:15:57Z
issue,104322,Illegal Memory Access on H100 `TestSparseCompressedTritonKernelsCUDA.test_triton_sampled_addmm_block_size_16_cuda_bfloat16`,https://github.com/pytorch/pytorch/issues/104322,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: sparse",scan when time allows,2023-06-29T17:23:17Z
issue,104163,Nested Tensor with PyG dataset custom class,https://github.com/pytorch/pytorch/issues/104163,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nestedtensor,scan when time allows,2023-06-29T08:05:08Z
issue,58466,`set_per_process_memory_fraction()` does not ensure max used GPU memory below fraction,https://github.com/pytorch/pytorch/issues/58466,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2023-06-28T21:15:01Z
issue,104282,Rename `topic: not user facing` to `release notes: not user facing`,https://github.com/pytorch/pytorch/issues/104282,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2023-06-28T16:50:05Z
issue,41839,"[Feature Request] Add `to`, `cpu`, and `cuda` method to optimizer",https://github.com/pytorch/pytorch/issues/41839,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2023-06-28T05:54:13Z
issue,104252,[RFC] Make `_HYBRID_SHARD_ZERO2` public as `HYBRID_SHARD_GRAD_OP`,https://github.com/pytorch/pytorch/issues/104252,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fsdp,scan when time allows,2023-06-27T18:11:14Z
issue,103589,`interpolate` with `antialias=True` on CUDA doesn't work if the difference of spatial size is large,https://github.com/pytorch/pytorch/issues/103589,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: interpolation",scan when time allows,2023-06-27T08:43:54Z
issue,3625,Proposal: combine requires_grad and retain_grad(),https://github.com/pytorch/pytorch/issues/3625,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2023-06-26T22:16:06Z
issue,103960,CODEOWNERS file has errors due to non existent people being referred to,https://github.com/pytorch/pytorch/issues/103960,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2023-06-26T18:24:55Z
issue,98495,Strided to batch BSR/BSC conversion fails when the number of zeros per block varies while the number of blocks per patch is constant,https://github.com/pytorch/pytorch/issues/98495,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2023-06-26T16:28:04Z
issue,103959,"Need the full ""Release Compatibility Matrix"" of torch",https://github.com/pytorch/pytorch/issues/103959,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, oncall: releng",scan when time allows,2023-06-25T01:37:54Z
issue,99797,Automatic broadcasting for sparse csr tensors,https://github.com/pytorch/pytorch/issues/99797,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2023-06-23T21:35:02Z
issue,103820,Upgrading SpGEMM algorithm to resolve Cusparse SpGEMM insufficient resources problem,https://github.com/pytorch/pytorch/issues/103820,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: sparse",scan when time allows,2023-06-23T20:42:27Z
issue,101069,"torch.autograd.detect_anomaly should report the original forward trace as part of the error, rather than as out of band warning",https://github.com/pytorch/pytorch/issues/101069,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2023-06-23T16:16:56Z
issue,97827,Memory leak when saving an input tensor returned as-is if mark_dirty and running with dual tensors,https://github.com/pytorch/pytorch/issues/97827,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: edge cases, module: forward ad",scan when time allows,2023-06-23T16:16:00Z
issue,91996,autograd.functional.jacobian : Imaginary part is lost for functions with real input and complex output.,https://github.com/pytorch/pytorch/issues/91996,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2023-06-23T15:42:30Z
issue,94990,interactions between views + autograd.Function + AOTAutograd causes memory leak,https://github.com/pytorch/pytorch/issues/94990,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2023-06-23T15:29:58Z
issue,94111,`svd` triggers INTERNAL ASSERT FAILED when computing jacobian in forward mode,https://github.com/pytorch/pytorch/issues/94111,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: complex, module: forward ad, module: linear algebra",scan when time allows,2023-06-23T15:28:56Z
issue,96766,Follow-ups to do after adding nested checkpoint,https://github.com/pytorch/pytorch/issues/96766,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2023-06-23T15:21:58Z
issue,95132,Differentiate with regard a subset of the input,https://github.com/pytorch/pytorch/issues/95132,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2023-06-23T15:21:19Z
issue,97575,Multi-output derivative formulas can save unnecessary tensors,https://github.com/pytorch/pytorch/issues/97575,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: nestedtensor",scan when time allows,2023-06-23T15:14:21Z
issue,103802,Eager PTDQ Performs Worse Than Non-Quantized Linear Layer on CPU(in Terms of Speed),https://github.com/pytorch/pytorch/issues/103802,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2023-06-23T09:19:28Z
issue,103803,Support `Sequence` type in JIT,https://github.com/pytorch/pytorch/issues/103803,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: typing, oncall: jit",scan when time allows,2023-06-23T03:25:16Z
issue,60401,Need workaround to support multiprocess CUDA tensor sharing on Jetson Platforms,https://github.com/pytorch/pytorch/issues/60401,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: multiprocessing,scan when time allows,2023-06-22T20:45:04Z
issue,104029,SummaryWriter.add_embedding not working for RGBA images,https://github.com/pytorch/pytorch/issues/104029,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tensorboard,scan when time allows,2023-06-22T20:42:50Z
issue,103895,Issue with loading similar checkpoints in a distributed fashion,https://github.com/pytorch/pytorch/issues/103895,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: serialization,scan when time allows,2023-06-22T20:31:36Z
issue,104033,Reproducibility documentation should be updated,https://github.com/pytorch/pytorch/issues/104033,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: numerical-reproducibility",scan when time allows,2023-06-22T17:23:55Z
issue,46739,torch.utils.tensorboard.SummaryWriter.add_embedding fails for some label_img sizes,https://github.com/pytorch/pytorch/issues/46739,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tensorboard,scan when time allows,2023-06-22T15:17:06Z
issue,104040,Improved error checking for custom Function when saving intermediates,https://github.com/pytorch/pytorch/issues/104040,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: double backwards",scan when time allows,2023-06-22T13:35:41Z
issue,14489,"Batch matmul with sparse matrix, dense vector",https://github.com/pytorch/pytorch/issues/14489,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2023-06-22T09:59:55Z
issue,41128,Memory bug for backward on torch.sparse.mm?,https://github.com/pytorch/pytorch/issues/41128,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: sparse",scan when time allows,2023-06-22T09:51:09Z
issue,101934,Parameter gradient is not moved parameter is moved across devices,https://github.com/pytorch/pytorch/issues/101934,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: docs, module: nn",scan when time allows,2023-06-22T01:52:48Z
issue,103860,"Long PR description leads to ""Argument list too long"" error from docker",https://github.com/pytorch/pytorch/issues/103860,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ci, module: devx",scan when time allows,2023-06-21T23:24:06Z
issue,103370,torchgen/gen_backend_stubs.py compatibility with DispatchStubs,https://github.com/pytorch/pytorch/issues/103370,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: codegen, module: dispatch, module: structured kernels",scan when time allows,2023-06-21T20:16:08Z
issue,103868,"row.device().is_cpu() INTERNAL ASSERT FAILED at ""csrc/cpu/diag_cpu.cpp"":7",https://github.com/pytorch/pytorch/issues/103868,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: macos,scan when time allows,2023-06-21T20:08:02Z
issue,103847,Some parameters are missing type descriptions,https://github.com/pytorch/pytorch/issues/103847,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2023-06-21T16:37:58Z
issue,103846,"The document style is inconsistent with other documents, and the parameter type is not clearly highlight",https://github.com/pytorch/pytorch/issues/103846,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2023-06-21T16:37:28Z
issue,103837,type conflict,https://github.com/pytorch/pytorch/issues/103837,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp,scan when time allows,2023-06-21T16:31:44Z
issue,103832,Torch 1.13 for GPU breaks if libcublas is already present.,https://github.com/pytorch/pytorch/issues/103832,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docker,scan when time allows,2023-06-21T16:07:45Z
issue,103798,Type misalignments in `nn.functional` (also JIT related),https://github.com/pytorch/pytorch/issues/103798,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: typing, oncall: jit",scan when time allows,2023-06-21T16:00:56Z
issue,103787,Generate complete annotations for `torch._C._nn`,https://github.com/pytorch/pytorch/issues/103787,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: typing,scan when time allows,2023-06-21T15:58:39Z
issue,54389,"torch.lerp and torch._foreach_lerp should support uint8 inputs (for vision), int16 (for audio), int32/int64 (for generality) without (up)casting to float for and/or dtype argument",https://github.com/pytorch/pytorch/issues/54389,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: interpolation,scan when time allows,2023-06-21T08:16:18Z
issue,78487,torch.fx: symbolic_trace: ones() received an invalid combination of arguments,https://github.com/pytorch/pytorch/issues/78487,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2023-06-21T06:54:03Z
issue,103643,"""addmm_out_sparse_csr_impl_mkl"" not implemented for 'Byte'",https://github.com/pytorch/pytorch/issues/103643,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2023-06-21T00:59:19Z
issue,103640,Disclose C++ ATen ops type promotion rules under OpOverload in Python,https://github.com/pytorch/pytorch/issues/103640,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2023-06-20T21:45:34Z
issue,55366,bool_tensor.sum(dtype=torch.int32) creates int32-copy of the original int8 tensor,https://github.com/pytorch/pytorch/issues/55366,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: boolean tensor, module: reductions",scan when time allows,2023-06-20T21:29:27Z
issue,103161,Calling jacrev with LSTM and functional_call gives error,https://github.com/pytorch/pytorch/issues/103161,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: functorch,scan when time allows,2023-06-20T20:18:59Z
issue,78210,Performance with MPS on AMD GPUs are worse than CPU,https://github.com/pytorch/pytorch/issues/78210,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: mps, module: performance",scan when time allows,2023-06-20T15:47:34Z
issue,53191,RuntimeError: cuDNN error: CUDNN_STATUS_INTERNAL_ERROR,https://github.com/pytorch/pytorch/issues/53191,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: cudnn",scan when time allows,2023-06-20T06:01:01Z
issue,86076,Figure out the future of Metal backend given the existence of MPS,https://github.com/pytorch/pytorch/issues/86076,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: arm, module: build, module: mps",scan when time allows,2023-06-18T12:26:36Z
issue,103761,Merge type stubs for `torch.nn.functional`,https://github.com/pytorch/pytorch/issues/103761,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: typing,scan when time allows,2023-06-17T16:31:44Z
issue,74034,[torch.onnx] ONNX export failed on adaptive_avg_pool2d because input size not accessible not supported,https://github.com/pytorch/pytorch/issues/74034,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2023-06-17T13:27:37Z
issue,72283,Add Sparsemax function to libtorch,https://github.com/pytorch/pytorch/issues/72283,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2023-06-16T10:24:07Z
issue,103608,Installing Torch on AMD Platform Leads to Huge Docker Image,https://github.com/pytorch/pytorch/issues/103608,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rocm,scan when time allows,2023-06-15T17:34:52Z
issue,103671,"""Y.getIntrusivePtr()->set_storage(X.getIntrusivePtr()->storage()); "" in C++ is not supported",https://github.com/pytorch/pytorch/issues/103671,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp,scan when time allows,2023-06-15T17:30:21Z
issue,103518,`gradcheck` produces false positives with sparse inputs when `masked=False`.,https://github.com/pytorch/pytorch/issues/103518,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: sparse",scan when time allows,2023-06-14T21:40:24Z
issue,103241,Image Processing with Pytorch,https://github.com/pytorch/pytorch/issues/103241,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2023-06-14T16:50:31Z
issue,101369,Theme update,https://github.com/pytorch/pytorch/issues/101369,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: doc infra, module: docs",scan when time allows,2023-06-14T16:34:21Z
issue,103498,ERROR:torch.distributed.elastic.multiprocessing.api:failed (exitcode: 1) local_rank: 0 (pid: 6047) of binary: /home/win10-ubuntu/anaconda3/envs/vicuna-7b/bin/python,https://github.com/pytorch/pytorch/issues/103498,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2023-06-14T03:44:01Z
issue,103372,ImportError: cannot import name 'Store' from 'torch.distributed',https://github.com/pytorch/pytorch/issues/103372,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2023-06-14T03:43:13Z
issue,102953,TypeError: (): incompatible function arguments,https://github.com/pytorch/pytorch/issues/102953,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2023-06-14T03:35:20Z
issue,103382,NotImplementedError Could not run 'c10d::alltoall_' with arguments from the 'Meta' backend.,https://github.com/pytorch/pytorch/issues/103382,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2023-06-13T17:10:35Z
issue,20686,In-source build causes repeating filename annotations (Windows doesn't support out-of-source build),https://github.com/pytorch/pytorch/issues/20686,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: windows",scan when time allows,2023-06-13T13:26:38Z
issue,102423,Add default device function interface for device-aware apis,https://github.com/pytorch/pytorch/issues/102423,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dispatch,scan when time allows,2023-06-13T11:17:45Z
issue,100054,Add compile option -Werror=return-type compile error,https://github.com/pytorch/pytorch/issues/100054,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2023-06-13T08:49:22Z
issue,101110,Tensorboard graph tracing with torch fx API,https://github.com/pytorch/pytorch/issues/101110,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: fx, module: tensorboard",scan when time allows,2023-06-13T07:20:37Z
issue,103393,Typing missing on arithmetic ops on `Tensor`,https://github.com/pytorch/pytorch/issues/103393,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: typing,scan when time allows,2023-06-12T21:02:23Z
issue,103375,Inplace binary ops on tensor subclasses can cause mypy error,https://github.com/pytorch/pytorch/issues/103375,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: typing,scan when time allows,2023-06-12T20:54:52Z
issue,99806,`cat` gradgrad tests failing,https://github.com/pytorch/pytorch/issues/99806,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: edge cases",scan when time allows,2023-06-12T17:48:43Z
issue,30386,Exponentiated gradient descent,https://github.com/pytorch/pytorch/issues/30386,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2023-06-12T13:37:17Z
issue,102437,discuss.pytorch.org signup issue,https://github.com/pytorch/pytorch/issues/102437,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2023-06-11T13:20:30Z
issue,99035,Issue on building from source: Remove -mfpu=neon option on MacOS with Apple silicon,https://github.com/pytorch/pytorch/issues/99035,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2023-06-11T07:43:51Z
issue,103272,oneDNN kernel fails to compile,https://github.com/pytorch/pytorch/issues/103272,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: mkldnn,scan when time allows,2023-06-10T04:05:35Z
issue,103316,binary_cross_entropy (loss) seems to be giving incorrect values for very negative logits,https://github.com/pytorch/pytorch/issues/103316,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2023-06-09T23:48:17Z
issue,103212,PyTorch can not be compiled with MKLDNN if system compiler is clang,https://github.com/pytorch/pytorch/issues/103212,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: mkldnn",scan when time allows,2023-06-09T23:36:26Z
issue,102977,raise `RuntimeError` faster when loading an object with a torch CUDA tensor on a CPU-only machine,https://github.com/pytorch/pytorch/issues/102977,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: serialization",scan when time allows,2023-06-09T19:22:14Z
issue,102971,Discussion and Design for Masked Loss Functions which can be used with PackedSequence training (but not exclusively),https://github.com/pytorch/pytorch/issues/102971,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: loss, module: masked operators",scan when time allows,2023-06-09T16:48:09Z
issue,103060,Symbolic trace error about torch.nn.functional.pad,https://github.com/pytorch/pytorch/issues/103060,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2023-06-09T16:01:07Z
issue,103306,DTensor uneven sharding corner cases.,https://github.com/pytorch/pytorch/issues/103306,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2023-06-09T11:26:35Z
issue,102162,NotImplementedError in backprop on on dense-sparse matrices,https://github.com/pytorch/pytorch/issues/102162,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2023-06-09T11:07:05Z
issue,103160,Allow overriding __repr__ to call dataclass_repr (infinite recursion right now),https://github.com/pytorch/pytorch/issues/103160,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: codegen,scan when time allows,2023-06-08T15:14:48Z
issue,103131,"error: ‘aligned_alloc’ was not declared in this scope static_cast<char*>(aligned_alloc(FLATBUFFERS_MAX_ALIGNMENT, size)), free);",https://github.com/pytorch/pytorch/issues/103131,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2023-06-08T15:01:04Z
issue,72948,Feature: a consistent Python and C++ logging facility that handles different classes of warnings,https://github.com/pytorch/pytorch/issues/72948,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: logging,scan when time allows,2023-06-08T08:12:58Z
issue,35529,Support bool input tensors for argmax / argmin / sort / topk and other functions,https://github.com/pytorch/pytorch/issues/35529,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: reductions,scan when time allows,2023-06-07T00:01:58Z
issue,16797,"Should be a way to unpickle an object with a torch cuda tensor on a CPU-only machine when using plain ""pickle""",https://github.com/pytorch/pytorch/issues/16797,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: serialization,scan when time allows,2023-06-06T14:33:47Z
issue,71117,Rollup: forward-mode AD operator coverage,https://github.com/pytorch/pytorch/issues/71117,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: forward ad",scan when time allows,2023-06-06T00:26:29Z
issue,102741,Unlocking PyTorch's Power: README.md in Multiple Languages!,https://github.com/pytorch/pytorch/issues/102741,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2023-06-05T21:54:15Z
issue,7313,[proposal] [discussion] Refactor pruning/weight_norm using new Reparametrization functionality + actually deprecate old impl of SpectralNorm,https://github.com/pytorch/pytorch/issues/7313,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2023-06-05T19:55:20Z
issue,102921,Unbox expectedml,https://github.com/pytorch/pytorch/issues/102921,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2023-06-05T16:20:20Z
issue,91395,PyObject preservation and resurrection for `StorageImpl`,https://github.com/pytorch/pytorch/issues/91395,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: internals,scan when time allows,2023-06-05T13:40:31Z
issue,102830,Unknow error when using `make_graphed_callables`,https://github.com/pytorch/pytorch/issues/102830,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda graphs,scan when time allows,2023-06-05T13:16:44Z
issue,74940,Feature Request: Add J0 J1 J2 H0 H1 H2 Bessel functions,https://github.com/pytorch/pytorch/issues/74940,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: special,scan when time allows,2023-06-03T08:01:17Z
issue,101291,fused torch.optim.AdamW isn't faster than the unfused version,https://github.com/pytorch/pytorch/issues/101291,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2023-06-03T03:02:47Z
issue,102811,`torch.poisson(torch.tensor([torch.inf))` returns 0,https://github.com/pytorch/pytorch/issues/102811,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: distributions, module: edge cases, module: error checking",scan when time allows,2023-06-02T22:17:08Z
issue,102753,Error: no matching constructor for initialization of 'at::OptionalIntArrayRef',https://github.com/pytorch/pytorch/issues/102753,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: macos",scan when time allows,2023-06-02T17:32:10Z
issue,102761,Pytorch Build images for RISCV64 Devices in the nightly builds,https://github.com/pytorch/pytorch/issues/102761,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2023-06-02T16:10:40Z
issue,102812,[DTensor] Error in distribute_module with module._apply,https://github.com/pytorch/pytorch/issues/102812,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2023-06-02T13:02:21Z
issue,97757,oneDNN 3.0+ support,https://github.com/pytorch/pytorch/issues/97757,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: intel,scan when time allows,2023-06-02T07:48:47Z
issue,102732,Building NCCL with `make -l $MAX_JOBS` slows down builds,https://github.com/pytorch/pytorch/issues/102732,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2023-06-01T17:20:05Z
issue,67586,SequentialLR have a question and why it use `step(epoch)`,https://github.com/pytorch/pytorch/issues/67586,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: lrscheduler, module: optimizer",scan when time allows,2023-06-01T16:54:59Z
issue,102609,[prims] torch.ops.aten.le decomposition confuses scalars and tensors,https://github.com/pytorch/pytorch/issues/102609,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: primtorch,scan when time allows,2023-06-01T16:09:49Z
issue,53758,ZeroDivisionError: float division by zero in Adam (bias_correction1 is zero),https://github.com/pytorch/pytorch/issues/53758,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: numerical-stability, module: optimizer",scan when time allows,2023-06-01T14:19:03Z
issue,99432,[DTensor] parallelize_module failed with nn.Transformer and the PairwiseParallel plan,https://github.com/pytorch/pytorch/issues/99432,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2023-05-31T21:58:47Z
issue,102610,Cannot invoke prims.sum with output_dtype,https://github.com/pytorch/pytorch/issues/102610,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: primtorch,scan when time allows,2023-05-31T13:23:50Z
issue,102515,pytorch java api documentation is not clear and does not cover example,https://github.com/pytorch/pytorch/issues/102515,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, oncall: java, oncall: mobile",scan when time allows,2023-05-31T04:05:22Z
issue,102556,test_functional_autograd_benchmark.py::TestFunctionalAutogradBenchmark::test_fast_tasks passes with all NaNs,https://github.com/pytorch/pytorch/issues/102556,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: functorch",scan when time allows,2023-05-31T00:04:34Z
issue,102501,[Utils][tensorboard]Enhancement: Include 'max_outputs' parameter in torch.utils.tensorboard.summary's 'image' method,https://github.com/pytorch/pytorch/issues/102501,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tensorboard,scan when time allows,2023-05-31T00:01:58Z
issue,102533,Segfault when running vulkan program linked against libtorch,https://github.com/pytorch/pytorch/issues/102533,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: vulkan,scan when time allows,2023-05-30T23:50:31Z
issue,102511,Faster BatchSampler with big batch size,https://github.com/pytorch/pytorch/issues/102511,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2023-05-30T23:48:47Z
issue,102459,Matrix multiplication performance regression in case of an additional dimension of size 1,https://github.com/pytorch/pytorch/issues/102459,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cublas, module: dependency bug",scan when time allows,2023-05-30T23:43:52Z
issue,102435,torch.distributed.all_reduce() has inconsistent behavior,https://github.com/pytorch/pytorch/issues/102435,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2023-05-30T21:16:41Z
issue,96111,Static size boolean masking,https://github.com/pytorch/pytorch/issues/96111,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: advanced indexing,scan when time allows,2023-05-30T19:41:57Z
issue,102272,"after add /path_to_libtorch/libtorch/lib to LD_LIBRARY_PATH, I can't import torch_scatter.",https://github.com/pytorch/pytorch/issues/102272,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: abi, module: build, module: cpp",scan when time allows,2023-05-27T00:41:42Z
issue,102368,Issue with ShufflerIterDataPipe in torch 1.13.1,https://github.com/pytorch/pytorch/issues/102368,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: data, module: dataloader",scan when time allows,2023-05-26T19:54:39Z
issue,102081,CPU Fallback does not convert Tensor?[],https://github.com/pytorch/pytorch/issues/102081,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: internals,scan when time allows,2023-05-25T22:56:18Z
issue,101653,Unexpected behavior comparing uint8 tensor to value greater than 255,https://github.com/pytorch/pytorch/issues/101653,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: edge cases, module: type promotion",scan when time allows,2023-05-25T14:26:21Z
issue,101890,Can't reproduce/non-deterministic results with CUDA,https://github.com/pytorch/pytorch/issues/101890,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: determinism",scan when time allows,2023-05-24T20:15:18Z
issue,95121,cuda 12 support request.,https://github.com/pytorch/pytorch/issues/95121,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2023-05-24T16:29:41Z
issue,14701,CI: Flaky download from download.pytorch.org,https://github.com/pytorch/pytorch/issues/14701,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ci, module: flaky-tests",scan when time allows,2023-05-23T19:36:59Z
issue,55607,pytorch inference lead to memory leak in cpu,https://github.com/pytorch/pytorch/issues/55607,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: memory usage,scan when time allows,2023-05-23T13:47:37Z
issue,100062,add github check that diffs generated code,https://github.com/pytorch/pytorch/issues/100062,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: codegen, module: infra",scan when time allows,2023-05-23T12:17:00Z
issue,101968,CrossEntropyLoss output difference on Windows vs. Linux,https://github.com/pytorch/pytorch/issues/101968,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: windows,scan when time allows,2023-05-23T08:18:39Z
issue,100785,compile torch2.0 in debug mode,https://github.com/pytorch/pytorch/issues/100785,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2023-05-23T00:43:15Z
issue,99311,Functorch pytrees with custom iterables,https://github.com/pytorch/pytorch/issues/99311,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: functorch, module: pytree",scan when time allows,2023-05-22T17:50:36Z
issue,36217,Rectify docs for MultiLabelSoftMarginLoss,https://github.com/pytorch/pytorch/issues/36217,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: nn",scan when time allows,2023-05-21T23:47:53Z
issue,38271,"PyTorch's flip returns a new tensor, but NumPy's flip returns a view",https://github.com/pytorch/pytorch/issues/38271,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: numpy,scan when time allows,2023-05-20T10:44:28Z
issue,101871,Can group convolution support other grouping methods?,https://github.com/pytorch/pytorch/issues/101871,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: convolution,scan when time allows,2023-05-19T20:29:23Z
issue,101428,2D inputs to linear layers run up to 25% slower than 4D ones on some Nvidia GPUs,https://github.com/pytorch/pytorch/issues/101428,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cublas,scan when time allows,2023-05-19T08:50:21Z
issue,10615,Request for better memory management,https://github.com/pytorch/pytorch/issues/10615,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: memory usage,scan when time allows,2023-05-18T19:31:56Z
issue,101781,Add support for bfloat16 in torch.from_numpy(),https://github.com/pytorch/pytorch/issues/101781,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bfloat16, module: numpy",scan when time allows,2023-05-18T17:24:02Z
issue,101536,"torch.cuda.set_device cannot use to set cpu device, but give an ambiguity hint",https://github.com/pytorch/pytorch/issues/101536,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2023-05-18T07:08:24Z
issue,101716,"[TorchScript] aten::__and__ with argument types: Tensor, bool not supported",https://github.com/pytorch/pytorch/issues/101716,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2023-05-17T23:26:21Z
issue,70166,[feature request] Autocast module and function wrappers,https://github.com/pytorch/pytorch/issues/70166,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision),scan when time allows,2023-05-17T20:35:13Z
issue,101669,[refs] inplace references resize the input to match the broadcasted input shape,https://github.com/pytorch/pytorch/issues/101669,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: primtorch,scan when time allows,2023-05-17T17:59:15Z
issue,101586,"/Users/davidlaxer/pytorch/third_party/tensorpipe/third_party/libuv/src/unix/getaddrinfo.c:165:10: error: implicit declaration of function 'uv__idna_toascii' [-Werror,-Wimplicit-function-declaration] rc = uv__idna_toascii(hostname,",https://github.com/pytorch/pytorch/issues/101586,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2023-05-17T17:47:54Z
issue,101564,not yet implemented the batching rule for torchaudio::_lfilter,https://github.com/pytorch/pytorch/issues/101564,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: batching, module: functorch",scan when time allows,2023-05-17T17:45:28Z
issue,101506,Support pipeline parallelism with PyG,https://github.com/pytorch/pytorch/issues/101506,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2023-05-17T17:40:18Z
issue,101470,[bazel] add inductor to bazel build,https://github.com/pytorch/pytorch/issues/101470,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bazel, module: build",scan when time allows,2023-05-17T17:39:08Z
issue,101319,Deprecated File bug,https://github.com/pytorch/pytorch/issues/101319,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2023-05-17T17:36:45Z
issue,101415,import functorch.dim monkeypatches torch,https://github.com/pytorch/pytorch/issues/101415,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: functorch,scan when time allows,2023-05-17T17:36:25Z
issue,96743,Pruning under channels_last format,https://github.com/pytorch/pytorch/issues/96743,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: memory format, module: pruning",scan when time allows,2023-05-17T14:09:51Z
issue,101505,Investigate random sequence number broadcast initially incorrect,https://github.com/pytorch/pytorch/issues/101505,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2023-05-16T16:59:06Z
issue,87159,"torch/csrc/utils/python_arg_parser.h:424:94: error: format ‘%ld’ expects argument of type ‘long int’, but argument 7 has type ‘int’",https://github.com/pytorch/pytorch/issues/87159,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: arm, module: build",scan when time allows,2023-05-16T16:55:28Z
issue,79967,[feature request] LazyTensor that provides/loads/computes its contents only upon request to be returned from torch.load,https://github.com/pytorch/pytorch/issues/79967,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: lazy,scan when time allows,2023-05-15T23:35:58Z
issue,93206,Build from Source Issues on MacOS Ventura 13.2,https://github.com/pytorch/pytorch/issues/93206,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: macos",scan when time allows,2023-05-15T18:02:34Z
issue,101251,Mac m2 MPSNDArray.mm:78: failed assertion `[MPSNDArrayDescriptor sliceDimension:withSubrange:] error: dimension index (2) not within number of dimensions (2) Dimension indices are 0-based',https://github.com/pytorch/pytorch/issues/101251,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: mps,scan when time allows,2023-05-15T16:21:01Z
issue,101359,RuntimeError in Scaled Dot Product Attention Tutorial Code,https://github.com/pytorch/pytorch/issues/101359,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpu,scan when time allows,2023-05-15T13:48:59Z
issue,101334,Functions for Calculating Skewness and Kurtosis,https://github.com/pytorch/pytorch/issues/101334,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: reductions,scan when time allows,2023-05-15T10:25:05Z
issue,94816,grid_sample with relative grid,https://github.com/pytorch/pytorch/issues/94816,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: interpolation, module: nn",scan when time allows,2023-05-15T07:04:10Z
issue,51429,"dist.init_process_group('nccl', ...) hangs in some combinations of pytorch+python+cuda version",https://github.com/pytorch/pytorch/issues/51429,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2023-05-14T23:20:20Z
issue,101177,version 4.26.1 to 4.29.0 has two bugs,https://github.com/pytorch/pytorch/issues/101177,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2023-05-14T03:58:48Z
issue,101210,GLCM implementation in pytorch C++ api and cuda,https://github.com/pytorch/pytorch/issues/101210,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp,scan when time allows,2023-05-12T19:53:58Z
issue,80296,Sort tensors inplace,https://github.com/pytorch/pytorch/issues/80296,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sorting and selection,scan when time allows,2023-05-12T18:26:17Z
issue,101246,Tool for identifying where in eager model an operation is nondeterministic,https://github.com/pytorch/pytorch/issues/101246,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: determinism,scan when time allows,2023-05-12T14:21:45Z
issue,81749,BatchNorm for complex tensor,https://github.com/pytorch/pytorch/issues/81749,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: complex, module: primtorch",scan when time allows,2023-05-12T13:47:37Z
issue,51623,[docs] FractionalMaxPool2d : _random_samples arg is present in signature but undocumented.,https://github.com/pytorch/pytorch/issues/51623,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: nn, module: pooling",scan when time allows,2023-05-12T08:49:18Z
issue,100957,PyTorch installs the file mkldnn.cmake that looks for the package MKLDNN that doesn't exist,https://github.com/pytorch/pytorch/issues/100957,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: mkldnn",scan when time allows,2023-05-12T08:10:55Z
issue,101189,`dense -> sparse compressed` to work with empty batches.,https://github.com/pytorch/pytorch/issues/101189,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2023-05-12T01:05:34Z
issue,101159,Inconsistency between GPU memory usage in torch.cuda.memory_summary and nvidia-smi,https://github.com/pytorch/pytorch/issues/101159,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2023-05-12T01:03:24Z
issue,101080,Sequence annotation in type hints is wrong,https://github.com/pytorch/pytorch/issues/101080,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: typing,scan when time allows,2023-05-11T23:09:31Z
issue,100123,Error building Pytorch from source,https://github.com/pytorch/pytorch/issues/100123,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: rocm",scan when time allows,2023-05-11T20:05:29Z
issue,100617,GPU VRAM usage significantly higher for Lenet5 models when compared to other frameworks,https://github.com/pytorch/pytorch/issues/100617,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2023-05-11T14:18:28Z
issue,37092,nn.Bilinear cannot be used inside nn.Sequential,https://github.com/pytorch/pytorch/issues/37092,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2023-05-11T07:48:29Z
issue,101070,Lazily format C++ stack trace if it is not used,https://github.com/pytorch/pytorch/issues/101070,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: performance,scan when time allows,2023-05-11T04:00:21Z
issue,101106,"addmv doesn't do type promotion correctly,",https://github.com/pytorch/pytorch/issues/101106,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: linear algebra, module: type promotion",scan when time allows,2023-05-11T03:37:34Z
issue,101073,Windows fatal exception: stack overflow while using pytorch for computing,https://github.com/pytorch/pytorch/issues/101073,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: functorch,scan when time allows,2023-05-11T03:23:07Z
issue,100850,[BUG] Poor torch.bmm performance on H100,https://github.com/pytorch/pytorch/issues/100850,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cublas, module: cuda, module: performance",scan when time allows,2023-05-10T08:04:02Z
issue,100932,torch.concat fails with float16 input in autocast(device_type=cpu) context,https://github.com/pytorch/pytorch/issues/100932,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision), module: bfloat16, module: cpu",scan when time allows,2023-05-10T06:27:04Z
issue,100705,torch.cuda.amp.GradScaler initialization,https://github.com/pytorch/pytorch/issues/100705,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision),scan when time allows,2023-05-10T05:39:57Z
issue,67680,OOM with a lot of GPU memory left,https://github.com/pytorch/pytorch/issues/67680,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: memory usage",scan when time allows,2023-05-10T02:19:29Z
issue,96637,Not allow force merge when lint fails and not because of broken trunk,https://github.com/pytorch/pytorch/issues/96637,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2023-05-09T19:01:13Z
issue,95645,Create a new Docker image with all inductor benchmarks and pre-trained models downloaded,https://github.com/pytorch/pytorch/issues/95645,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ci, module: devx",scan when time allows,2023-05-09T18:55:20Z
issue,98948,behaviour of `torch.tensor()` changes after editing `Tensor.__getitem__`,https://github.com/pytorch/pytorch/issues/98948,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python frontend,scan when time allows,2023-05-09T18:32:36Z
issue,99544,Cross compile Pytorch for ARM in Bazel,https://github.com/pytorch/pytorch/issues/99544,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: arm, module: build",scan when time allows,2023-05-09T17:53:55Z
issue,30574,`index_select` with multidimensional `index`,https://github.com/pytorch/pytorch/issues/30574,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: advanced indexing,scan when time allows,2023-05-09T15:07:38Z
issue,90261,Why torch.mode return different value between CPU and GPU,https://github.com/pytorch/pytorch/issues/90261,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2023-05-09T09:16:14Z
issue,100801,There is a performance drop because we have not yet implemented the batching rule for aten::native_dropout_backward,https://github.com/pytorch/pytorch/issues/100801,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: functorch,scan when time allows,2023-05-08T22:18:12Z
issue,100795,Quickstart notebook fails to train properly with ROCm,https://github.com/pytorch/pytorch/issues/100795,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rocm,scan when time allows,2023-05-08T22:16:50Z
issue,100904,Revise glossary,https://github.com/pytorch/pytorch/issues/100904,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2023-05-08T22:10:19Z
issue,35633,Integration of Large Model Support in PyTorch,https://github.com/pytorch/pytorch/issues/35633,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: internals, module: memory usage",scan when time allows,2023-05-08T19:22:36Z
issue,50402,ValueError: optimizer got an empty parameter list,https://github.com/pytorch/pytorch/issues/50402,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2023-05-08T17:35:02Z
issue,100807,[torch.compile] returns output with WRONG SHAPE after `cat_slice_cat`,https://github.com/pytorch/pytorch/issues/100807,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2023-05-07T11:49:17Z
issue,100584,[doc] torch.scalar_tensor doc is missing,https://github.com/pytorch/pytorch/issues/100584,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2023-05-05T13:35:36Z
issue,84972,Add unit tests for test decorators,https://github.com/pytorch/pytorch/issues/84972,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tests,scan when time allows,2023-05-05T02:06:17Z
issue,66787,loading large model not finished after 16 hours,https://github.com/pytorch/pytorch/issues/66787,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: performance, module: serialization",scan when time allows,2023-05-04T05:20:59Z
issue,99640,Ban GradScaler scale from being less than 1,https://github.com/pytorch/pytorch/issues/99640,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision),scan when time allows,2023-05-03T17:23:22Z
issue,100425,Higher GPU consumption for Lenet-5 and LSTM models when compared to other frameworks,https://github.com/pytorch/pytorch/issues/100425,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: memory usage",scan when time allows,2023-05-03T15:28:52Z
issue,99630,Parameterisation of MultivariateNormal distribution using Cholesky decomposition of precision matrix,https://github.com/pytorch/pytorch/issues/99630,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distributions,scan when time allows,2023-05-03T15:26:18Z
issue,99287,Add `TORCH_ASSERT_ONLY_METHOD_OPERATORS` to functorch codebase,https://github.com/pytorch/pytorch/issues/99287,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: functorch,scan when time allows,2023-05-03T14:16:02Z
issue,99558,Batching rule not implemented for aten::narrow.Tensor,https://github.com/pytorch/pytorch/issues/99558,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: functorch,scan when time allows,2023-05-02T21:32:06Z
issue,18220,Context Manager that disables training mode with in a nn.Module.,https://github.com/pytorch/pytorch/issues/18220,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2023-05-01T19:19:22Z
issue,100244,torch.utils._content_store will deduplicate storage with identical contents; may be problematic for mutation,https://github.com/pytorch/pytorch/issues/100244,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: serialization,scan when time allows,2023-05-01T15:32:00Z
issue,87433,index_select() applied in sparse tensor can't backprop,https://github.com/pytorch/pytorch/issues/87433,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2023-05-01T14:33:11Z
issue,100145,This flag not work : torch.backends.cudnn.allow_tf32 = False,https://github.com/pytorch/pytorch/issues/100145,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2023-04-28T09:05:32Z
issue,94115,`cat` fails the gradient computation in forward mode with empty tensors when used with legacy vmap,https://github.com/pytorch/pytorch/issues/94115,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: edge cases, module: forward ad",scan when time allows,2023-04-27T15:33:03Z
issue,66357,OpInfos disabled for batched forward grad computation,https://github.com/pytorch/pytorch/issues/66357,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: forward ad, module: tests, module: vmap",scan when time allows,2023-04-27T02:59:50Z
issue,99299,There has implmenet bug in LTC IrBuilder's MakeSizeMul method.,https://github.com/pytorch/pytorch/issues/99299,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: lazy,scan when time allows,2023-04-26T19:33:01Z
issue,37246,`max_norm` parameter on nn.Embedding will fail inside nn.DataParallel,https://github.com/pytorch/pytorch/issues/37246,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: data parallel, module: nn",scan when time allows,2023-04-26T19:08:53Z
issue,93884,Importing tensorflow (2.12) before torch (2.0) hangs at import torch,https://github.com/pytorch/pytorch/issues/93884,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: binaries,scan when time allows,2023-04-26T16:49:46Z
issue,100012,Dataloader multiprocess loading with num_worker > 0 calls __main__ file to run,https://github.com/pytorch/pytorch/issues/100012,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2023-04-26T14:04:29Z
issue,98073,Add PrivateUse1 folder in aten/src/ATen,https://github.com/pytorch/pytorch/issues/98073,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: backend,scan when time allows,2023-04-26T14:01:11Z
issue,98414,"PyTorch 1.12, high failure rate for test_optim/test_nadam",https://github.com/pytorch/pytorch/issues/98414,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2023-04-25T05:15:21Z
issue,99790,Cannot compile torch 1.10 in CentOS 7.3,https://github.com/pytorch/pytorch/issues/99790,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2023-04-25T00:45:43Z
issue,99770,Deformable Convolution export to onnx,https://github.com/pytorch/pytorch/issues/99770,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2023-04-24T23:55:21Z
issue,99812,cuda.is_available() error,https://github.com/pytorch/pytorch/issues/99812,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2023-04-24T23:54:52Z
issue,99836,NTK notebook calculates wrong object - wrong output dimensions,https://github.com/pytorch/pytorch/issues/99836,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: functorch,scan when time allows,2023-04-24T23:52:23Z
issue,99681,`torch.ops.aten.empty` is not discoverable from `dir(torch.ops.aten)` until explicitly calling getattr,https://github.com/pytorch/pytorch/issues/99681,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: library,scan when time allows,2023-04-24T20:53:16Z
issue,98924,Integrate open device privateuse1 customized method registration,https://github.com/pytorch/pytorch/issues/98924,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: backend,scan when time allows,2023-04-24T17:50:11Z
issue,99710,pca_lowrank and svd_lowrank broken under automatic mixed precision.,https://github.com/pytorch/pytorch/issues/99710,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision), module: cuda, module: half, module: linear algebra",scan when time allows,2023-04-24T11:31:13Z
issue,99515,Support polyphase channelizer,https://github.com/pytorch/pytorch/issues/99515,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fft,scan when time allows,2023-04-24T07:40:55Z
issue,99690,"gpu training work well, but cpu training not work",https://github.com/pytorch/pytorch/issues/99690,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: fft",scan when time allows,2023-04-23T22:30:34Z
issue,96738,Many padding Module fail memory_format tests,https://github.com/pytorch/pytorch/issues/96738,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: intel, module: memory format, module: nn",scan when time allows,2023-04-23T07:14:28Z
issue,95945,CPU time performance is unstable,https://github.com/pytorch/pytorch/issues/95945,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: performance",scan when time allows,2023-04-23T05:02:53Z
issue,76751,Default repr of __get__ methods in __torch_function__ is bad,https://github.com/pytorch/pytorch/issues/76751,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: __torch_function__,scan when time allows,2023-04-22T15:58:55Z
issue,29637,Compiled functions can't take variable number of arguments,https://github.com/pytorch/pytorch/issues/29637,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2023-04-21T21:56:22Z
issue,99684,In torchelastic support running worker rank 0 on agent rank 0 consistently,https://github.com/pytorch/pytorch/issues/99684,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: elastic, oncall: distributed",scan when time allows,2023-04-21T20:11:09Z
issue,98200,torch.nn.init functions with `generator` argument,https://github.com/pytorch/pytorch/issues/98200,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: random",scan when time allows,2023-04-21T19:26:37Z
issue,96855,Performance Drop for linalg_ldl_factor and ldl_solve,https://github.com/pytorch/pytorch/issues/96855,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: functorch, module: vmap",scan when time allows,2023-04-20T20:39:33Z
issue,48486,[complex] torch.abs: does not match numpy,https://github.com/pytorch/pytorch/issues/48486,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: complex, module: nans and infs, module: numpy",scan when time allows,2023-04-20T10:50:14Z
issue,99138,torch.cuda.is_available() return False,https://github.com/pytorch/pytorch/issues/99138,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: docs",scan when time allows,2023-04-20T08:42:12Z
issue,67684,[feature request] torch.clamp on BoolTensors,https://github.com/pytorch/pytorch/issues/67684,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: boolean tensor, module: numpy",scan when time allows,2023-04-19T21:20:34Z
issue,71552,[docs] Tensor.uniform_ docs are not clear about whether from/to boundary values are included in sampling or not,https://github.com/pytorch/pytorch/issues/71552,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: random",scan when time allows,2023-04-19T21:12:18Z
issue,99410,"torch.nn.functional.multilabel_margin_loss cuda lacks checking of ""out of bound""",https://github.com/pytorch/pytorch/issues/99410,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: error checking, module: nn",scan when time allows,2023-04-19T20:20:41Z
issue,99351,we should make semantically meaningless positional arguments positional only in our operator API,https://github.com/pytorch/pytorch/issues/99351,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: python array api, module: python frontend",scan when time allows,2023-04-19T12:41:36Z
issue,27030,Tensorboard logging image WITH LABEL,https://github.com/pytorch/pytorch/issues/27030,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tensorboard,scan when time allows,2023-04-19T09:31:27Z
issue,52495,link failed when using custom build pytorch on Android,https://github.com/pytorch/pytorch/issues/52495,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: android, module: arm, module: undefined reference",scan when time allows,2023-04-19T06:57:09Z
issue,31815,ImportError: /home/xx/anaconda3/lib/python3.7/site-packages/torch/lib/libtorch_python.so: undefined symbol: _ZNK5torch3jit5Graph8toStringE,https://github.com/pytorch/pytorch/issues/31815,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2023-04-19T05:22:03Z
issue,53440,Add complex support for torch.unique,https://github.com/pytorch/pytorch/issues/53440,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: complex, module: numpy, module: python array api",scan when time allows,2023-04-19T05:01:54Z
issue,99407,Torch.fx.symbolic_trace removes some of the keys from module state_dict,https://github.com/pytorch/pytorch/issues/99407,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2023-04-19T02:22:26Z
issue,99414,The meta implementation of `index_put` does not do any check,https://github.com/pytorch/pytorch/issues/99414,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: meta tensors,scan when time allows,2023-04-19T02:05:13Z
issue,98792,DataLoader doesn't accept non-cpu device for loading.,https://github.com/pytorch/pytorch/issues/98792,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2023-04-18T15:04:02Z
issue,99359,[c++17] Replace lock_guard with scoped_lock,https://github.com/pytorch/pytorch/issues/99359,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: internals,scan when time allows,2023-04-18T14:36:10Z
issue,99316,torch.linalg.lstsq doc arguments error,https://github.com/pytorch/pytorch/issues/99316,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: linear algebra",scan when time allows,2023-04-18T12:31:42Z
issue,25092,No way to correctly reset weights of a model with spectral norm,https://github.com/pytorch/pytorch/issues/25092,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2023-04-18T09:07:45Z
issue,99265,the error message of torch.addcmul is wrong,https://github.com/pytorch/pytorch/issues/99265,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: error checking,scan when time allows,2023-04-17T17:13:30Z
issue,99310,Torch func Documentation for trees,https://github.com/pytorch/pytorch/issues/99310,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: functorch",scan when time allows,2023-04-17T16:00:03Z
issue,99248,tools PYTHONPATH trick in run_test.py does not work reliably,https://github.com/pytorch/pytorch/issues/99248,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: testing,scan when time allows,2023-04-17T15:26:48Z
issue,99147,`torch.sparse.sum` backward fails when reducing over dense dimensions.,https://github.com/pytorch/pytorch/issues/99147,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: sparse",scan when time allows,2023-04-17T08:22:47Z
issue,99160,"Run ChatRWKV on MBP(intel CPU)+eGPU[rx6800 16G], returna a very big number -9223372036854775808, looks like overflow",https://github.com/pytorch/pytorch/issues/99160,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: mps,scan when time allows,2023-04-14T20:21:05Z
issue,99042,Memory allocation issues in distributions.multivariate_normal.MultivariateNormal,https://github.com/pytorch/pytorch/issues/99042,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distributions,scan when time allows,2023-04-14T17:19:14Z
issue,99142,"More Nested Tensor Functionality (layer_norm, cross_entropy / log_softmax&nll_loss)",https://github.com/pytorch/pytorch/issues/99142,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nestedtensor,scan when time allows,2023-04-14T16:29:52Z
issue,99023,Dtype changes while going from FX graph -> Torchscript,https://github.com/pytorch/pytorch/issues/99023,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2023-04-14T15:49:23Z
issue,36041,[DISCUSSION] Better user experience for debugging on Windows,https://github.com/pytorch/pytorch/issues/36041,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: windows",scan when time allows,2023-04-13T22:34:03Z
issue,98917,Change module to module_ in torch/csrc/api/include/torch/python.h,https://github.com/pytorch/pytorch/issues/98917,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2023-04-13T21:47:57Z
issue,98977,[BE] Dedup the functorch skipOps mechanism and the common_method_invocations one,https://github.com/pytorch/pytorch/issues/98977,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: testing,scan when time allows,2023-04-13T19:58:54Z
issue,98942,[torch.fx] Upgrade on node info,https://github.com/pytorch/pytorch/issues/98942,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2023-04-13T07:55:40Z
issue,98863,Problem with instalation torch2 on a100+cu12.1,https://github.com/pytorch/pytorch/issues/98863,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: cuda",scan when time allows,2023-04-12T21:06:08Z
issue,1794,Feature request: reverse_padded_sequence,https://github.com/pytorch/pytorch/issues/1794,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rnn,scan when time allows,2023-04-12T16:04:14Z
issue,98907,Move template code to header,https://github.com/pytorch/pytorch/issues/98907,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp,scan when time allows,2023-04-12T04:02:19Z
issue,98805,Discrepancy of supported Python versions between Get Started page and index of pre-built binaries for PIP installation,https://github.com/pytorch/pytorch/issues/98805,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2023-04-11T23:40:12Z
issue,98673,[ux] Non-blocking tensor constructors,https://github.com/pytorch/pytorch/issues/98673,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tensor creation,scan when time allows,2023-04-11T21:05:27Z
issue,98871,Document the user-facing API for the component-level logging system,https://github.com/pytorch/pytorch/issues/98871,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2023-04-11T20:54:47Z
issue,98587,Cannot use AT_CUDA_DRIVER_CHECK from user code,https://github.com/pytorch/pytorch/issues/98587,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cpp-extensions, module: cuda, module: internals",scan when time allows,2023-04-11T19:19:40Z
issue,98557,torch.jit.script codegen warning with cuda and vmap,https://github.com/pytorch/pytorch/issues/98557,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: functorch, oncall: jit",scan when time allows,2023-04-11T12:42:15Z
issue,98808,[FSDP] move up the first all gather,https://github.com/pytorch/pytorch/issues/98808,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: fsdp, oncall: distributed",scan when time allows,2023-04-11T12:37:52Z
issue,91334,Pytorch 1.13 conda package with cuda requires too many unneccessary packages,https://github.com/pytorch/pytorch/issues/91334,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: cuda",scan when time allows,2023-04-11T07:33:33Z
issue,98617,Add test/distributed/test_c10d_mpi.py,https://github.com/pytorch/pytorch/issues/98617,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2023-04-11T05:39:48Z
issue,98286,"When I use the DDP model, I use a custom loss function, when the batch size changes during training, the process will be stuck.",https://github.com/pytorch/pytorch/issues/98286,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2023-04-11T05:28:29Z
issue,98515,torch.cond should work with expressions involving SymInt,https://github.com/pytorch/pytorch/issues/98515,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: functorch,scan when time allows,2023-04-10T17:04:39Z
issue,98541,Memory corruption using torch.ops.* to access re-registered operator,https://github.com/pytorch/pytorch/issues/98541,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: internals,scan when time allows,2023-04-10T16:56:57Z
issue,98499,`torch.nn.utils.rnn.unpad_sequence` modifies arguments in-place,https://github.com/pytorch/pytorch/issues/98499,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: rnn",scan when time allows,2023-04-10T16:56:18Z
issue,89730,torch.addbmm throws different exception differences on CPU and GPU.,https://github.com/pytorch/pytorch/issues/89730,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: error checking",scan when time allows,2023-04-10T13:13:07Z
issue,88144,AvgPool2D output shapes are inconsistent when ceil_mode=True,https://github.com/pytorch/pytorch/issues/88144,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: pooling",scan when time allows,2023-04-10T05:42:31Z
issue,41164,Pytorch.js would be nice!,https://github.com/pytorch/pytorch/issues/41164,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2023-04-09T17:45:08Z
issue,89393,Edge case: CPU bool abs is not supported,https://github.com/pytorch/pytorch/issues/89393,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: edge cases, module: numpy",scan when time allows,2023-04-09T12:14:18Z
issue,42095,[RFC] Add Windows support to torch.distributed package,https://github.com/pytorch/pytorch/issues/42095,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: windows, oncall: distributed",scan when time allows,2023-04-09T09:16:45Z
issue,37487,Log-linear version of cumsum and cumprod,https://github.com/pytorch/pytorch/issues/37487,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: performance,scan when time allows,2023-04-08T06:29:07Z
issue,91608,SSL: CERTIFICATE_VERIFY_FAILED while trying to download pretrained model within a company that transforms SSL certificates for security purposes,https://github.com/pytorch/pytorch/issues/91608,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: hub,scan when time allows,2023-04-08T01:45:46Z
issue,97498,Function Registry for extending collate_fn,https://github.com/pytorch/pytorch/issues/97498,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2023-04-07T22:32:49Z
issue,98498,"Higher order derivatives not working when setting compute device to `torch.device(""mps"")`",https://github.com/pytorch/pytorch/issues/98498,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: mps",scan when time allows,2023-04-07T14:33:31Z
issue,24468,[Tensorboard] Write summaries to S3 or GCS bucket,https://github.com/pytorch/pytorch/issues/24468,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tensorboard,scan when time allows,2023-04-07T08:05:03Z
issue,28594,[BUG] Can't Deepcopy module with weightnorm,https://github.com/pytorch/pytorch/issues/28594,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2023-04-07T07:08:50Z
issue,94912,PR #88607 breaks build for POWER9 CPU,https://github.com/pytorch/pytorch/issues/94912,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: power",scan when time allows,2023-04-06T15:38:31Z
issue,98484,Improvements to FSDP debugability,https://github.com/pytorch/pytorch/issues/98484,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: fsdp, oncall: distributed",scan when time allows,2023-04-06T11:30:35Z
issue,88464,MaxPool1D output shapes can be negative when ceil_mode=True,https://github.com/pytorch/pytorch/issues/88464,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: pooling",scan when time allows,2023-04-06T07:37:40Z
issue,92128,Forward arguments are not updated in DDP,https://github.com/pytorch/pytorch/issues/92128,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2023-04-06T06:55:35Z
issue,98169,The operator 'aten::_weight_norm_interface' is not currently implemented for the MPS device.,https://github.com/pytorch/pytorch/issues/98169,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: mps,scan when time allows,2023-04-05T19:28:37Z
issue,98409,`torch.Tensor.layout` is not documented,https://github.com/pytorch/pytorch/issues/98409,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: python frontend",scan when time allows,2023-04-05T17:17:39Z
issue,98075,Unexpected results with torch.nn.functional.layer_norm,https://github.com/pytorch/pytorch/issues/98075,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: norms and normalization, module: numerical-stability",scan when time allows,2023-04-05T15:05:12Z
issue,92251,"`torch.load(..., map_location=""cuda:0"")` allocates memory on both CPU and GPU",https://github.com/pytorch/pytorch/issues/92251,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: python frontend, module: serialization",scan when time allows,2023-04-05T14:56:37Z
issue,98338,[functorch] [vmap] tests fail when `_set_vmap_fallback_enabled(False)`.,https://github.com/pytorch/pytorch/issues/98338,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: functorch,scan when time allows,2023-04-05T14:18:19Z
issue,95369,build failed when strictly following the guidelines,https://github.com/pytorch/pytorch/issues/95369,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2023-04-05T10:01:02Z
issue,73554,Build failure using GCC 11.2.0,https://github.com/pytorch/pytorch/issues/73554,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: mkldnn, module: third_party",scan when time allows,2023-04-05T09:12:04Z
issue,95024,cuDNN doesn't support convolutions with more than `INT_MAX` elements and native kernel uses too much memory,https://github.com/pytorch/pytorch/issues/95024,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: cudacachingallocator, module: cudnn",scan when time allows,2023-04-05T03:48:48Z
issue,98070,Request custom backend device memory Allocator.,https://github.com/pytorch/pytorch/issues/98070,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: backend, module: cudacachingallocator, module: memory usage",scan when time allows,2023-04-05T00:44:08Z
issue,98004,matmul with CSR matrix in inference mode throws an exception,https://github.com/pytorch/pytorch/issues/98004,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: sparse",scan when time allows,2023-04-05T00:40:45Z
issue,97902,Multiple model init using OpenMP in c++ does not speed up,https://github.com/pytorch/pytorch/issues/97902,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: multiprocessing,scan when time allows,2023-04-05T00:40:43Z
issue,97638,Regression in jit for f-strings with new lines,https://github.com/pytorch/pytorch/issues/97638,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2023-04-05T00:40:41Z
issue,97595,can get_submodule be called within a ScriptFunction ?,https://github.com/pytorch/pytorch/issues/97595,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2023-04-05T00:40:39Z
issue,73638,addcdiv is failing the ASAN test for zero divisors,https://github.com/pytorch/pytorch/issues/73638,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: error checking, module: tests",scan when time allows,2023-04-05T00:40:38Z
issue,89420,Edge case: torch.baddbmm supports double x int8 x int8 inputs on CPU but not CUDA,https://github.com/pytorch/pytorch/issues/89420,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: edge cases, module: linear algebra, module: type promotion",scan when time allows,2023-04-05T00:39:00Z
issue,89419,torch.equal can still run successfully when the parameter types are different.,https://github.com/pytorch/pytorch/issues/89419,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: numpy, module: type promotion",scan when time allows,2023-04-05T00:26:21Z
issue,97236,Non-deterministic results when training a model on GPU with MPS backend,https://github.com/pytorch/pytorch/issues/97236,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: determinism, module: mps",scan when time allows,2023-04-04T21:55:03Z
issue,89418,"torch.floor_divide: The dividend of torch.floor_divide is set to 0, but it can still run on the GPU.",https://github.com/pytorch/pytorch/issues/89418,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: error checking,scan when time allows,2023-04-04T15:52:18Z
issue,89381,torch.nn.ReplicationPad1d:The description of the exception information thrown is not accurate,https://github.com/pytorch/pytorch/issues/89381,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: error checking, module: padding",scan when time allows,2023-04-04T15:30:42Z
issue,98260,Broken mypy check in test_type_hints.py::TestTypeHints::test_doc_examples,https://github.com/pytorch/pytorch/issues/98260,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: typing,scan when time allows,2023-04-04T14:06:34Z
issue,98193,"RuntimeError: CUDA error: an illegal memory access was encountered, torch/cuda/streams.py"", line 94, in synchronize",https://github.com/pytorch/pytorch/issues/98193,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2023-04-04T13:47:35Z
issue,70073,Feature request: [STFT] Add warning message if signal length is not a multiple of hop_length in torch.stft,https://github.com/pytorch/pytorch/issues/70073,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: complex, module: fft",scan when time allows,2023-04-04T12:32:58Z
issue,1512,[feature request] Support tensors of different sizes as batch elements in DataLoader,https://github.com/pytorch/pytorch/issues/1512,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dataloader, module: nestedtensor",scan when time allows,2023-04-04T10:43:01Z
issue,89369,RuntimeError: CUDA error: device-side assert triggered,https://github.com/pytorch/pytorch/issues/89369,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: loss,scan when time allows,2023-04-04T01:30:25Z
issue,89342,torch.mm: Exceptions thrown on the CPU and GPU are inconsistent,https://github.com/pytorch/pytorch/issues/89342,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: error checking",scan when time allows,2023-04-04T01:06:10Z
issue,97849,TORCH_LIBRARIES variable leads to undefined reference function error in compiling while using libtorch in c++,https://github.com/pytorch/pytorch/issues/97849,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp,scan when time allows,2023-04-03T17:09:55Z
issue,18993,ninja: build stopped: subcommand failed.,https://github.com/pytorch/pytorch/issues/18993,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2023-04-03T01:59:00Z
issue,77253,lintrunner not working,https://github.com/pytorch/pytorch/issues/77253,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: lint,scan when time allows,2023-04-02T18:21:39Z
issue,55399,[FX][testing] Test symbolic tracing of detectron2,https://github.com/pytorch/pytorch/issues/55399,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2023-04-02T02:40:06Z
issue,97961,[CI/Infra] Record keeping: runner shutdown spike,https://github.com/pytorch/pytorch/issues/97961,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2023-04-02T01:27:30Z
issue,93947,[RFC] Support Huge Model Init Without mallocs for Compile/Distributed Use Cases,https://github.com/pytorch/pytorch/issues/93947,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2023-04-01T00:17:01Z
issue,97030,DDP static graph fails for static model,https://github.com/pytorch/pytorch/issues/97030,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ddp, oncall: distributed",scan when time allows,2023-03-31T19:27:25Z
issue,83244,torch.nn.Upsample's error message is inconsistent with the documentation,https://github.com/pytorch/pytorch/issues/83244,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: nn",scan when time allows,2023-03-31T15:35:28Z
issue,97966,torch.randn signature is missing generator,https://github.com/pytorch/pytorch/issues/97966,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: python frontend, module: random",scan when time allows,2023-03-31T14:44:53Z
issue,97676,Pytorch 2 compile + fsdp + transformers crash,https://github.com/pytorch/pytorch/issues/97676,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: xla,scan when time allows,2023-03-30T21:20:19Z
issue,97991,functional collective should respect the whole mesh,https://github.com/pytorch/pytorch/issues/97991,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2023-03-30T16:40:04Z
issue,7617,"checkpoint(function, *args) should have the same requires_grad as function(*args)",https://github.com/pytorch/pytorch/issues/7617,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: activation checkpointing, module: autograd",scan when time allows,2023-03-30T14:16:59Z
issue,92285,InstanceNorm operator support for Vulkan devices,https://github.com/pytorch/pytorch/issues/92285,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: vulkan,scan when time allows,2023-03-30T09:34:19Z
issue,95934,arange bug,https://github.com/pytorch/pytorch/issues/95934,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: numerical-stability,scan when time allows,2023-03-30T03:26:42Z
issue,97915,Investigate Lazy{*}Norm{*}d modules no batch dim support,https://github.com/pytorch/pytorch/issues/97915,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2023-03-29T21:54:36Z
issue,97823,Using `param in param_list` can trigger `non-singleton dimension` error?,https://github.com/pytorch/pytorch/issues/97823,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python frontend,scan when time allows,2023-03-29T18:34:16Z
issue,97852,Functionalize crashes on train_step GraphModule,https://github.com/pytorch/pytorch/issues/97852,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: functionalization,scan when time allows,2023-03-29T18:09:21Z
issue,97500,.set_ operation on a view (detach()) of the view tensor changes grad_fn of the original view tensor from ViewBackward0 to AsStridedBackward0,https://github.com/pytorch/pytorch/issues/97500,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2023-03-29T15:59:40Z
issue,81412,num_worker and prefetch_factor in DataLoader do not scale,https://github.com/pytorch/pytorch/issues/81412,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dataloader, module: multiprocessing",scan when time allows,2023-03-29T02:05:18Z
issue,97802,Some c++ library docstrings incorrectly linked/repeated,https://github.com/pytorch/pytorch/issues/97802,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: docs, module: mps",scan when time allows,2023-03-28T18:53:12Z
issue,97761,torch.testing.assert_close: allow check to fail on part on the input,https://github.com/pytorch/pytorch/issues/97761,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: testing,scan when time allows,2023-03-28T14:42:44Z
issue,90284,Public API definition is not compatible with `torch.testing`,https://github.com/pytorch/pytorch/issues/90284,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: python frontend, module: testing",scan when time allows,2023-03-28T14:27:55Z
issue,97503,torch.cppExtension won't work with wsl2,https://github.com/pytorch/pytorch/issues/97503,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: wsl,scan when time allows,2023-03-28T07:55:00Z
issue,97659,How do I get the original object wrapped by the torch.fx.proxy class？,https://github.com/pytorch/pytorch/issues/97659,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: fx,scan when time allows,2023-03-28T02:10:58Z
issue,97656,[bug] Internal assert failed when using pyro,https://github.com/pytorch/pytorch/issues/97656,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: distributions, module: linear algebra, module: macos, module: python frontend",scan when time allows,2023-03-27T20:58:01Z
issue,97653,transposed 2d copy bfloat16 support,https://github.com/pytorch/pytorch/issues/97653,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2023-03-27T18:17:08Z
issue,97631,torch.zeros_like on a zero-sized BSR/BSC tensor results invalid tensor,https://github.com/pytorch/pytorch/issues/97631,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: sparse",scan when time allows,2023-03-27T18:07:10Z
issue,97597,Insufficient MPS Documentation,https://github.com/pytorch/pytorch/issues/97597,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: mps",scan when time allows,2023-03-27T17:37:24Z
issue,97189,torch.randint range for torch.int64 dtype seems wrong,https://github.com/pytorch/pytorch/issues/97189,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python frontend,scan when time allows,2023-03-27T17:28:45Z
issue,97504,"Burn benchmark suites into CI docker image. Not only this saves test time, but also it will get rid of occasional model installation failures. (@weiwangmeta )",https://github.com/pytorch/pytorch/issues/97504,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2023-03-27T16:45:42Z
issue,65153,Unification of model initialization methods / naming across domain libraries + support of skip_init,https://github.com/pytorch/pytorch/issues/65153,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: initialization,scan when time allows,2023-03-27T15:48:57Z
issue,97484,[DTensor] Add a unittest to cover default PG condition for DeviceMesh,https://github.com/pytorch/pytorch/issues/97484,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2023-03-27T14:49:13Z
issue,97114,[Feature Proposal: New Distributed Training Algorithms] LSGD and EASGD,https://github.com/pytorch/pytorch/issues/97114,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2023-03-27T14:40:19Z
issue,97670,[FSDP] test model.eval() + keep_low_precision_grads,https://github.com/pytorch/pytorch/issues/97670,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2023-03-27T14:14:24Z
issue,24904,tensorboard add_graph error,https://github.com/pytorch/pytorch/issues/24904,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tensorboard,scan when time allows,2023-03-27T10:00:19Z
issue,95921,ROCm distributed flaky on test_distributed_spawn,https://github.com/pytorch/pytorch/issues/95921,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rocm,scan when time allows,2023-03-24T22:23:53Z
issue,97329,"torch.ops.aten.pow(2.0, 3) return unexpected value with complex type",https://github.com/pytorch/pytorch/issues/97329,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: custom-operators, module: internals",scan when time allows,2023-03-24T17:31:32Z
issue,76244,Initial integration of ZenDNN as backend into PyTorch,https://github.com/pytorch/pytorch/issues/76244,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: convolution,scan when time allows,2023-03-24T17:08:43Z
issue,97118,nn.Conv function to compute conv formula,https://github.com/pytorch/pytorch/issues/97118,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: nn",scan when time allows,2023-03-24T16:54:10Z
issue,80017,RPC init fails and crashes when world_size is greater than 18,https://github.com/pytorch/pytorch/issues/80017,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: rpc, oncall: distributed",scan when time allows,2023-03-24T10:38:15Z
issue,97397,Import fails when both `USE_TENSORPIPE=OFF` and `USE_DISTRIBUTED=ON`.,https://github.com/pytorch/pytorch/issues/97397,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: tensorpipe",scan when time allows,2023-03-24T05:42:55Z
issue,97408,Further memcopy improvement at FX body level,https://github.com/pytorch/pytorch/issues/97408,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2023-03-24T05:41:53Z
issue,96613,The output of torch.histc is incorrect on both CPU and CUDA,https://github.com/pytorch/pytorch/issues/96613,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: edge cases,scan when time allows,2023-03-24T01:29:21Z
issue,97421,"After the release of pytorch 2.0.0, the compilation of ACLs is problematic.",https://github.com/pytorch/pytorch/issues/97421,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2023-03-23T20:22:52Z
issue,97283,`jacfwd` fails when computing the gradient for `channels_last` tensor,https://github.com/pytorch/pytorch/issues/97283,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: functorch, module: memory format",scan when time allows,2023-03-23T08:04:27Z
issue,40087,Make Scaling in BatchNorm optional,https://github.com/pytorch/pytorch/issues/40087,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2023-03-23T04:20:07Z
issue,97068,[RFC] CPU float16 performance optimization on eager mode.,https://github.com/pytorch/pytorch/issues/97068,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: half",scan when time allows,2023-03-23T02:52:42Z
issue,40967,Inconsistent behaviour when parameter appears multiple times in parameter list,https://github.com/pytorch/pytorch/issues/40967,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: optimizer",scan when time allows,2023-03-22T16:24:55Z
issue,33248,PyTorch 1.4.0 CUDA initialization error with CPU-only (multiprocessing) on Python 3.7.5,https://github.com/pytorch/pytorch/issues/33248,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: cuda",scan when time allows,2023-03-22T11:23:31Z
issue,97286,`jacrev` and `jacfwd` raise an error that `Sparse CSR tensors do not have strides`,https://github.com/pytorch/pytorch/issues/97286,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: functorch, module: sparse",scan when time allows,2023-03-21T22:33:29Z
issue,81750,Modernize logging tensor in torch.testing._internal,https://github.com/pytorch/pytorch/issues/81750,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: internals, module: logging",scan when time allows,2023-03-21T19:46:30Z
issue,67913,Tracker: ModuleInfo-based testing,https://github.com/pytorch/pytorch/issues/67913,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: tests",scan when time allows,2023-03-21T17:14:21Z
issue,97196,Sequential/Partial unpickling and loading of models,https://github.com/pytorch/pytorch/issues/97196,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: pickle,scan when time allows,2023-03-21T09:58:33Z
issue,97004,Small learning rate with `capturable=True` causes Adam optimizer to blow up model parameters.,https://github.com/pytorch/pytorch/issues/97004,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2023-03-21T09:04:43Z
issue,97088,A Segment Fault can be triggered in torch.embedding,https://github.com/pytorch/pytorch/issues/97088,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: edge cases,scan when time allows,2023-03-20T23:56:22Z
issue,97087,A Segment Fault can be triggered in torch.adjoint,https://github.com/pytorch/pytorch/issues/97087,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: edge cases,scan when time allows,2023-03-20T23:56:07Z
issue,97026,How to get list of all valid devices?,https://github.com/pytorch/pytorch/issues/97026,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2023-03-20T23:49:13Z
issue,97014,deprecate integral and boolean dtype support torch.logit and torch.special.logit,https://github.com/pytorch/pytorch/issues/97014,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: deprecation, module: special",scan when time allows,2023-03-20T15:55:20Z
issue,81692,Failed to static link latest cuDNN while compiling,https://github.com/pytorch/pytorch/issues/81692,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cudnn",scan when time allows,2023-03-18T12:32:46Z
issue,96908,Doing inplace on a inplace view of tensor that retains_grad triggers internal assert,https://github.com/pytorch/pytorch/issues/96908,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2023-03-17T21:03:44Z
issue,38624,Wheels not manylinux1 compliant,https://github.com/pytorch/pytorch/issues/38624,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: binaries,scan when time allows,2023-03-17T13:37:15Z
issue,94333,cpu log1p for bfloat16 gives wrong result.,https://github.com/pytorch/pytorch/issues/94333,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bfloat16, module: cpu",scan when time allows,2023-03-17T02:45:40Z
issue,96769,`sparse.mm` triggers INTERNAL ASSERT FAILED when backwarding,https://github.com/pytorch/pytorch/issues/96769,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: sparse",scan when time allows,2023-03-16T16:12:24Z
issue,70555,[feature request] Exponential moving average (EMA) of a tensor across a dimension,https://github.com/pytorch/pytorch/issues/70555,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: numpy,scan when time allows,2023-03-16T09:02:24Z
issue,96789,`cumprod` triggers INTERNAL ASSERT FAILED when `out` is a tensor on cuda but input is on cpu,https://github.com/pytorch/pytorch/issues/96789,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: assert failure, module: error checking",scan when time allows,2023-03-16T01:01:37Z
issue,95776,`torch.Tensor.is_set_to` raises `NotImplementedError` when inputs contain sparse tensor,https://github.com/pytorch/pytorch/issues/95776,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2023-03-15T16:12:34Z
issue,76689,Performance bad on ARM AArch64 for PyTorch C++,https://github.com/pytorch/pytorch/issues/76689,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: arm, module: build",scan when time allows,2023-03-15T13:15:04Z
issue,91760,Inconsistent rank among torch.distributed primitives,https://github.com/pytorch/pytorch/issues/91760,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2023-03-15T10:28:08Z
issue,95622,"Testing InvokeAI 2.3.1.post1, using mps, with PyTorch nightly dev20230226 yields RuntimeError cross-device copies are not allowed!)",https://github.com/pytorch/pytorch/issues/95622,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: mps,scan when time allows,2023-03-15T02:15:43Z
issue,95756,`torch.nanmedian` return a negative value when input is empty,https://github.com/pytorch/pytorch/issues/95756,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: error checking,scan when time allows,2023-03-15T00:55:43Z
issue,94779,Better Numpy API (interoperability between ML frameworks),https://github.com/pytorch/pytorch/issues/94779,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: numpy,scan when time allows,2023-03-14T21:57:09Z
issue,96704,`logical_xx` operations trigger INTERNAL ASSERT FAIL when `input` is complex tensor on cuda and `other` is on cpu,https://github.com/pytorch/pytorch/issues/96704,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: complex,scan when time allows,2023-03-14T19:18:38Z
issue,96585,Proposal: Disable GC in test suite; GC after every test case,https://github.com/pytorch/pytorch/issues/96585,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: devx, module: flaky-tests, module: infra, module: testing",scan when time allows,2023-03-14T19:12:52Z
issue,96305,"Reuse autograd.grad graph for rapid, repeated gradient calculation",https://github.com/pytorch/pytorch/issues/96305,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: cuda graphs",scan when time allows,2023-03-14T13:17:17Z
issue,31461,Make RRef.to_here() non-blocking,https://github.com/pytorch/pytorch/issues/31461,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rpc,scan when time allows,2023-03-14T10:16:58Z
issue,90998,"Crash in `index_select` with singleton `self`, non-singleton `index`",https://github.com/pytorch/pytorch/issues/90998,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2023-03-14T08:11:00Z
issue,96036,Questions and Possible Features: Pytorch RPC 'future.wait()' will not release GIL which will block other thread's execution when using multithreading.,https://github.com/pytorch/pytorch/issues/96036,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: multithreading,scan when time allows,2023-03-14T04:31:54Z
issue,96615,Tensor Permutation Along Given Axis,https://github.com/pytorch/pytorch/issues/96615,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python frontend,scan when time allows,2023-03-13T22:37:18Z
issue,96629,Dataloader should kill & restart workers when timeout is hit,https://github.com/pytorch/pytorch/issues/96629,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2023-03-13T22:35:10Z
issue,83509,Build and Run QNNPACK on X86,https://github.com/pytorch/pytorch/issues/83509,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2023-03-13T14:56:55Z
issue,95572,Support datatype argument for torch.distributed.all_gather() (And the whole distributed module),https://github.com/pytorch/pytorch/issues/95572,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: c10d, module: nccl",scan when time allows,2023-03-13T10:55:47Z
issue,94441,`get_debug_state` a script function causes INTERNAL ASSERT FAILED,https://github.com/pytorch/pytorch/issues/94441,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2023-03-13T05:03:39Z
issue,96523,"MPS Backend Doc, model = YourFavoriteNet() not defined",https://github.com/pytorch/pytorch/issues/96523,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: mps",scan when time allows,2023-03-12T12:43:43Z
issue,74901,"torch.fx.wrap will not work, when encapsulate the code",https://github.com/pytorch/pytorch/issues/74901,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2023-03-11T07:25:59Z
issue,93121,[RFC] Flop counters in PyTorch,https://github.com/pytorch/pytorch/issues/93121,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python frontend,scan when time allows,2023-03-10T22:59:28Z
issue,57776,[rfc] Build a debug tool to help users find which parameters are unused during DDP training,https://github.com/pytorch/pytorch/issues/57776,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2023-03-10T21:51:41Z
issue,96494,The sign of torch.distributions.transforms.PowerTransform seems to be incorrect,https://github.com/pytorch/pytorch/issues/96494,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distributions,scan when time allows,2023-03-10T21:04:58Z
issue,69520,race condition of agent's TCP store on exit_barrier(),https://github.com/pytorch/pytorch/issues/69520,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: r2p,scan when time allows,2023-03-10T16:28:03Z
issue,96518,fft should ignore dims with shape 1,https://github.com/pytorch/pytorch/issues/96518,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fft,scan when time allows,2023-03-10T16:18:21Z
issue,96447,Information about CPU in `collect_env` is too verbose,https://github.com/pytorch/pytorch/issues/96447,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: collect_env.py,scan when time allows,2023-03-10T16:13:19Z
issue,96428,Compressed sparse constructor allows mixed `int32/int64` indices which leads to dtype promotion/demotion in conversions.,https://github.com/pytorch/pytorch/issues/96428,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2023-03-10T16:12:00Z
issue,96412,Proxy Options for Pytorch Hub,https://github.com/pytorch/pytorch/issues/96412,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: hub,scan when time allows,2023-03-10T16:11:55Z
issue,95953,The torch.sparse document's typo error,https://github.com/pytorch/pytorch/issues/95953,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: sparse",scan when time allows,2023-03-09T17:40:29Z
issue,94675,"`UnsupportedOperatorError`, `OnnxExporterError` and `SymbolicValueError` related to MultiheadAttention export to onnx with torch.jit.script",https://github.com/pytorch/pytorch/issues/94675,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2023-03-09T17:37:24Z
issue,58812,"Transferring tensor to the gpu and converting dtype in a single call to .to() is slower than first transferring, then converting.",https://github.com/pytorch/pytorch/issues/58812,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: performance",scan when time allows,2023-03-09T13:47:04Z
issue,96236,nn.interpolate scale_factor floors output size with floating,https://github.com/pytorch/pytorch/issues/96236,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: edge cases, module: nn",scan when time allows,2023-03-09T12:02:19Z
issue,96318,Support managed memory backed dlpack with torch.from_dlpack,https://github.com/pytorch/pytorch/issues/96318,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dlpack,scan when time allows,2023-03-09T12:01:35Z
issue,95815,Let Nested Tensor Metadata be cached on GPU,https://github.com/pytorch/pytorch/issues/95815,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nestedtensor,scan when time allows,2023-03-09T00:05:45Z
issue,49370,"PyEval_SaveThread: the function must be called with the GIL held, but the GIL is released",https://github.com/pytorch/pytorch/issues/49370,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: pybind,scan when time allows,2023-03-08T22:19:18Z
issue,94395,`func.jacrev()` should be implemented as `func.jacfwd().mT.contiguous()`,https://github.com/pytorch/pytorch/issues/94395,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: complex, module: functorch",scan when time allows,2023-03-08T20:47:41Z
issue,96337,pytreeify decorators,https://github.com/pytorch/pytorch/issues/96337,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: functorch, module: pytree",scan when time allows,2023-03-08T20:41:54Z
issue,94593,Memory corruptions can be triggered in torch._remove_batch_dim,https://github.com/pytorch/pytorch/issues/94593,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: vmap,scan when time allows,2023-03-08T20:39:29Z
issue,95872,corrupted size vs prev size error,https://github.com/pytorch/pytorch/issues/95872,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ddp,scan when time allows,2023-03-08T18:25:35Z
issue,96140,mkldnn matmul kernel may be slower than openblas kernel for very small tensor shapes,https://github.com/pytorch/pytorch/issues/96140,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: mkldnn, module: performance",scan when time allows,2023-03-08T17:26:27Z
issue,89675,[ONNX] torch.onnx.export snapshots the grads as constants in onnx when op is in cuda device,https://github.com/pytorch/pytorch/issues/89675,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2023-03-08T13:54:11Z
issue,89634,[ONNX] torch.onnx.export can not export the grad of conv when the op is in CPU,https://github.com/pytorch/pytorch/issues/89634,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2023-03-08T13:54:04Z
issue,44551,PyTorch 1.6 DataParallel causes CUDNN_STATUS_BAD_PARAM in backward pass,https://github.com/pytorch/pytorch/issues/44551,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: cudnn, module: data parallel",scan when time allows,2023-03-08T11:24:00Z
issue,95207,"new backend privateuseone with ""to"" op",https://github.com/pytorch/pytorch/issues/95207,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: backend,scan when time allows,2023-03-07T18:47:19Z
issue,78248,"USE_NATIVE_ARCH flag causes nvcc build failure due to ""'arch=native': expected a number""",https://github.com/pytorch/pytorch/issues/78248,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2023-03-07T18:35:02Z
issue,96187,`gradgradcheck` does not work with sparse inputs.,https://github.com/pytorch/pytorch/issues/96187,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: sparse",scan when time allows,2023-03-07T18:13:06Z
issue,70920,`unique` should be split into four partial functions,https://github.com/pytorch/pytorch/issues/70920,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python array api,scan when time allows,2023-03-07T09:32:58Z
issue,70914,support setting `step` in `arange` without setting `end`,https://github.com/pytorch/pytorch/issues/70914,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: python array api, module: tensor creation",scan when time allows,2023-03-07T09:32:01Z
issue,61649,Move `tensordot` into `torch.linalg`,https://github.com/pytorch/pytorch/issues/61649,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: linear algebra, module: python array api",scan when time allows,2023-03-07T09:30:58Z
issue,58742,Python Array API New Operators Tracker,https://github.com/pytorch/pytorch/issues/58742,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python array api,scan when time allows,2023-03-07T09:28:34Z
issue,70910,`eye` should support other diagonals than the main one,https://github.com/pytorch/pytorch/issues/70910,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: python array api, module: tensor creation",scan when time allows,2023-03-07T09:20:17Z
issue,63935,Inconsistent behaviour of index put when assigning from other type,https://github.com/pytorch/pytorch/issues/63935,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: advanced indexing, module: type promotion",scan when time allows,2023-03-07T08:50:49Z
issue,55454,Include the cuda headers in the libtorch pre-built packages,https://github.com/pytorch/pytorch/issues/55454,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: build, module: cuda",scan when time allows,2023-03-07T02:48:56Z
issue,96136,`torch.utils.checkpoint` should avoid updating BatchNorm statistics twice,https://github.com/pytorch/pytorch/issues/96136,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: activation checkpointing,scan when time allows,2023-03-07T02:10:27Z
issue,71683,EMA optimizer: class-form and function-form (using new foreach_lerp) - can be used for explicit robust updates of BatchNorm stats,https://github.com/pytorch/pytorch/issues/71683,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: mta, module: optimizer",scan when time allows,2023-03-06T22:16:27Z
issue,95944,training hangs at line torch.cuda.synchronize(),https://github.com/pytorch/pytorch/issues/95944,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: deadlock",scan when time allows,2023-03-06T15:50:19Z
issue,95956,[vulkan] missing aten::reflection_pad1d.out operator,https://github.com/pytorch/pytorch/issues/95956,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: vulkan,scan when time allows,2023-03-06T15:45:46Z
issue,90827,Support for Pylint,https://github.com/pytorch/pytorch/issues/90827,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: lint,scan when time allows,2023-03-06T15:40:13Z
issue,88893,RuntimeError: derivative for aten::mps_max_pool2d_backward is not implemented,https://github.com/pytorch/pytorch/issues/88893,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: mps,scan when time allows,2023-03-06T03:43:14Z
issue,79851,[bazel] [ci] `//:module_test` CUDA error: CUDA driver version is insufficient for CUDA runtime version,https://github.com/pytorch/pytorch/issues/79851,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bazel, module: cuda",scan when time allows,2023-03-03T18:12:06Z
issue,70870,Add flag for functional.Jacobian to return output as well,https://github.com/pytorch/pytorch/issues/70870,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2023-03-03T14:09:31Z
issue,76522,`torch.clamp` does not distribute gradients as element-wise`min/max` do,https://github.com/pytorch/pytorch/issues/76522,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2023-03-03T11:16:22Z
issue,95718,functorch.compile.memory_efficient_fusion errors with: RuntimeError: forward() Expected a value of type 'Tensor (inferred)' for argument 'primals_356' but instead found type 'int'.,https://github.com/pytorch/pytorch/issues/95718,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: functorch,scan when time allows,2023-03-03T07:10:43Z
issue,95917,torchvision Caltech101 collate_fn error,https://github.com/pytorch/pytorch/issues/95917,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2023-03-03T00:00:40Z
issue,27522,[feature request] Reduction (torch.add / torch.logaddexp / torch.max / torch.min / torch.mean) of several tensors without extra copies/allocations / memory accesses } TensorList inputs support,https://github.com/pytorch/pytorch/issues/27522,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2023-03-01T18:29:00Z
issue,95786,Static asserts on accessor templates,https://github.com/pytorch/pytorch/issues/95786,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp-extensions,scan when time allows,2023-03-01T16:38:16Z
issue,95768,Inconsistent behaviour of torch.all(),https://github.com/pytorch/pytorch/issues/95768,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2023-03-01T16:34:39Z
issue,95696,Can only import torch after Tensorflow accessed its gpu device,https://github.com/pytorch/pytorch/issues/95696,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2023-03-01T16:33:51Z
issue,95648,torch needs to SHOW that it support sm_89 even if functionally the same as sm_86,https://github.com/pytorch/pytorch/issues/95648,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2023-03-01T16:32:13Z
issue,95779,SymInt'ify _gather_sparse_backward,https://github.com/pytorch/pytorch/issues/95779,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2023-03-01T13:14:37Z
issue,50201,Excessive memory usage caused by Samplers storing lists of indices,https://github.com/pytorch/pytorch/issues/50201,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dataloader, module: memory usage",scan when time allows,2023-02-28T21:50:08Z
issue,95244,Make this ridiculously long error message more user friendly,https://github.com/pytorch/pytorch/issues/95244,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: infra,scan when time allows,2023-02-28T19:42:55Z
issue,94806,ImportError: cannot import name 'Backend' from 'torch._C._distributed_c10d' (unknown location),https://github.com/pytorch/pytorch/issues/94806,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2023-02-28T19:34:23Z
issue,95309,`torch.distributed.Store` triggers INTERNAL ASSER FAILED when seting,https://github.com/pytorch/pytorch/issues/95309,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2023-02-28T19:31:13Z
issue,95434,It seems that `torch.Tensor.addmv` and `torch.Tensor.addr` will check some inputs' dtype if and only if in `backward()`,https://github.com/pytorch/pytorch/issues/95434,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: complex, module: linear algebra, module: type promotion",scan when time allows,2023-02-28T17:09:19Z
issue,95485,Enrich shape operations with nested tensors,https://github.com/pytorch/pytorch/issues/95485,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nestedtensor,scan when time allows,2023-02-28T03:19:36Z
issue,94704,`where` triggers INTERNAL ASSERT FAILED when `out` is a long tensor due to mixed types,https://github.com/pytorch/pytorch/issues/94704,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: error checking, module: type promotion",scan when time allows,2023-02-27T18:15:01Z
issue,95590,Confusing error messages from `torch.nn.LazyLinear` in different versions.,https://github.com/pytorch/pytorch/issues/95590,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: error checking, module: lazy",scan when time allows,2023-02-27T16:34:18Z
issue,95548,"torch.distributions.kumaraswamy.Kumaraswamy generates samples outside its support (0,1)",https://github.com/pytorch/pytorch/issues/95548,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: distributions, module: nans and infs",scan when time allows,2023-02-27T16:27:44Z
issue,95463,`add/add_` for CSC: errors when trying to access non-existent `crow_indices`.,https://github.com/pytorch/pytorch/issues/95463,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2023-02-27T16:25:57Z
issue,95462,Extend docs - Fixing out of memory with python garbage collection,https://github.com/pytorch/pytorch/issues/95462,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: memory usage",scan when time allows,2023-02-27T16:23:16Z
issue,95403,test_ddp_apply_optim_in_backward in distributed_test.py fails for gloo backend,https://github.com/pytorch/pytorch/issues/95403,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2023-02-27T16:18:14Z
issue,95290,Continuous dropout layer,https://github.com/pytorch/pytorch/issues/95290,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2023-02-27T12:19:26Z
issue,91653,Stochastic Illegal Memory Access error mid-epoch on AWS p4d instances,https://github.com/pytorch/pytorch/issues/91653,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cublas, module: cuda",scan when time allows,2023-02-26T22:18:46Z
issue,85621,Gloo DDP SocketTimeout error on Windows,https://github.com/pytorch/pytorch/issues/85621,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: c10d, oncall: distributed",scan when time allows,2023-02-26T12:36:06Z
issue,51112,"torch.tensor(x) fails inconsistently, assumes 0 index exists in 'Series'",https://github.com/pytorch/pytorch/issues/51112,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tensor creation,scan when time allows,2023-02-26T05:16:33Z
issue,95562,test_layer_norm_backward and test_layer_norm_backward_5d run OOM in slow gradcheck,https://github.com/pytorch/pytorch/issues/95562,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nestedtensor,scan when time allows,2023-02-26T01:33:46Z
issue,62350,NotImplementedError: Could not run 'torchvision::nms' with arguments from the 'CUDA' backend.,https://github.com/pytorch/pytorch/issues/62350,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: vision",scan when time allows,2023-02-24T00:41:05Z
issue,95276,tabulate is used by `torch.fx.graph_module.GraphModule.print_tabular` but is not installed when installing pytorch,https://github.com/pytorch/pytorch/issues/95276,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2023-02-23T22:49:48Z
issue,89421,fx.wrap is ignored with make_fx proxy tensor tracer,https://github.com/pytorch/pytorch/issues/89421,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2023-02-23T14:22:04Z
issue,95129,Default value of `validate_args` is set to `True` when passed as `None` in `Multinomial`,https://github.com/pytorch/pytorch/issues/95129,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distributions,scan when time allows,2023-02-23T01:57:40Z
issue,95058,Upsampling ResBlock GPU memory spike,https://github.com/pytorch/pytorch/issues/95058,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: memory usage",scan when time allows,2023-02-22T23:03:08Z
issue,95122,CosineAnnealingWarmRestarts but restarts are becoming more frequent,https://github.com/pytorch/pytorch/issues/95122,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: lrscheduler,scan when time allows,2023-02-22T23:01:45Z
issue,95124,"`INTERNAL ASSERT FAILED` -When using the PyTorch docker environment released by pytorch, a Vulcan support issue occurs",https://github.com/pytorch/pytorch/issues/95124,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: docker",scan when time allows,2023-02-22T22:50:17Z
issue,95135,Add local version identifier to wheel file names,https://github.com/pytorch/pytorch/issues/95135,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2023-02-22T22:48:24Z
issue,95146,Whether to consider native support for intel gpu？,https://github.com/pytorch/pytorch/issues/95146,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: intel,scan when time allows,2023-02-22T22:45:19Z
issue,95237,test_foreach failing cuda memory leak check,https://github.com/pytorch/pytorch/issues/95237,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: mta",scan when time allows,2023-02-22T22:34:38Z
issue,95077,Implement a `torch.cuda.visible_device_indexes` function.,https://github.com/pytorch/pytorch/issues/95077,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2023-02-22T22:34:14Z
issue,95304,`torch.cartesian_prod` returns inconsistent dimensions with only one input,https://github.com/pytorch/pytorch/issues/95304,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: linear algebra,scan when time allows,2023-02-22T22:33:46Z
issue,95169,COO @ COO tries to allocate way too much memory on CUDA,https://github.com/pytorch/pytorch/issues/95169,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: sparse",scan when time allows,2023-02-22T22:33:46Z
issue,32293,Truncated normal distribution,https://github.com/pytorch/pytorch/issues/32293,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distributions,scan when time allows,2023-02-21T20:22:31Z
issue,79091,Error with Named Tensors and multiple threads,https://github.com/pytorch/pytorch/issues/79091,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: named tensor,scan when time allows,2023-02-21T16:56:35Z
issue,71156,Slow backward for matrix multiplication of two sparse COO tensors on CPU,https://github.com/pytorch/pytorch/issues/71156,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2023-02-20T16:10:14Z
issue,94801,Compiling PyTorch from Source on Xavier,https://github.com/pytorch/pytorch/issues/94801,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: vectorization",scan when time allows,2023-02-19T23:59:39Z
issue,50940,SequentialSampler getting slower as time passing by,https://github.com/pytorch/pytorch/issues/50940,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2023-02-16T19:35:01Z
issue,52625,Add slicing to distributions,https://github.com/pytorch/pytorch/issues/52625,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distributions,scan when time allows,2023-02-16T17:34:56Z
issue,81899,Dependency header directory is not properly expanded in the utils.cpp_extention in ninja mode,https://github.com/pytorch/pytorch/issues/81899,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp-extensions,scan when time allows,2023-02-16T12:34:25Z
issue,93275,Bug in torch.linalg.svd,https://github.com/pytorch/pytorch/issues/93275,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: mkl,scan when time allows,2023-02-16T01:04:43Z
issue,94457,LSTM on CPU is significantly slower on PyTorch compared to other frameworks,https://github.com/pytorch/pytorch/issues/94457,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: performance",scan when time allows,2023-02-15T23:02:12Z
issue,18630,documentation for C++ / libtorch autograd profiler,https://github.com/pytorch/pytorch/issues/18630,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: docs",scan when time allows,2023-02-15T22:19:40Z
issue,94827,Linking error with Libtorch,https://github.com/pytorch/pytorch/issues/94827,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: mkl",scan when time allows,2023-02-15T16:15:01Z
issue,91536,The speed of matrix inversion is relatively slow for many small matrices,https://github.com/pytorch/pytorch/issues/91536,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: linear algebra, module: performance",scan when time allows,2023-02-15T00:59:38Z
issue,87353,Speed of torch.istft,https://github.com/pytorch/pytorch/issues/87353,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: fft, module: performance",scan when time allows,2023-02-14T03:45:04Z
issue,94654,Missing FX documents for some modules,https://github.com/pytorch/pytorch/issues/94654,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2023-02-14T03:19:57Z
issue,94705,Split getitem OpInfo into dynamic and non-dynamic inputs,https://github.com/pytorch/pytorch/issues/94705,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tests,scan when time allows,2023-02-14T03:06:53Z
issue,94739,Update PyTorch's default C standard to C17 from C11,https://github.com/pytorch/pytorch/issues/94739,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2023-02-14T02:59:25Z
issue,94591,Issue with `upsample_nearest2d` decomposition,https://github.com/pytorch/pytorch/issues/94591,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: decompositions,scan when time allows,2023-02-14T02:58:51Z
issue,44931,Issue while writing scalars o tensorboard using writer.add_scalars(....),https://github.com/pytorch/pytorch/issues/44931,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tensorboard,scan when time allows,2023-02-13T19:30:05Z
issue,94590,A Segment Fault can be triggered in torch.affine_grid_generator,https://github.com/pytorch/pytorch/issues/94590,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: edge cases,scan when time allows,2023-02-13T17:46:59Z
issue,94614,Jetson CI needs Updates,https://github.com/pytorch/pytorch/issues/94614,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: jetson,scan when time allows,2023-02-13T17:43:58Z
issue,94131,faster WeightedRandomSampler implementation based on alias method,https://github.com/pytorch/pytorch/issues/94131,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2023-02-13T15:43:33Z
issue,94388,Inconsistent results when using torch.Tensor.bernoulli with float instead of Tensor probabilities,https://github.com/pytorch/pytorch/issues/94388,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: determinism, module: distributions, module: random",scan when time allows,2023-02-10T14:42:31Z
issue,68323,sparse.mm: CUDA error: internal error when calling `cusparseSpGEMM_workEstimation [...]`,https://github.com/pytorch/pytorch/issues/68323,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: sparse",scan when time allows,2023-02-10T10:44:28Z
issue,71678,torch.bmm backward with sparse input,https://github.com/pytorch/pytorch/issues/71678,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2023-02-10T09:23:01Z
issue,27588,RuntimeError: cuDNN error: CUDNN_STATUS_MAPPING_ERROR,https://github.com/pytorch/pytorch/issues/27588,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: cudnn",scan when time allows,2023-02-10T05:07:09Z
issue,62255,Support `torch.linalg.trace`,https://github.com/pytorch/pytorch/issues/62255,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python array api,scan when time allows,2023-02-09T22:55:35Z
issue,52668,Independent axis support for torch.trace,https://github.com/pytorch/pytorch/issues/52668,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2023-02-09T22:52:19Z
issue,94451,"`jacrev` raise ""Cannot access storage of TensorWrapper"" error when computing the grad of `storage`",https://github.com/pytorch/pytorch/issues/94451,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: functorch",scan when time allows,2023-02-09T17:09:49Z
issue,94474,pybind11 SymNode binding is a footgun py::cast,https://github.com/pytorch/pytorch/issues/94474,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: pybind,scan when time allows,2023-02-09T15:22:03Z
issue,94504,[mypy] skipping mypy for a few torch/fx and torch/_subclass files,https://github.com/pytorch/pytorch/issues/94504,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: lint,scan when time allows,2023-02-09T15:11:30Z
issue,94454,Document and promise reproducibility torch.randn / torch.rand / torch.randint family behavior on CPU devices,https://github.com/pytorch/pytorch/issues/94454,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: random,scan when time allows,2023-02-09T14:48:35Z
issue,94443,A better error msg for `cuda.jiterator` when input is on `cpu`,https://github.com/pytorch/pytorch/issues/94443,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: jiterator,scan when time allows,2023-02-09T13:06:01Z
issue,94208,`zeros_like` + `fill_` makes the gradient computation in forward mode fail,https://github.com/pytorch/pytorch/issues/94208,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: forward ad,scan when time allows,2023-02-09T13:04:15Z
issue,94161,JIT: Dropout fails codegen on the third forward passes,https://github.com/pytorch/pytorch/issues/94161,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nvfuser,scan when time allows,2023-02-09T13:02:28Z
issue,94132,Asking for a LAZYMODULEMIXIN warning,https://github.com/pytorch/pytorch/issues/94132,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: lazy, module: molly-guard, module: nn",scan when time allows,2023-02-09T13:01:51Z
issue,94085,`unfold` fails in forward mode when unfolding a scalar tensor,https://github.com/pytorch/pytorch/issues/94085,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: forward ad,scan when time allows,2023-02-09T12:59:32Z
issue,49095,NCCL Error when training with 2x 3090s.,https://github.com/pytorch/pytorch/issues/49095,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: multi-gpu",scan when time allows,2023-02-09T03:35:52Z
issue,70669,CUSOLVER_STATUS_EXECUTION_FAILED when using the torch.logdet(),https://github.com/pytorch/pytorch/issues/70669,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: linear algebra",scan when time allows,2023-02-09T01:55:06Z
issue,62630,Instructions for CMAKE_PREFIX_PATH seem to be broken,https://github.com/pytorch/pytorch/issues/62630,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: docs",scan when time allows,2023-02-09T01:13:58Z
issue,68041,[RFC] APEX style fused optimizers in PyTorch,https://github.com/pytorch/pytorch/issues/68041,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: mta, module: optimizer, module: performance",scan when time allows,2023-02-08T23:03:03Z
issue,92260,distributions.Beta returning incorrect results at 0 and 1,https://github.com/pytorch/pytorch/issues/92260,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distributions,scan when time allows,2023-02-08T22:08:41Z
issue,94238,`PyTorchFileWriter` should drop the GIL while writing files,https://github.com/pytorch/pytorch/issues/94238,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: serialization,scan when time allows,2023-02-08T19:47:11Z
issue,94304,"RuntimeError: p.block != nullptr && p.block->ptr != nullptr INTERNAL ASSERT FAILED at ""../c10/cuda/CUDACachingAllocator.cpp"":1275, please report a bug to PyTorch.",https://github.com/pytorch/pytorch/issues/94304,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: assert failure, module: cudacachingallocator",scan when time allows,2023-02-08T19:46:07Z
issue,94374,[fx] const_fold.split_const_subgraphs leads to UserWarning,https://github.com/pytorch/pytorch/issues/94374,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2023-02-08T19:25:44Z
issue,49825,Ensure inplace views are properly handled,https://github.com/pytorch/pytorch/issues/49825,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2023-02-08T16:26:51Z
issue,72408,Matrix multiplication is 30 times slower for integers than floats on CPU,https://github.com/pytorch/pytorch/issues/72408,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: linear algebra, module: performance",scan when time allows,2023-02-08T09:17:13Z
issue,69063,make: Makefile: No such file or directory,https://github.com/pytorch/pytorch/issues/69063,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2023-02-08T02:10:25Z
issue,61281,DataLoader with IterativeDataset throws an error when providing a BatchSampler,https://github.com/pytorch/pytorch/issues/61281,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2023-02-07T23:41:12Z
issue,93240,USE_CUDNN=1 doesn't force cmake to fail if cudnn is not found,https://github.com/pytorch/pytorch/issues/93240,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: build warnings, module: cuda",scan when time allows,2023-02-07T22:44:19Z
issue,94003,test_nccl_warn_not_in_group_debug_detail is flaky,https://github.com/pytorch/pytorch/issues/94003,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: flaky-tests, oncall: distributed",scan when time allows,2023-02-07T17:17:25Z
issue,85157,NestedTensor 2.0 issue tracking,https://github.com/pytorch/pytorch/issues/85157,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nestedtensor,scan when time allows,2023-02-07T16:12:57Z
issue,94185,Addition of CSC/BSR/BSC tensors raises RuntimeError exceptions,https://github.com/pytorch/pytorch/issues/94185,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2023-02-07T15:03:35Z
issue,50036,channels_last format convolution is slower than normal NCHW,https://github.com/pytorch/pytorch/issues/50036,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: memory format, module: performance",scan when time allows,2023-02-07T14:56:51Z
issue,17745,"distributed data parallel, gloo backend works, but nccl deadlock",https://github.com/pytorch/pytorch/issues/17745,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: deadlock, module: nccl, oncall: distributed",scan when time allows,2023-02-07T06:22:45Z
issue,94021,Set AVX2 is minimum supported instruction set for Linux X86,https://github.com/pytorch/pytorch/issues/94021,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: intel,scan when time allows,2023-02-06T18:24:26Z
issue,93944,error: no member named 'residual_with_sum_zero_point' in 'ideep::attr_t,https://github.com/pytorch/pytorch/issues/93944,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: macos",scan when time allows,2023-02-06T15:00:57Z
issue,93347,"when I want to use a new backend, how to deal with the op with 'device' argument?",https://github.com/pytorch/pytorch/issues/93347,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: backend,scan when time allows,2023-02-06T14:44:50Z
issue,90547,Unable to link LibTorch against CUDA and CUDNN statically,https://github.com/pytorch/pytorch/issues/90547,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: static linking,scan when time allows,2023-02-06T09:43:32Z
issue,93955,"Enable Link Time Optimization in PyTorch 2.0 Release Binaries - Smaller, Faster, Better Binaries",https://github.com/pytorch/pytorch/issues/93955,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: build, module: performance, oncall: releng",scan when time allows,2023-02-05T20:26:57Z
issue,93523,[functorch] [vmap] [SymInt][fake tensor],https://github.com/pytorch/pytorch/issues/93523,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: functorch,scan when time allows,2023-02-03T17:29:20Z
issue,38487,expected scalar type Half but found Float with torch.cuda.amp and torch.nn.DataParallel,https://github.com/pytorch/pytorch/issues/38487,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision), module: data parallel",scan when time allows,2023-02-03T12:35:00Z
issue,92884,"[RFC] Make more operations inplace (GELU, BatchNorm, LayerNorm)",https://github.com/pytorch/pytorch/issues/92884,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2023-02-02T23:52:57Z
issue,62395,[FX] Does PyTorch want to support for dynamic control flow for torch.fx?,https://github.com/pytorch/pytorch/issues/62395,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2023-02-02T22:50:34Z
issue,93319,Minifier should not use pickle to save state into minifier launcher,https://github.com/pytorch/pytorch/issues/93319,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: functorch, module: fx",scan when time allows,2023-02-01T15:09:19Z
issue,93227,squeezenet1_1 fails accuracy with AMP (but not on CI and dashboard); minifier does not work (when not using cuDNN?),https://github.com/pytorch/pytorch/issues/93227,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision),scan when time allows,2023-02-01T02:40:32Z
issue,93579,torchdynamo.export doesn't work with data-dependent control flow,https://github.com/pytorch/pytorch/issues/93579,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2023-02-01T01:06:34Z
issue,93458,minified code can not produce fp64_ref result,https://github.com/pytorch/pytorch/issues/93458,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2023-02-01T01:02:09Z
issue,93446,[Inductor] [CPU] Vectorization not supporting python pass-in scalar double in speech_transformer,https://github.com/pytorch/pytorch/issues/93446,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2023-02-01T01:01:55Z
issue,93444,[Inductor] [CPU] Maxpooling is not vectorized in shufflenet_v2_x1_0,https://github.com/pytorch/pytorch/issues/93444,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2023-02-01T01:01:53Z
issue,62168,Error installing from source,https://github.com/pytorch/pytorch/issues/62168,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2023-01-31T22:06:02Z
issue,93231,"Minifier produces minifier script that doesn't fail accuracy on Background_Matting (dynamic shapes, inductor, inference)",https://github.com/pytorch/pytorch/issues/93231,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: functorch, module: fx",scan when time allows,2023-01-31T21:39:24Z
issue,93317,Minifier doesn't save/load functorch config,https://github.com/pytorch/pytorch/issues/93317,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: functorch, module: fx",scan when time allows,2023-01-31T21:16:10Z
issue,62998,Add nn.Module full pre-backward hooks,https://github.com/pytorch/pytorch/issues/62998,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2023-01-31T19:10:57Z
issue,89597,Caching a model's weights and state_dict to disk to save RAM,https://github.com/pytorch/pytorch/issues/89597,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2023-01-31T16:38:15Z
issue,93362,Make torch.testing functions overrideable with torch_function?,https://github.com/pytorch/pytorch/issues/93362,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: __torch_function__, module: testing",scan when time allows,2023-01-31T16:25:39Z
issue,72880,Bazel fails in an obscure way if submodules are not initialized,https://github.com/pytorch/pytorch/issues/72880,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bazel, module: third_party",scan when time allows,2023-01-31T11:14:42Z
issue,91553,Segmentation fault when running torch.nn.AdaptiveMaxPool2d,https://github.com/pytorch/pytorch/issues/91553,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: edge cases,scan when time allows,2023-01-30T21:14:00Z
issue,91552,Overflow when running torch.nn.AdaptiveMaxPool2d,https://github.com/pytorch/pytorch/issues/91552,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: edge cases, module: pooling",scan when time allows,2023-01-30T21:13:38Z
issue,91940,torchrun elastic always “address already in use” error,https://github.com/pytorch/pytorch/issues/91940,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: oncall: distributed, oncall: r2p",scan when time allows,2023-01-30T02:41:25Z
issue,91697,Build from source fails: undefined reference to caffe2::DeviceQuery,https://github.com/pytorch/pytorch/issues/91697,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2023-01-29T10:20:21Z
issue,91943,elastic job failed when scale down,https://github.com/pytorch/pytorch/issues/91943,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: oncall: distributed, oncall: r2p",scan when time allows,2023-01-28T06:40:30Z
issue,90793,nn.CrossEntropyLoss error out when the sample size is large,https://github.com/pytorch/pytorch/issues/90793,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: edge cases, module: memory usage, module: nn",scan when time allows,2023-01-27T17:29:46Z
issue,92778,Process get killed when running torch.normal,https://github.com/pytorch/pytorch/issues/92778,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: numpy,scan when time allows,2023-01-27T07:02:32Z
issue,93017,numpy v1.24 does not work with `writer.add_histogram`,https://github.com/pytorch/pytorch/issues/93017,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tensorboard,scan when time allows,2023-01-26T23:48:17Z
issue,91325,Support for saving multiple storages/tensors that view same data as different dtypes,https://github.com/pytorch/pytorch/issues/91325,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: serialization,scan when time allows,2023-01-26T19:26:11Z
issue,24081,Multi-gpu example freeze and is not killable,https://github.com/pytorch/pytorch/issues/24081,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: data parallel, module: deadlock, module: dependency bug",scan when time allows,2023-01-26T17:16:55Z
issue,53937,Only tensor can be proxied in builtin operators with `__torch_function__`,https://github.com/pytorch/pytorch/issues/53937,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: __torch_function__, module: fx",scan when time allows,2023-01-26T13:23:47Z
issue,92967,Tracking issue for segfaults and floating point exceptions on 1.12.0,https://github.com/pytorch/pytorch/issues/92967,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2023-01-26T06:52:07Z
issue,84309,Unable to run a single convolutional layer in different CUDA-contexts,https://github.com/pytorch/pytorch/issues/84309,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2023-01-25T21:26:07Z
issue,66482,BAR1 memory of GPU is not released when main process is killed.,https://github.com/pytorch/pytorch/issues/66482,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: dataloader, module: memory usage",scan when time allows,2023-01-25T17:19:39Z
issue,83388,"zero-numel tensor has ""RuntimeError: strides[cur - 1] == sizes[cur] * strides[cur] INTERNAL ASSERT FAILED"" in multi-thread.",https://github.com/pytorch/pytorch/issues/83388,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nvfuser, oncall: jit",scan when time allows,2023-01-25T16:13:34Z
issue,92907,Unable to find an engine to execute when using pip to install but not with conda,https://github.com/pytorch/pytorch/issues/92907,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: convolution, module: cudnn",scan when time allows,2023-01-25T16:06:09Z
issue,83376,Complex-Valued Gaussian distributions,https://github.com/pytorch/pytorch/issues/83376,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: complex, module: distributions, module: random",scan when time allows,2023-01-25T11:52:16Z
issue,92920,Add Stride Argument For Constructors,https://github.com/pytorch/pytorch/issues/92920,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tensor creation,scan when time allows,2023-01-25T07:30:37Z
issue,92781,Floating point exception when running torch.nn.AdaptiveMaxPool3d,https://github.com/pytorch/pytorch/issues/92781,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2023-01-25T07:24:21Z
issue,57423,internal compiler error compiling PyTorch master on Fedora 33,https://github.com/pytorch/pytorch/issues/57423,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2023-01-25T06:56:33Z
issue,92835,Double free when running torch.linalg.ldl_solve,https://github.com/pytorch/pytorch/issues/92835,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: edge cases,scan when time allows,2023-01-25T06:27:01Z
issue,92828,segfault when running torch.igamma,https://github.com/pytorch/pytorch/issues/92828,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2023-01-25T06:14:40Z
issue,35071,Options for printing the shape with print(tensor),https://github.com/pytorch/pytorch/issues/35071,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: printing,scan when time allows,2023-01-25T01:56:59Z
issue,63855,Include tensor shape in its default print formatting,https://github.com/pytorch/pytorch/issues/63855,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: printing,scan when time allows,2023-01-25T01:54:57Z
issue,71210,axis to dim remapping is not working for flip and roll,https://github.com/pytorch/pytorch/issues/71210,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python array api,scan when time allows,2023-01-24T21:36:09Z
issue,92801,"[BE] move _apply_to_tensors from FSDP to torch.distributed.utils, use in _recursive_to",https://github.com/pytorch/pytorch/issues/92801,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ddp, oncall: distributed",scan when time allows,2023-01-24T18:09:45Z
issue,78880,fatal_signal_asan_no_sig_test in current master hang.,https://github.com/pytorch/pytorch/issues/78880,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: deadlock, module: testing",scan when time allows,2023-01-24T14:06:50Z
issue,72393,Some system-installed headers are mistakenly used.,https://github.com/pytorch/pytorch/issues/72393,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2023-01-24T13:58:26Z
issue,92252,"`model.to(""cuda:0"")` does not release all CPU memory",https://github.com/pytorch/pytorch/issues/92252,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: memory usage,scan when time allows,2023-01-23T21:58:35Z
issue,91556,Overflow when running torch.nn.AdaptiveMaxPool3d on torch 1.12.0 and 1.13.1,https://github.com/pytorch/pytorch/issues/91556,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: edge cases,scan when time allows,2023-01-23T17:17:51Z
issue,92811,[complex] Jacobian of a non-holomorphic complex valued function,https://github.com/pytorch/pytorch/issues/92811,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2023-01-23T15:49:17Z
issue,28733,CPU MaxPool2d is very slow,https://github.com/pytorch/pytorch/issues/28733,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, module: cpu, module: performance, module: pooling",scan when time allows,2023-01-23T02:08:41Z
issue,92683,Torchscript troubles with complex values. RuntimeError: isInt() INTERNAL ASSERT FAILED,https://github.com/pytorch/pytorch/issues/92683,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: complex, oncall: jit",scan when time allows,2023-01-21T03:39:03Z
issue,92542,Pytorch AMP performance issue.,https://github.com/pytorch/pytorch/issues/92542,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision), module: memory format",scan when time allows,2023-01-20T01:50:37Z
issue,92594,INTERNAL ASSERT FAILED when mixed dtypes for `addcmul_`,https://github.com/pytorch/pytorch/issues/92594,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: assert failure, module: type promotion",scan when time allows,2023-01-20T01:32:14Z
issue,92173,"""multi device"" tests get skipped in standard CI",https://github.com/pytorch/pytorch/issues/92173,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ci, module: tests",scan when time allows,2023-01-19T22:32:32Z
issue,89160,Got many TestDTensorOpsCUDA.test_dtensor_op_db_X test failures,https://github.com/pytorch/pytorch/issues/89160,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2023-01-19T16:22:43Z
issue,49038,Make the forward function of a nn.Module and/or a certain function to work with only float32 when used with autocast,https://github.com/pytorch/pytorch/issues/49038,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision), module: nn",scan when time allows,2023-01-19T01:14:44Z
issue,92223,Improve make_fx tracing speed,https://github.com/pytorch/pytorch/issues/92223,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: performance, module: proxytensor",scan when time allows,2023-01-18T15:11:13Z
issue,92217,"false INTERNAL ASSERT FAILED at ""../c10/cuda/CUDAGraphsC10Utils.h"":73, please report a bug to PyTorch. Unknown CUDA graph CaptureStatus32680",https://github.com/pytorch/pytorch/issues/92217,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda graphs,scan when time allows,2023-01-18T13:49:37Z
issue,92350,TypeError: no implementation found for 'torch._ops.aten.max.default' on types that implement __torch_dispatch__: [<class 'torch.masked.maskedtensor.core.MaskedTensor'>],https://github.com/pytorch/pytorch/issues/92350,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: masked operators,scan when time allows,2023-01-18T12:48:39Z
issue,92259,[discussion] Fused MLPs,https://github.com/pytorch/pytorch/issues/92259,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",scan when time allows,2023-01-18T12:48:20Z
issue,71301,Why AMP make backward speed more slow?,https://github.com/pytorch/pytorch/issues/71301,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2023-01-18T12:47:26Z
issue,64535,Memory Leak in MKL OpenMP on AVX2 machine,https://github.com/pytorch/pytorch/issues/64535,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: intel, module: memory usage, module: mkl",scan when time allows,2023-01-18T08:12:33Z
issue,91810,[Bug/functorch] Cannot use `tensor.detach().numpy()` for `GradTrackingTensor`: Cannot access data pointer of Tensor that doesn't have storage,https://github.com/pytorch/pytorch/issues/91810,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: functorch,scan when time allows,2023-01-17T20:23:32Z
issue,91856,torchrun --help is too slow,https://github.com/pytorch/pytorch/issues/91856,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: oncall: distributed, oncall: r2p",scan when time allows,2023-01-17T18:15:49Z
issue,91855,torchrun default value of command line options,https://github.com/pytorch/pytorch/issues/91855,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: oncall: distributed, oncall: r2p",scan when time allows,2023-01-17T18:15:30Z
issue,91889,ToTensor deadlock in subprocess,https://github.com/pytorch/pytorch/issues/91889,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: multiprocessing,scan when time allows,2023-01-16T10:36:11Z
issue,92086,Error while building pytorch mobile binaries from source,https://github.com/pytorch/pytorch/issues/92086,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: mobile,scan when time allows,2023-01-16T09:37:45Z
issue,46374,Complex Number support for torch.nn modules,https://github.com/pytorch/pytorch/issues/46374,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: complex, module: nn",scan when time allows,2023-01-15T20:01:33Z
issue,85656,"[functorch] conv.{1, 2, 3}d should raise errors",https://github.com/pytorch/pytorch/issues/85656,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: functorch,scan when time allows,2023-01-15T08:28:01Z
issue,89105,Bad string in GLSL shader,https://github.com/pytorch/pytorch/issues/89105,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: vulkan,scan when time allows,2023-01-15T04:06:03Z
issue,92148,Occassional OverflowError with mps running yolov7,https://github.com/pytorch/pytorch/issues/92148,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: mps,scan when time allows,2023-01-13T18:34:59Z
issue,90147,[Feature Proposal] Extend torch hub to better support cloud serving and edge deployment,https://github.com/pytorch/pytorch/issues/90147,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: hub,scan when time allows,2023-01-13T11:33:50Z
issue,91716,linspace (and arange) behaves differently on GPU and CPU,https://github.com/pytorch/pytorch/issues/91716,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2023-01-13T08:11:00Z
issue,53103,ModuleList not recognised as reversible by mypy,https://github.com/pytorch/pytorch/issues/53103,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: typing",scan when time allows,2023-01-12T23:03:08Z
issue,92075,"Hessian produces wrong results, but works if I add a perturbation",https://github.com/pytorch/pytorch/issues/92075,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2023-01-12T20:31:20Z
issue,92072,Implement forward AD with grid_sampler_2d,https://github.com/pytorch/pytorch/issues/92072,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: forward ad,scan when time allows,2023-01-12T19:33:23Z
issue,73563,Pytorch Installation from source fails,https://github.com/pytorch/pytorch/issues/73563,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2023-01-12T01:44:07Z
issue,56862,Non-symbolic FX tracer,https://github.com/pytorch/pytorch/issues/56862,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2023-01-11T14:26:01Z
issue,58736,type promotion with 0d-tensors diverges from array API specification,https://github.com/pytorch/pytorch/issues/58736,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: python array api, module: type promotion",scan when time allows,2023-01-11T08:28:12Z
issue,91958,CUDA error `CUBLAS_STATUS_NOT_INITIALIZED`,https://github.com/pytorch/pytorch/issues/91958,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cublas, module: cuda",scan when time allows,2023-01-11T06:42:18Z
issue,91809,Better API for `torch.cov` (and `Tensor.cov`),https://github.com/pytorch/pytorch/issues/91809,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python frontend,scan when time allows,2023-01-10T14:20:16Z
issue,91842,Adding a page for subfolder/subfile overview/descriptions in the developer wiki,https://github.com/pytorch/pytorch/issues/91842,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2023-01-10T14:10:14Z
issue,78489,[1.9.1] [collect_env] collect_env does not collect actual runtime-loaded cudnn version,https://github.com/pytorch/pytorch/issues/78489,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: collect_env.py, module: cudnn",scan when time allows,2023-01-09T20:42:42Z
issue,91887,Only the first logged trace in a given log dir is visible in tensorboard.,https://github.com/pytorch/pytorch/issues/91887,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tensorboard,scan when time allows,2023-01-09T18:59:13Z
issue,91300,[🚀 Feature Request] pdf and sampling from Alpha-stable distribution,https://github.com/pytorch/pytorch/issues/91300,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distributions,scan when time allows,2023-01-09T18:18:54Z
issue,82076,Autocast documentation examples would break,https://github.com/pytorch/pytorch/issues/82076,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision), module: docs",scan when time allows,2023-01-08T17:11:35Z
issue,91759,"Error while building pytorch from source on windows - Ninja Build Stopped, Subcommand Failed",https://github.com/pytorch/pytorch/issues/91759,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2023-01-06T04:23:04Z
issue,91678,Wrong in building torch from source,https://github.com/pytorch/pytorch/issues/91678,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2023-01-06T04:20:44Z
issue,91624,trainer,https://github.com/pytorch/pytorch/issues/91624,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2023-01-06T04:18:27Z
issue,91751,[bazel] replace //c10:headers dependency by //c10 dependency,https://github.com/pytorch/pytorch/issues/91751,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: bazel,scan when time allows,2023-01-06T03:29:36Z
issue,87861,RuntimeError: unable to mmap 29764 bytes from file </torch_10182_3020184674_63991>: Cannot allocate memory (12),https://github.com/pytorch/pytorch/issues/87861,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tensor creation,scan when time allows,2023-01-05T13:22:31Z
issue,91615,"sparse.mm(coo, dense) produces wrong results on T4/V100 GPUs",https://github.com/pytorch/pytorch/issues/91615,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2023-01-04T18:19:10Z
issue,91593,Division by zero error when running torch.nn.functional.lp_pool1d,https://github.com/pytorch/pytorch/issues/91593,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: edge cases,scan when time allows,2023-01-04T16:14:11Z
issue,91618,There is no developer documentation about getting started with MPS native debugging,https://github.com/pytorch/pytorch/issues/91618,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: mps",scan when time allows,2023-01-04T15:33:48Z
issue,91599,"`mul(CSC, CSC)` fails with layout mismatch between the inputs and the output.",https://github.com/pytorch/pytorch/issues/91599,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2023-01-04T15:28:14Z
issue,62672,Add Convolution support for lazy tensor,https://github.com/pytorch/pytorch/issues/62672,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: lazy",scan when time allows,2023-01-03T22:20:34Z
issue,68058,Some lr-schedulers docs seems to have typos/missing information,https://github.com/pytorch/pytorch/issues/68058,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: lrscheduler, module: optimizer",scan when time allows,2023-01-03T18:52:59Z
issue,91574,Add BlockWise Distribution Support to the torch.distributions Package,https://github.com/pytorch/pytorch/issues/91574,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distributions,scan when time allows,2023-01-03T15:22:26Z
issue,91437,using Tensor subclass between vmap layers,https://github.com/pytorch/pytorch/issues/91437,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: __torch_dispatch__, module: functorch",scan when time allows,2023-01-03T15:21:18Z
issue,91447,"Degenerate ranges are allowed in NumPy, but not in PyTorch.",https://github.com/pytorch/pytorch/issues/91447,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: numpy,scan when time allows,2023-01-03T10:11:33Z
issue,91582,Softmax function slows down for data with large range,https://github.com/pytorch/pytorch/issues/91582,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: performance",scan when time allows,2023-01-03T08:48:26Z
issue,91557,Segmentation fault when running torch.nn.AdaptiveMaxPool3d,https://github.com/pytorch/pytorch/issues/91557,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: edge cases,scan when time allows,2023-01-03T02:04:16Z
issue,73175,[feature request] Support tensor count vector argument in torch.split,https://github.com/pytorch/pytorch/issues/73175,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: bootcamp,scan when time allows,2023-01-02T22:34:29Z
issue,71329,"[RFC] Implement array methods (extend, insert, pop, etc) for container classes",https://github.com/pytorch/pytorch/issues/71329,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2023-01-02T06:10:27Z
issue,41327,torch.nn.parallel.scatter_gather.gather can't gather outputs that are dataclasses,https://github.com/pytorch/pytorch/issues/41327,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: data parallel, oncall: distributed",scan when time allows,2022-12-31T01:24:15Z
issue,70394,Intel MKL FATAL ERROR: This system does not meet the minimum requirements for use of the Intel(R) Math Kernel Library.,https://github.com/pytorch/pytorch/issues/70394,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: dependency bug, module: intel, module: macos",scan when time allows,2022-12-30T12:56:39Z
issue,91509,[functorch] make batch norm docs point to UX limitations,https://github.com/pytorch/pytorch/issues/91509,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: functorch,scan when time allows,2022-12-29T16:31:52Z
issue,91471,Clean up nt impl duplicates where one can,https://github.com/pytorch/pytorch/issues/91471,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nestedtensor,scan when time allows,2022-12-28T22:22:34Z
issue,19117,[jit] Can't `torch.jit.script` a lambda,https://github.com/pytorch/pytorch/issues/19117,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2022-12-28T20:14:59Z
issue,90412,Importing numpy makes Tensor min max crash,https://github.com/pytorch/pytorch/issues/90412,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: numpy,scan when time allows,2022-12-28T19:42:23Z
issue,91417,[bazel] error: use of undeclared identifier 'cudaGraphDebugDotPrint',https://github.com/pytorch/pytorch/issues/91417,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cuda",scan when time allows,2022-12-28T14:00:21Z
issue,89344,AdaptiveAvgPool1d failed in the lower version,https://github.com/pytorch/pytorch/issues/89344,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: pooling",scan when time allows,2022-12-28T13:29:14Z
issue,90481,Internal assert when ctx.saved_tensors fails when saving results of an intermediate view tensor with torch.utils.checkpoint and use_reentrant=False,https://github.com/pytorch/pytorch/issues/90481,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: activation checkpointing, module: autograd",scan when time allows,2022-12-28T10:17:37Z
issue,91406,Pytorch clang-tidy header-filter is still broken,https://github.com/pytorch/pytorch/issues/91406,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2022-12-28T05:02:28Z
issue,91374,[RFC] `quantile` should work for `float16`/`half` on the GPU,https://github.com/pytorch/pytorch/issues/91374,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: half",scan when time allows,2022-12-27T23:27:32Z
issue,91375,Internal Assert failed,https://github.com/pytorch/pytorch/issues/91375,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: cuda graphs",scan when time allows,2022-12-27T23:26:59Z
issue,91039,Simple deleting from the sys cache fails on reimport,https://github.com/pytorch/pytorch/issues/91039,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python frontend,scan when time allows,2022-12-26T20:12:58Z
issue,55437,Error building docker image: No module named 'typing_extensions',https://github.com/pytorch/pytorch/issues/55437,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: docker",scan when time allows,2022-12-25T14:14:10Z
issue,52295,Printing a Tensor that is being vmap'ed over in C++ raises an error,https://github.com/pytorch/pytorch/issues/52295,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: vmap,scan when time allows,2022-12-24T10:07:37Z
issue,38536,"build libtorch problem: Configuring incomplete, errors occurred! _mm256_abs_epi16",https://github.com/pytorch/pytorch/issues/38536,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2022-12-24T08:59:38Z
issue,78475,`with torch.backends.cudnn.flags(deterministic=True)` doesn't give an exception for ctc_loss backward on CUDA,https://github.com/pytorch/pytorch/issues/78475,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cudnn, module: determinism",scan when time allows,2022-12-23T03:36:58Z
issue,91156,`quantile` fails for `float16`/`half` inputs,https://github.com/pytorch/pytorch/issues/91156,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: half",scan when time allows,2022-12-23T00:45:43Z
issue,91280,custom Function that supports functorch jvp doesn't work with in-place,https://github.com/pytorch/pytorch/issues/91280,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: forward ad, module: functorch",scan when time allows,2022-12-22T21:45:50Z
issue,89907,nn.CrossEntropy/nn.NLLLoss : Request for option to specify invalid ignore_index for perf. optimization,https://github.com/pytorch/pytorch/issues/89907,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: loss,scan when time allows,2022-12-22T17:30:57Z
issue,91199,functorch.functionalize doesn't work with torch.autograd.grad,https://github.com/pytorch/pytorch/issues/91199,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: functorch,scan when time allows,2022-12-22T03:18:22Z
issue,46409,Custom ops get stuck in multiprocess data loader under certain environments,https://github.com/pytorch/pytorch/issues/46409,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: custom-operators, module: dataloader, module: deadlock",scan when time allows,2022-12-21T23:01:01Z
issue,91244,first class dims leak memeory,https://github.com/pytorch/pytorch/issues/91244,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: functorch,scan when time allows,2022-12-21T13:38:21Z
issue,70505,`UninitializedParameter.to(device='meta')` creates a zero-sized meta tensor instead of remaining uninitialized,https://github.com/pytorch/pytorch/issues/70505,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: lazy, module: meta tensors",scan when time allows,2022-12-21T05:59:23Z
issue,91210,Odd/hand-wavy mathematical notation for Conv2D,https://github.com/pytorch/pytorch/issues/91210,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2022-12-21T04:57:09Z
issue,89068,view_copy out= does not reshape zero element tensors,https://github.com/pytorch/pytorch/issues/89068,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: functionalization, module: viewing and reshaping",scan when time allows,2022-12-20T23:16:33Z
issue,90999,overflow (?) on cuda tensor after matrix multiplication,https://github.com/pytorch/pytorch/issues/90999,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: numerical-stability",scan when time allows,2022-12-20T18:01:39Z
issue,90526,valgrind failure `Conditional jump or move depends on uninitialised value(s)`,https://github.com/pytorch/pytorch/issues/90526,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sanitizers,scan when time allows,2022-12-20T08:56:47Z
issue,90905,Retrieve Tensor from Tensor.data_ptr(),https://github.com/pytorch/pytorch/issues/90905,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-12-20T08:54:10Z
issue,90663,[Feature Request] An alternative sampling routine for Dirichlet to fix Dirichlet and Beta sampling bugs,https://github.com/pytorch/pytorch/issues/90663,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distributions,scan when time allows,2022-12-17T12:15:18Z
issue,90992,as_strided_scatter : INTERNAL_ASSERT_FAILED for requires_grad=True and non-config input,https://github.com/pytorch/pytorch/issues/90992,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: functionalization,scan when time allows,2022-12-16T15:31:12Z
issue,90555,[RFC] Add torch.backends.tbb.is_available(),https://github.com/pytorch/pytorch/issues/90555,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tbb,scan when time allows,2022-12-15T23:49:10Z
issue,90895,ModuleNotFoundError: No module named 'torch._C._distributed_c10d'; 'torch._C' is not a package,https://github.com/pytorch/pytorch/issues/90895,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2022-12-15T22:21:04Z
issue,90832,Inference Mode docs,https://github.com/pytorch/pytorch/issues/90832,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2022-12-15T04:01:05Z
issue,68377,Misleading Error Message from nn.RNN when Passing Incorrect Data Type,https://github.com/pytorch/pytorch/issues/68377,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rnn,scan when time allows,2022-12-15T00:40:10Z
issue,90708,PixelShuffle/Unshuffle Channels Last Support on GPU,https://github.com/pytorch/pytorch/issues/90708,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: memory format,scan when time allows,2022-12-15T00:37:49Z
issue,90751,pytorch prune in libtorch,https://github.com/pytorch/pytorch/issues/90751,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp,scan when time allows,2022-12-13T16:10:16Z
issue,33354,Long torchscript warmup times can be problematic for production serving,https://github.com/pytorch/pytorch/issues/33354,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2022-12-13T10:23:19Z
issue,87448,Semantics of sparse operations clarification - Sparsity of the gradient with respect to a sparse tensor input,https://github.com/pytorch/pytorch/issues/87448,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: docs, module: masked operators, module: sparse",scan when time allows,2022-12-13T05:41:07Z
issue,87956,Autograd doesn't stop executing backward graph early enough in situations involving set_,https://github.com/pytorch/pytorch/issues/87956,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2022-12-13T05:38:55Z
issue,87755,Autograd precision for CONV + BN between pytorch version 1.11.0 and 1.12.1,https://github.com/pytorch/pytorch/issues/87755,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: numerical-reproducibility, module: numerical-stability",scan when time allows,2022-12-13T05:31:32Z
issue,90742,Adopt full_backward_pre_hook in DDP,https://github.com/pytorch/pytorch/issues/90742,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ddp, oncall: distributed",scan when time allows,2022-12-13T03:12:00Z
issue,52040,torch.distributed.TCPStore doesn't work with dual IPv4/IPv6 network interface,https://github.com/pytorch/pytorch/issues/52040,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2022-12-13T00:08:11Z
issue,87992,torch.rand(...) is not consistent for large shape dimensions across GPUs (with the same random seed),https://github.com/pytorch/pytorch/issues/87992,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: random",scan when time allows,2022-12-12T11:02:00Z
issue,43250,Support for Multi-Categorical in torch.distributions,https://github.com/pytorch/pytorch/issues/43250,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distributions,scan when time allows,2022-12-11T12:08:02Z
issue,36706,Reduction for `torch.int8` is super slow on CUDA,https://github.com/pytorch/pytorch/issues/36706,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: performance, module: tensoriterator",scan when time allows,2022-12-10T15:36:16Z
issue,90398,IValue(c10::List<IValue>) constructor is confusing and undocumented,https://github.com/pytorch/pytorch/issues/90398,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: internals,scan when time allows,2022-12-09T19:35:39Z
issue,90447,Different behavior for complex numbers operations with numpy,https://github.com/pytorch/pytorch/issues/90447,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: complex, module: nans and infs, module: numpy",scan when time allows,2022-12-09T16:22:48Z
issue,90485,Tensor indexing and slicing documentation should explicitly state that indexing follows numpy semantics and link to the numpy indexing documentation.,https://github.com/pytorch/pytorch/issues/90485,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: numpy",scan when time allows,2022-12-09T16:09:19Z
issue,90057,[GradScaler] Inconsistent scale values across different GPUs caused by uneven inputs for AMP DDP training,https://github.com/pytorch/pytorch/issues/90057,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision),scan when time allows,2022-12-09T12:35:04Z
issue,88265,build: failure when upgrade oneTBB to 2021.7.0,https://github.com/pytorch/pytorch/issues/88265,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: tbb",scan when time allows,2022-12-09T03:01:59Z
issue,90065,forward-mode AD formula for torch.add (and possibly others) accidentally upcasts float32 to float64,https://github.com/pytorch/pytorch/issues/90065,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: forward ad, module: functorch, module: performance",scan when time allows,2022-12-08T21:34:01Z
issue,89408,torch.nn.MultiLabelMarginLoss has different performance on CPU and GPU.,https://github.com/pytorch/pytorch/issues/89408,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: loss,scan when time allows,2022-12-08T13:21:26Z
issue,51720,Support for 64-bit (ILP64) LAPACK,https://github.com/pytorch/pytorch/issues/51720,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: linear algebra",scan when time allows,2022-12-08T13:07:12Z
issue,90395,"Cannot add target-level dependencies to non-existent target ""gloo_cuda"".",https://github.com/pytorch/pytorch/issues/90395,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2022-12-08T01:13:49Z
issue,90256,LibTorch static build from source missing libshm.so,https://github.com/pytorch/pytorch/issues/90256,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2022-12-07T02:47:52Z
issue,19349,Different behavior when trace model.,https://github.com/pytorch/pytorch/issues/19349,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2022-12-07T01:12:16Z
issue,71806,Migrate master to main: https://github.com/pytorch/pytorch,https://github.com/pytorch/pytorch/issues/71806,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: infra,scan when time allows,2022-12-06T19:20:05Z
issue,62540,[Meta] Change default branch name to `main` for repos in `pytorch` project,https://github.com/pytorch/pytorch/issues/62540,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, module: infra",scan when time allows,2022-12-06T19:19:53Z
issue,90126,torch._dynamo.exc.BackendCompilerFailed: compile_fx raised TypeError: tqdm.__init__() got an unexpected keyword argument 'desc',https://github.com/pytorch/pytorch/issues/90126,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: hub,scan when time allows,2022-12-06T18:11:09Z
issue,90114,documentation need to be as per python version,https://github.com/pytorch/pytorch/issues/90114,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2022-12-06T18:00:03Z
issue,87090,MPS memory usage significantly higher than on CPU,https://github.com/pytorch/pytorch/issues/87090,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: memory usage, module: mps",scan when time allows,2022-12-06T12:04:52Z
issue,68916,Crash loading fairseq based model,https://github.com/pytorch/pytorch/issues/68916,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: hub,scan when time allows,2022-12-05T23:37:07Z
issue,84944,test_warp_softmax_64bit_indexing_cuda_float16 takes ~147GB of CPU memory and is very slow,https://github.com/pytorch/pytorch/issues/84944,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: memory usage, module: testing",scan when time allows,2022-12-05T20:02:34Z
issue,47908,The speed of pytorch with cudatoolkit 11.0 is slower than cudatoolkit 10.2,https://github.com/pytorch/pytorch/issues/47908,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: performance",scan when time allows,2022-12-05T17:21:30Z
issue,76105,Failed to build on Ubuntu 18.04 due to bad MPI linker flags,https://github.com/pytorch/pytorch/issues/76105,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2022-12-05T15:48:10Z
issue,89942,Random K compression hook in PyTorch DDP,https://github.com/pytorch/pytorch/issues/89942,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ddp,scan when time allows,2022-12-05T04:32:41Z
issue,76069,`init_process_group` hanging on HPC multi-node system w GPU,https://github.com/pytorch/pytorch/issues/76069,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2022-12-04T16:47:37Z
issue,62661,Rewrite PyTorch CUDA backend in Triton,https://github.com/pytorch/pytorch/issues/62661,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2022-12-02T21:11:17Z
issue,85485,"`max_pool2d_with_indices(self, ...)` shouldn't need to save `self` for backward",https://github.com/pytorch/pytorch/issues/85485,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2022-12-02T15:47:39Z
issue,61167,Torch .pth model export to ONNX returned empty graph,https://github.com/pytorch/pytorch/issues/61167,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2022-12-02T11:13:15Z
issue,55967,"Cannot init, destroy, and then re-init process groups",https://github.com/pytorch/pytorch/issues/55967,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2022-12-02T00:01:18Z
issue,90000,"Can not use x=torch.tensor(b), to create a Tensor out of a List[List[Tensor]] (A List of Lists of Tensors)",https://github.com/pytorch/pytorch/issues/90000,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2022-12-01T22:28:59Z
issue,89609,NVFuser failing masked.{amax|amin|sum} extremal and correctness tests,https://github.com/pytorch/pytorch/issues/89609,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nvfuser,scan when time allows,2022-12-01T08:37:16Z
issue,72138,[RFC] PyTorch Sharder for distributed training,https://github.com/pytorch/pytorch/issues/72138,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2022-11-30T17:36:03Z
issue,26781,Matrix corresponding to convolution by a 2D kernel (convmtx2),https://github.com/pytorch/pytorch/issues/26781,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2022-11-30T07:19:24Z
issue,89686,MultiProcess tests fail when run on nodes with 1 GPU,https://github.com/pytorch/pytorch/issues/89686,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2022-11-29T16:46:49Z
issue,89054,Add a `device` keyword argument to `torch.manual_seed`,https://github.com/pytorch/pytorch/issues/89054,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: random,scan when time allows,2022-11-29T00:30:23Z
issue,85300,torchrun substitutes host names for IP addresses,https://github.com/pytorch/pytorch/issues/85300,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: elastic, oncall: distributed",scan when time allows,2022-11-28T22:42:34Z
issue,82479,Cloning conjugate tensor in torch_dispatch context produces non equality.,https://github.com/pytorch/pytorch/issues/82479,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: __torch_dispatch__, module: complex",scan when time allows,2022-11-28T18:52:21Z
issue,88942,`torch.svd_lowrank` Trigger RuntimeError under UndefinedBehaviorSanitizer,https://github.com/pytorch/pytorch/issues/88942,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: linear algebra, module: sanitizers, module: sparse",scan when time allows,2022-11-28T18:11:44Z
issue,88941,`torch.linalg.lstsq` Trigger RuntimeError under UndefinedBehaviorSanitizer,https://github.com/pytorch/pytorch/issues/88941,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: edge cases, module: linear algebra, module: sanitizers",scan when time allows,2022-11-28T18:10:02Z
issue,82534,Use NestedTensor in RNN models,https://github.com/pytorch/pytorch/issues/82534,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nestedtensor,scan when time allows,2022-11-28T18:01:03Z
issue,50718,dependency_links is deprecated,https://github.com/pytorch/pytorch/issues/50718,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: docs",scan when time allows,2022-11-28T16:11:08Z
issue,89684,`positive_semidefinite` constraint fails on CUDA 11.7,https://github.com/pytorch/pytorch/issues/89684,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: distributions, module: linear algebra",scan when time allows,2022-11-28T15:17:13Z
issue,69202,CPU only mode with cudatoolkit 11.3,https://github.com/pytorch/pytorch/issues/69202,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: binaries,scan when time allows,2022-11-28T12:27:01Z
issue,73050,aten::batch_norm schema does not mention mutation of running_mean/running_var,https://github.com/pytorch/pytorch/issues/73050,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-11-27T21:51:54Z
issue,68405,"Windows multi machine multi card training, card initialization in GLOO communication.",https://github.com/pytorch/pytorch/issues/68405,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: windows, oncall: distributed",scan when time allows,2022-11-25T15:59:35Z
issue,50471,linker error when trying to use Metal backend in PyTorch Mobile,https://github.com/pytorch/pytorch/issues/50471,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: arm, module: build",scan when time allows,2022-11-24T07:04:16Z
issue,88880,Add alphatensor support for faster matrix multiplication?,https://github.com/pytorch/pytorch/issues/88880,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-11-24T04:35:39Z
issue,89601,Building PyTorch with Vulkan backend fails (1.13 and master),https://github.com/pytorch/pytorch/issues/89601,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: vulkan,scan when time allows,2022-11-24T00:08:37Z
issue,89574,Finish deprecation of autograd decorator over class objects,https://github.com/pytorch/pytorch/issues/89574,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2022-11-23T16:51:04Z
issue,88805,[feature request] Get/set fastmath CPU bit (and some other FPU flags?),https://github.com/pytorch/pytorch/issues/88805,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: numerical-stability",scan when time allows,2022-11-23T05:19:21Z
issue,10604,"Bilinear interpolation behavior inconsistent with TF, CoreML and Caffe",https://github.com/pytorch/pytorch/issues/10604,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: interpolation,scan when time allows,2022-11-22T08:47:18Z
issue,89416,"When the torch.masked_select operator passes in the same parameters, it behaves differently on CPU and GPU.",https://github.com/pytorch/pytorch/issues/89416,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: masked operators,scan when time allows,2022-11-22T08:32:09Z
issue,89275,"torch.addcdiv: input, tensor1, and tensor2 parameters should be of the same type",https://github.com/pytorch/pytorch/issues/89275,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: assert failure, module: docs",scan when time allows,2022-11-22T01:57:50Z
issue,89125,binary_cross_entropy/bce_with_logits (+ other loss functions) for nested_tensor,https://github.com/pytorch/pytorch/issues/89125,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: loss, module: nestedtensor",scan when time allows,2022-11-21T18:02:22Z
issue,89320,`masked_fill` with `FloatTensor` mask will never mask but fails silently.,https://github.com/pytorch/pytorch/issues/89320,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: masked operators",scan when time allows,2022-11-21T15:35:54Z
issue,89301,Meta implementation for copy_ is wrong,https://github.com/pytorch/pytorch/issues/89301,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: meta tensors,scan when time allows,2022-11-21T15:33:54Z
issue,89293,fbgemm_avx512 build failure,https://github.com/pytorch/pytorch/issues/89293,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2022-11-21T15:32:02Z
issue,51450,torch.median slower than torch.sort on cpu,https://github.com/pytorch/pytorch/issues/51450,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: performance, module: reductions, module: sorting and selection",scan when time allows,2022-11-20T13:02:41Z
issue,75480,IndexError: Caught IndexError in replica 0 on device 0.,https://github.com/pytorch/pytorch/issues/75480,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2022-11-19T18:15:39Z
issue,89245,sm_80 support,https://github.com/pytorch/pytorch/issues/89245,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2022-11-19T05:17:55Z
issue,89124,Zero-copy way to make flat tensor into a nested_tensor given a shape,https://github.com/pytorch/pytorch/issues/89124,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nestedtensor,scan when time allows,2022-11-18T18:57:03Z
issue,86888,Broadcasting add for nested_tensor,https://github.com/pytorch/pytorch/issues/86888,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nestedtensor,scan when time allows,2022-11-18T18:55:56Z
issue,89218,"`torch.nn.LayerNorm` Abort with ""invalid device ordinal"" Error",https://github.com/pytorch/pytorch/issues/89218,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: error checking, module: nn, module: norms and normalization",scan when time allows,2022-11-18T08:22:04Z
issue,83313,torch.Tag doesn't have accurate mypy info,https://github.com/pytorch/pytorch/issues/83313,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: typing,scan when time allows,2022-11-17T21:13:44Z
issue,88563,`nn.functional.embedding_bag` Trigger out-of-bound Read under Compute Sanitizer,https://github.com/pytorch/pytorch/issues/88563,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2022-11-17T08:23:51Z
issue,88992,Incorrect version in the instructions on official website,https://github.com/pytorch/pytorch/issues/88992,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2022-11-16T16:43:29Z
issue,89080,Unable to backprop through dense weighted sum of sparse_coo_tensors,https://github.com/pytorch/pytorch/issues/89080,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2022-11-16T14:35:44Z
issue,7806,OOM Exception when using torch.nn.grad.conv2d_weight (apparently because CuDNN backwards is not used),https://github.com/pytorch/pytorch/issues/7806,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: cudnn, module: memory usage, module: performance",scan when time allows,2022-11-16T09:58:16Z
issue,88949,`torch.set_rng_state` Trigger RuntimeError under UndefinedBehaviorSanitizer,https://github.com/pytorch/pytorch/issues/88949,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sanitizers,scan when time allows,2022-11-16T04:24:56Z
issue,88950,`torch.nn.functional.embedding_bag` Trigger RuntimeError under UndefinedBehaviorSanitizer,https://github.com/pytorch/pytorch/issues/88950,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sanitizers,scan when time allows,2022-11-16T04:12:05Z
issue,88931,INTERNAL ASSERT FAILED. Missing scalar type infromation.,https://github.com/pytorch/pytorch/issues/88931,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nvfuser,scan when time allows,2022-11-15T23:36:20Z
issue,88991,nvprims.native_batch_norm doesn't support fake tensor inputs,https://github.com/pytorch/pytorch/issues/88991,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nvfuser, module: primtorch",scan when time allows,2022-11-15T23:18:24Z
issue,88980,Glog macro redefinition problem when including headers from both libtorch and glog,https://github.com/pytorch/pytorch/issues/88980,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2022-11-15T23:18:20Z
issue,88947,`torch.Tensor.msort` Trigger RuntimeError under UndefinedBehaviorSanitizer,https://github.com/pytorch/pytorch/issues/88947,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sanitizers,scan when time allows,2022-11-15T23:16:13Z
issue,88945,`torch.linalg.eigvals` Trigger RuntimeError under UndefinedBehaviorSanitizer,https://github.com/pytorch/pytorch/issues/88945,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sanitizers,scan when time allows,2022-11-15T23:16:06Z
issue,88944,`torch.topk` Trigger RuntimError under UndefinedBehaviorSanitizer,https://github.com/pytorch/pytorch/issues/88944,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sanitizers,scan when time allows,2022-11-15T23:15:58Z
issue,89060,Extend test_proxy_tensor tests to support ops test non floating point types,https://github.com/pytorch/pytorch/issues/89060,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: proxytensor, module: tests",scan when time allows,2022-11-15T23:07:26Z
issue,86131,Debuggability++: Share instructions for building exotic CI configurations,https://github.com/pytorch/pytorch/issues/86131,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2022-11-15T19:59:13Z
issue,88410,benchmark cache persist,https://github.com/pytorch/pytorch/issues/88410,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: cudnn",scan when time allows,2022-11-15T19:24:54Z
issue,65698,torch.onnx.export error,https://github.com/pytorch/pytorch/issues/65698,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2022-11-15T16:07:30Z
issue,65718,Migrate current Windows CI scripts off of batch,https://github.com/pytorch/pytorch/issues/65718,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ci, module: windows",scan when time allows,2022-11-15T03:59:49Z
issue,86340,Add complex support for SparseAdam and LBFGS optimizers,https://github.com/pytorch/pytorch/issues/86340,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: complex, module: optimizer",scan when time allows,2022-11-14T23:54:45Z
issue,73003,pin_memory hangs instead of throwing,https://github.com/pytorch/pytorch/issues/73003,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: dataloader",scan when time allows,2022-11-14T18:47:46Z
issue,88883,Investigate why `test_aot_autograd_symbolic_exhaustive_masked_median_cpu_float32` is flaky,https://github.com/pytorch/pytorch/issues/88883,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: flaky-tests,scan when time allows,2022-11-14T16:55:23Z
issue,88658,"nn.Linear allocate too many space which lead to CPUAllocator ""allocate memory failure"" if it's BF16. good for FP32.",https://github.com/pytorch/pytorch/issues/88658,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpu,scan when time allows,2022-11-14T12:54:22Z
issue,2001,Implement similar PyTorch function as model.summary() in keras?,https://github.com/pytorch/pytorch/issues/2001,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2022-11-14T10:19:06Z
issue,88565,What causes CPU to degrade when I load the weight with torch.hub.load(),https://github.com/pytorch/pytorch/issues/88565,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: hub",scan when time allows,2022-11-14T08:25:20Z
issue,87957,VS2022Preview ParallelCommon.cpp.obj : fatal error LNK1161: invalid export specification,https://github.com/pytorch/pytorch/issues/87957,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: static linking",scan when time allows,2022-11-14T04:48:16Z
issue,62178,Wrong implementation of method log_prob in torch.distributions.negative_binomial,https://github.com/pytorch/pytorch/issues/62178,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: distributions, module: docs, module: numpy",scan when time allows,2022-11-13T17:59:39Z
issue,87019,Rewrite `narrow_copy_dense_cpu_out` using `copy_` and `narrow`,https://github.com/pytorch/pytorch/issues/87019,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: functionalization,scan when time allows,2022-11-12T15:13:30Z
issue,88902,MPS test_numpy_ref_mps_nn_functional_group_norm_mps_float32 is flaky?,https://github.com/pytorch/pytorch/issues/88902,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: flaky-tests, module: mps",scan when time allows,2022-11-11T21:14:55Z
issue,88691,forward AD for _euclidean_dist,https://github.com/pytorch/pytorch/issues/88691,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: forward ad, module: functorch",scan when time allows,2022-11-11T18:21:29Z
issue,45115,Standardized Distributions,https://github.com/pytorch/pytorch/issues/45115,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distributions,scan when time allows,2022-11-10T21:01:45Z
issue,88612,Bernoulli uses legacy contiguous memory format,https://github.com/pytorch/pytorch/issues/88612,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: memory format,scan when time allows,2022-11-10T17:19:16Z
issue,88643,pack_sequence() always fail after set_default_tensor_type to CUDA,https://github.com/pytorch/pytorch/issues/88643,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rnn,scan when time allows,2022-11-10T17:18:34Z
issue,88639,CUDA unknown error after suspend during debugging,https://github.com/pytorch/pytorch/issues/88639,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2022-11-10T17:17:58Z
issue,88598,"Despite having aten::diag_embed.out, torch.diag_embed doesn't support out= argument",https://github.com/pytorch/pytorch/issues/88598,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: codegen,scan when time allows,2022-11-10T17:16:07Z
issue,88591,`pack_padded_sequence` not compatible with deterministic mode it calls `torch.scatter`,https://github.com/pytorch/pytorch/issues/88591,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: determinism, module: rnn",scan when time allows,2022-11-10T17:15:38Z
issue,88614,The libtorch test TestScalarTensor.TestScalarTensorMPS fails on Apple Silicon,https://github.com/pytorch/pytorch/issues/88614,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: m1, module: mps",scan when time allows,2022-11-10T17:14:39Z
issue,88616,The libtorch test SequentialTest.ModuleForwardMethodOptionalArg fails on Apple Silicon,https://github.com/pytorch/pytorch/issues/88616,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: m1",scan when time allows,2022-11-10T17:13:19Z
issue,88474,Mixed precision training fails due to NaN in batch norm running_mean,https://github.com/pytorch/pytorch/issues/88474,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision),scan when time allows,2022-11-08T06:36:19Z
issue,88309,ASAN shard 4 started to OOM after unrelated commit,https://github.com/pytorch/pytorch/issues/88309,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2022-11-07T23:15:33Z
issue,88002,Einsum Optimization Tracker,https://github.com/pytorch/pytorch/issues/88002,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: linear algebra, module: performance",scan when time allows,2022-11-07T16:38:40Z
issue,39009,Missing OneCycleLR and MultiplicativeLR in lr_scheduler.pyi,https://github.com/pytorch/pytorch/issues/39009,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: typing,scan when time allows,2022-11-07T04:31:48Z
issue,88142,Refactor `torch.return_types.topk` to behave like a `namedtuple` or a `dict`,https://github.com/pytorch/pytorch/issues/88142,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sorting and selection,scan when time allows,2022-11-06T19:55:05Z
issue,88389,Whether to support libtorch source code compilation of C++11 ？,https://github.com/pytorch/pytorch/issues/88389,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2022-11-06T14:38:23Z
issue,82219,[feature request] DataLoader to accept num_threads argument to auto-set number of threads for OpenMP / intra-op parallelism,https://github.com/pytorch/pytorch/issues/82219,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dataloader, module: openmp",scan when time allows,2022-11-04T12:52:26Z
issue,61110,Incorrect exponential calculation on Jetson devices with float32 dtype,https://github.com/pytorch/pytorch/issues/61110,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: arm, module: cuda, module: jetson",scan when time allows,2022-11-04T09:07:05Z
issue,88423,view_as_real and split_with_sizes links in Tensor Views docs are broken,https://github.com/pytorch/pytorch/issues/88423,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2022-11-03T18:18:14Z
issue,88408,Unit test with `--subprocess` command doesn't respect the `-k` filter flag and runs all available sub tests,https://github.com/pytorch/pytorch/issues/88408,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ci, module: testing, module: tests",scan when time allows,2022-11-03T18:12:30Z
issue,25767,"Model parallel with DDP get `Socket Timeout` error when using NCCL, while GLOO works fine",https://github.com/pytorch/pytorch/issues/25767,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2022-11-03T08:56:44Z
issue,88194,Add a config option to raise errors instead of warnings in nvFuser integration,https://github.com/pytorch/pytorch/issues/88194,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nvfuser,scan when time allows,2022-11-02T16:50:59Z
issue,88004,link error happen when intergrate libtorch to other tool,https://github.com/pytorch/pytorch/issues/88004,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cpp",scan when time allows,2022-11-02T16:09:19Z
issue,53261,Dimension argument names in torch.diag_embed/diagonal vs. transpose/transpose_,https://github.com/pytorch/pytorch/issues/53261,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: numpy",scan when time allows,2022-11-02T14:28:55Z
issue,50274,torch.equal is divergent from np.equal,https://github.com/pytorch/pytorch/issues/50274,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: deprecation, module: numpy, module: testing",scan when time allows,2022-11-02T13:32:34Z
issue,88304,AttributeError: module 'tensorboard.compat.tensorflow_stub.io.gfile' has no attribute 'MakeDirs',https://github.com/pytorch/pytorch/issues/88304,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tensorboard,scan when time allows,2022-11-02T13:18:15Z
issue,86529,[NvFuser] JIT model with `mul+atan+sgn` will access illegal memory on cuda when computing gradient,https://github.com/pytorch/pytorch/issues/86529,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nvfuser,scan when time allows,2022-11-02T01:08:12Z
issue,88192,[docs] torch.is_neg/torch.Tensor.is_neg not documented,https://github.com/pytorch/pytorch/issues/88192,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2022-11-01T19:46:05Z
issue,86644,"[NvFuser] INTERNAL ASSERT FAIL ""ScalarType should be static for Tensors in fusion for amp optimization""",https://github.com/pytorch/pytorch/issues/86644,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: assert failure, module: nvfuser",scan when time allows,2022-11-01T17:15:57Z
issue,88147,Conv2d is not deterministic when input tensor has different strides,https://github.com/pytorch/pytorch/issues/88147,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: convolution,scan when time allows,2022-11-01T15:30:36Z
issue,88191,`torch.nn.RReLU` not reporting `lower > upper` on CUDA,https://github.com/pytorch/pytorch/issues/88191,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2022-11-01T14:11:03Z
issue,88189,Moving tensor to GPU by .cuda() gets stucked when AMD Secure Encripted Virtualization (SEV) is activated,https://github.com/pytorch/pytorch/issues/88189,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: rocm",scan when time allows,2022-11-01T14:10:12Z
issue,88185,`torch.mm` Trigger RuntimeError with UndefinedBehaviorSanitizer,https://github.com/pytorch/pytorch/issues/88185,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sanitizers,scan when time allows,2022-11-01T14:08:25Z
issue,87841,`max_unpool3d` will trigger an assertion fail under compute sanitizer,https://github.com/pytorch/pytorch/issues/87841,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: pooling, module: sanitizers",scan when time allows,2022-11-01T13:49:35Z
issue,87955,AttributeError: '_MultiProcessingDataLoaderIter' object has no attribute 'next',https://github.com/pytorch/pytorch/issues/87955,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2022-11-01T13:42:41Z
issue,88062,Cannot import `traverse_dps` from torch.data.utils.graph,https://github.com/pytorch/pytorch/issues/88062,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: data,scan when time allows,2022-11-01T13:39:38Z
issue,87902,Permute,https://github.com/pytorch/pytorch/issues/87902,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2022-11-01T13:38:02Z
issue,87979,amp with `bf16`: backward happens in `f16` when using `@torch.cuda.amp.custom_bwd`,https://github.com/pytorch/pytorch/issues/87979,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision), module: bfloat16, module: half",scan when time allows,2022-10-31T23:32:55Z
issue,88081,out of memory with pytorch version after 1.8.1,https://github.com/pytorch/pytorch/issues/88081,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: cudacachingallocator, module: memory usage",scan when time allows,2022-10-31T21:46:14Z
issue,88027,"getting error error: namespace ""cub"" has no member ""Debug"" when try to build v1.8.2 with CUDA 11.6",https://github.com/pytorch/pytorch/issues/88027,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2022-10-31T19:20:33Z
issue,88053,Different behaviour in sparse matmul,https://github.com/pytorch/pytorch/issues/88053,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2022-10-31T19:12:08Z
issue,84234,Randomness should be consistent across devices with use_deterministic_algorithms,https://github.com/pytorch/pytorch/issues/84234,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: determinism, module: random",scan when time allows,2022-10-31T18:26:04Z
issue,88103,ProcessGroupNCCL watchdog can't catch NCCL comm initialization issues,https://github.com/pytorch/pytorch/issues/88103,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: c10d, oncall: distributed",scan when time allows,2022-10-31T17:56:55Z
issue,87706,C++ Adagrad optimizer doesn't initialize parameter state,https://github.com/pytorch/pytorch/issues/87706,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: optimizer",scan when time allows,2022-10-31T14:28:43Z
issue,62917,Use `linecache.lazycache` in torch.fx,https://github.com/pytorch/pytorch/issues/62917,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, module: fx",scan when time allows,2022-10-29T21:00:33Z
issue,18229,Add support for tuple type deduction in C++ custom operators,https://github.com/pytorch/pytorch/issues/18229,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, module: internals",scan when time allows,2022-10-29T14:11:39Z
issue,82259,[Reproducibility] Make tests say when unusual environment variables are set that change behavior of the test,https://github.com/pytorch/pytorch/issues/82259,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2022-10-28T21:56:45Z
issue,83081,[CI] Split up periodic.yml into forward-fixable.yml and periodic.yml,https://github.com/pytorch/pytorch/issues/83081,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2022-10-28T21:03:06Z
issue,39245,pytorch-crf model to onnx conversion,https://github.com/pytorch/pytorch/issues/39245,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2022-10-28T01:43:46Z
issue,87163,New APIs for cuda graph inspection and manipulation,https://github.com/pytorch/pytorch/issues/87163,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda graphs,scan when time allows,2022-10-27T19:39:06Z
issue,87539,RAM leak when copying tensor from cpu to cuda,https://github.com/pytorch/pytorch/issues/87539,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: memory usage",scan when time allows,2022-10-27T19:04:42Z
issue,87642,`libtorch_cpu.so` is exposing some LLVM symbols,https://github.com/pytorch/pytorch/issues/87642,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cpp-extensions",scan when time allows,2022-10-27T14:40:00Z
issue,86449,torch.cuda.empty_cache() is not working,https://github.com/pytorch/pytorch/issues/86449,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2022-10-27T14:14:42Z
issue,39716,Do not modify global random state,https://github.com/pytorch/pytorch/issues/39716,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: random",scan when time allows,2022-10-27T02:52:13Z
issue,19126,weight_norm doesn't support eta and returns nan for zero weights,https://github.com/pytorch/pytorch/issues/19126,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nans and infs, module: nn, module: norms and normalization",scan when time allows,2022-10-26T23:54:12Z
issue,87555,Improve `c10d::ReduceOp` & `torch.distributed.distributed_c10d.ReduceOp`,https://github.com/pytorch/pytorch/issues/87555,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2022-10-26T23:43:19Z
issue,87388,Document dist.new_subgroups,https://github.com/pytorch/pytorch/issues/87388,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, oncall: distributed",scan when time allows,2022-10-26T17:15:48Z
issue,87634,Add tests for ProcessGroup cpp extensions,https://github.com/pytorch/pytorch/issues/87634,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2022-10-26T16:32:55Z
issue,17798,Feature Request: deterministic CUDA torch.nn.CTCLoss,https://github.com/pytorch/pytorch/issues/17798,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: determinism, module: loss, module: nn",scan when time allows,2022-10-26T12:10:37Z
issue,87701,pytorch/pytorch cpu official Docker images,https://github.com/pytorch/pytorch/issues/87701,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docker,scan when time allows,2022-10-25T20:55:10Z
issue,87589,"#error ""Expected GLOO_USE_CUDA to be defined""",https://github.com/pytorch/pytorch/issues/87589,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: third_party",scan when time allows,2022-10-25T02:13:38Z
issue,87571,Testing insufficient to catch incorrect dispatch key for bernoulli.p re functionalization,https://github.com/pytorch/pytorch/issues/87571,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: functionalization,scan when time allows,2022-10-24T22:15:08Z
issue,68169,torch.histogram: Sum of PDFs is num of bins instead of 1,https://github.com/pytorch/pytorch/issues/68169,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: sorting and selection",scan when time allows,2022-10-24T21:16:59Z
issue,87451,Some operations do not keep `channels_last` memory format which yields accuracy drop,https://github.com/pytorch/pytorch/issues/87451,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: memory format,scan when time allows,2022-10-24T19:08:09Z
issue,86020,Functorch memory_efficient_fusion gives wrong output batch size,https://github.com/pytorch/pytorch/issues/86020,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: functorch",scan when time allows,2022-10-24T17:25:20Z
issue,87267,The installation commands given on the pytorch website will not install properly,https://github.com/pytorch/pytorch/issues/87267,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: binaries,scan when time allows,2022-10-24T13:08:56Z
issue,87494,Missing docstring for resize_as,https://github.com/pytorch/pytorch/issues/87494,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2022-10-23T19:26:37Z
issue,87551,`bmm` will return wrong result on cpu with in-place,https://github.com/pytorch/pytorch/issues/87551,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: edge cases,scan when time allows,2022-10-23T19:16:31Z
issue,87170,torch.linalg.cond gives inconsistent results on CPU/CUDA,https://github.com/pytorch/pytorch/issues/87170,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: linear algebra, module: nans and infs",scan when time allows,2022-10-21T15:19:33Z
issue,82687,Move nested-tensor tutorial from prototype,https://github.com/pytorch/pytorch/issues/82687,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nestedtensor,scan when time allows,2022-10-21T14:53:33Z
issue,87389,Installing PyTorch with BUILD_SPLIT_CUDA=ON and CUDNN fails on linker error,https://github.com/pytorch/pytorch/issues/87389,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cuda, module: cudnn",scan when time allows,2022-10-21T14:36:01Z
issue,87371,Better type annotations for `torch.Tensor` subclasses,https://github.com/pytorch/pytorch/issues/87371,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: typing,scan when time allows,2022-10-21T13:45:29Z
issue,87276,torch.save throws ValueError: ctypes objects containing pointers cannot be pickled,https://github.com/pytorch/pytorch/issues/87276,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: serialization,scan when time allows,2022-10-21T13:04:28Z
issue,87352,RuntimeError: Tensors of type TensorImpl do not have numel,https://github.com/pytorch/pytorch/issues/87352,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cuda, module: docker",scan when time allows,2022-10-21T13:03:01Z
issue,28743,Sampler for IterableDataset,https://github.com/pytorch/pytorch/issues/28743,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2022-10-21T09:05:24Z
issue,84336,torch 1.12.1 cuda 10.2 runs slower than torch 1.8.2 cuda 10.2,https://github.com/pytorch/pytorch/issues/84336,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: cudnn, module: performance",scan when time allows,2022-10-20T18:27:23Z
issue,87222,`custom_jvp` and `custom_vjp`,https://github.com/pytorch/pytorch/issues/87222,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: functorch,scan when time allows,2022-10-20T15:06:08Z
issue,84415,Deepcopy of FX graph fails with nested make_fx and constant tensors,https://github.com/pytorch/pytorch/issues/84415,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-10-20T13:06:19Z
issue,74539,Making PyTorch more pythonic: the PyTorch functions should be inspectable.,https://github.com/pytorch/pytorch/issues/74539,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-10-20T04:44:09Z
issue,76613,MeanVarianceNormalization,https://github.com/pytorch/pytorch/issues/76613,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: norms and normalization",scan when time allows,2022-10-19T23:11:28Z
issue,87041,"Segmentation fault: 11 when running ""import torch"" on Mac OS X",https://github.com/pytorch/pytorch/issues/87041,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: macos,scan when time allows,2022-10-19T18:25:55Z
issue,86265,TORCH_WARN is executed just once per set of parameters,https://github.com/pytorch/pytorch/issues/86265,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: logging,scan when time allows,2022-10-18T21:23:24Z
issue,87087,Failing periodic tests: test_dense_mask_index_cpu (__main__.CpuTests) & est_expanded_reduction_cpu (__main__.CpuTests),https://github.com/pytorch/pytorch/issues/87087,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: flaky-tests,scan when time allows,2022-10-18T19:56:35Z
issue,87178,The autogenerated out variants via `autogen:` do not check that the dtype of the `out` kwarg via `canCast`.,https://github.com/pytorch/pytorch/issues/87178,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: codegen,scan when time allows,2022-10-18T19:55:55Z
issue,87031,Improve Readability of error(s) when provided unexpected keyword arguments.,https://github.com/pytorch/pytorch/issues/87031,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: error checking, module: python frontend",scan when time allows,2022-10-18T17:52:36Z
issue,87003,Multiprocessing DataLoader pickles multiprocessing.Queues incorrectly,https://github.com/pytorch/pytorch/issues/87003,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2022-10-18T17:26:10Z
issue,86968,Drop deprecated behavior from NumPy-style `T`,https://github.com/pytorch/pytorch/issues/86968,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: deprecation, module: numpy",scan when time allows,2022-10-18T17:22:36Z
issue,87076,Nandense layer for missing values,https://github.com/pytorch/pytorch/issues/87076,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: masked operators, module: nn, module: sparse",scan when time allows,2022-10-18T14:58:47Z
issue,87196,Missing `docker` directory in `tools/`,https://github.com/pytorch/pytorch/issues/87196,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docker, module: docs",scan when time allows,2022-10-18T14:44:49Z
issue,87080,cppextension host compiler check ignores executable symbolic link in CUDA bin directory,https://github.com/pytorch/pytorch/issues/87080,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cpp-extensions",scan when time allows,2022-10-18T02:21:29Z
issue,87056,Pipe conveys inconsistent value in GPU env,https://github.com/pytorch/pytorch/issues/87056,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: multiprocessing,scan when time allows,2022-10-18T01:31:58Z
issue,50549,Channels last doesn't improve speed when using SyncBatchNorm,https://github.com/pytorch/pytorch/issues/50549,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: memory format, module: performance",scan when time allows,2022-10-17T22:47:54Z
issue,87033,Saving and loading from physical storage,https://github.com/pytorch/pytorch/issues/87033,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: memory usage, module: serialization",scan when time allows,2022-10-17T16:09:26Z
issue,84616,Autogenerated out functions are missing at::cpu:: and co bindings,https://github.com/pytorch/pytorch/issues/84616,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: codegen,scan when time allows,2022-10-16T03:06:05Z
issue,86890,View-based advanced indexing (Integer array/LongTensor indexing) of nested_tensor,https://github.com/pytorch/pytorch/issues/86890,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: advanced indexing,scan when time allows,2022-10-15T13:51:00Z
issue,83764,Missing header file,https://github.com/pytorch/pytorch/issues/83764,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-10-15T02:48:24Z
issue,67350,[FX] autowrap_functions doesn't work for module-scoped functions,https://github.com/pytorch/pytorch/issues/67350,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-10-14T19:23:15Z
issue,86965,Upgrade to a newer llvm-openmp version to avoid `/dev/shm` pollution,https://github.com/pytorch/pytorch/issues/86965,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: openmp",scan when time allows,2022-10-14T16:09:47Z
issue,86814,Expanding the parameters of `torch.svd_lowrank`,https://github.com/pytorch/pytorch/issues/86814,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: linear algebra,scan when time allows,2022-10-14T14:00:23Z
issue,86918,torch.cat on empty tensor is bogus,https://github.com/pytorch/pytorch/issues/86918,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: edge cases,scan when time allows,2022-10-14T13:57:13Z
issue,86819,Could not run select_backward [vmap] [dlrm] [functorch],https://github.com/pytorch/pytorch/issues/86819,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: functorch,scan when time allows,2022-10-13T19:48:59Z
issue,86798,TF32 conv_transpose2d with groups has bad precision compared to fp32,https://github.com/pytorch/pytorch/issues/86798,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: numerical-stability, module: tf32",scan when time allows,2022-10-13T14:47:09Z
issue,72556,"Clarify test dependencies (e.g., into a test-requirements.txt file)",https://github.com/pytorch/pytorch/issues/72556,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, module: ci",scan when time allows,2022-10-13T13:22:30Z
issue,86315,Feature request: Tests for `int` should be tests for `numbers.Integral`,https://github.com/pytorch/pytorch/issues/86315,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: numpy, module: python frontend",scan when time allows,2022-10-12T19:10:51Z
issue,56485,Sparse tensor CSR layout for CUDA,https://github.com/pytorch/pytorch/issues/56485,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: sparse",scan when time allows,2022-10-12T16:45:45Z
issue,9222,[feature request] Implementing Block Sparse Operations,https://github.com/pytorch/pytorch/issues/9222,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2022-10-12T16:26:38Z
issue,86381,OpInfo Tests To Validate that All Operators Are Being Tested With Strided Tensors,https://github.com/pytorch/pytorch/issues/86381,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tests,scan when time allows,2022-10-12T15:30:32Z
issue,86782,Poisson sampling on GPU fails for high rates,https://github.com/pytorch/pytorch/issues/86782,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: random",scan when time allows,2022-10-12T14:18:58Z
issue,75747,Depthwise Conv1d performance (a naive CUDA kernel is 10x faster),https://github.com/pytorch/pytorch/issues/75747,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2022-10-12T13:25:31Z
issue,66491,Normalize handling of scalar arguments,https://github.com/pytorch/pytorch/issues/66491,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tests,scan when time allows,2022-10-12T08:43:12Z
issue,86465,Wrong results with torch.linalg.inv on batched matrices when using cuda,https://github.com/pytorch/pytorch/issues/86465,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: cuda, module: linear algebra, module: magma",scan when time allows,2022-10-12T00:57:00Z
issue,86717,NVFuser `FusionRootMappingMultipleBroadcast_CUDA` raises exception on sm_80+,https://github.com/pytorch/pytorch/issues/86717,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ci, module: cuda, module: nvfuser",scan when time allows,2022-10-11T20:18:03Z
issue,86444,Dedicated function for shallow_copy_and_detach,https://github.com/pytorch/pytorch/issues/86444,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2022-10-11T16:11:55Z
issue,86676,Pytorch built for Jetson errors if CUDA is not found,https://github.com/pytorch/pytorch/issues/86676,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: arm, module: build, module: cuda",scan when time allows,2022-10-11T13:06:22Z
issue,36990,[docs] Explain active_bytes in torch.cuda.memory_stats and Cuda Memory Management,https://github.com/pytorch/pytorch/issues/36990,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2022-10-11T07:44:02Z
issue,85841,Setting the cuda device when using start_processes in Jupyter on Ampere leads to CUDA reinitialization error,https://github.com/pytorch/pytorch/issues/85841,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: multiprocessing",scan when time allows,2022-10-11T05:51:54Z
issue,86548,About autocast,https://github.com/pytorch/pytorch/issues/86548,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision),scan when time allows,2022-10-11T05:51:13Z
issue,86467,torch.tensor obj automatically moved to shared memory upon Process launch,https://github.com/pytorch/pytorch/issues/86467,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: multiprocessing,scan when time allows,2022-10-11T05:48:56Z
issue,86601,NVFuser batch norm with prims: internal assert failure from test suite,https://github.com/pytorch/pytorch/issues/86601,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nvfuser, module: primtorch, module: tests",scan when time allows,2022-10-11T05:41:44Z
issue,86660,libtorch make failed,https://github.com/pytorch/pytorch/issues/86660,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2022-10-11T05:41:28Z
issue,86544,Compile failed at allreduce without gloo,https://github.com/pytorch/pytorch/issues/86544,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2022-10-11T00:03:16Z
issue,86493,`mem_get_info` reserves memory and can not be destroyed / deallocated.,https://github.com/pytorch/pytorch/issues/86493,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2022-10-10T20:38:09Z
issue,86271,AUTOGRAD is not working on IOS,https://github.com/pytorch/pytorch/issues/86271,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: ios, oncall: mobile",scan when time allows,2022-10-10T18:17:16Z
issue,86551,`MultiLabelMarginLoss` will return incorrect values in JIT after the first run on cuda,https://github.com/pytorch/pytorch/issues/86551,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: nvfuser, oncall: jit",scan when time allows,2022-10-10T17:59:48Z
issue,86563,[functorch] colab links on functorch 0.2.0 website should be linked to a permalinked version of the colabs,https://github.com/pytorch/pytorch/issues/86563,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: functorch",scan when time allows,2022-10-10T16:54:17Z
issue,86554,[NvFuser] would change the output for some inaccurate dtype,https://github.com/pytorch/pytorch/issues/86554,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nvfuser,scan when time allows,2022-10-10T15:07:41Z
issue,86192,fmt/src/os.cc: error: unknown type name 'error_code'; did you mean 'std::error_code'?,https://github.com/pytorch/pytorch/issues/86192,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2022-10-10T12:00:36Z
issue,86204,How to perform unstructured interpolation,https://github.com/pytorch/pytorch/issues/86204,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: interpolation,scan when time allows,2022-10-10T11:59:06Z
issue,86194,path in WORKSPACE,https://github.com/pytorch/pytorch/issues/86194,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: bazel,scan when time allows,2022-10-10T11:57:45Z
issue,15457,Improve one_hot,https://github.com/pytorch/pytorch/issues/15457,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2022-10-10T04:29:41Z
issue,44938,DataParallel on CPU,https://github.com/pytorch/pytorch/issues/44938,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: data parallel,scan when time allows,2022-10-08T14:32:22Z
issue,86111,When will the torch.sparse module be usable?,https://github.com/pytorch/pytorch/issues/86111,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2022-10-07T20:24:12Z
issue,86298,AOT Autograd Device Partitioning,https://github.com/pytorch/pytorch/issues/86298,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: functorch,scan when time allows,2022-10-06T22:33:21Z
issue,86270,Autocast with BF16 on CPU slows down model more than 2X,https://github.com/pytorch/pytorch/issues/86270,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision), module: bfloat16, module: performance",scan when time allows,2022-10-06T20:07:12Z
issue,86326,`torch.special.round` doesn't support the same dtypes as `torch.round`,https://github.com/pytorch/pytorch/issues/86326,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: special,scan when time allows,2022-10-06T17:20:27Z
issue,86356,`conv_transpose` is not similar to `nn.grad.conv_input` when `output_padding` is passed with non-default values.,https://github.com/pytorch/pytorch/issues/86356,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2022-10-06T17:18:58Z
issue,86339,Add `maximize` support to LBFGS optimizer,https://github.com/pytorch/pytorch/issues/86339,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2022-10-05T23:51:39Z
issue,32694,ninja: build stopped: subcommand failed.,https://github.com/pytorch/pytorch/issues/32694,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2022-10-05T22:05:25Z
issue,85775,Timed out receiving the shared seed from the distribtued store on Rank 2,https://github.com/pytorch/pytorch/issues/85775,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dataloader, oncall: distributed",scan when time allows,2022-10-05T13:01:48Z
issue,86225,[Feature] Dispatching PyTorch Distributed Collectives,https://github.com/pytorch/pytorch/issues/86225,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: c10d, oncall: distributed",scan when time allows,2022-10-05T11:31:18Z
issue,69505,"with the same environment ,pytorch 1.8 worked but 1.10 can't work",https://github.com/pytorch/pytorch/issues/69505,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: cuda, module: wsl",scan when time allows,2022-10-04T22:50:29Z
issue,85607,[Distributed: RPC] Failed to initialize RPC with >18 workers,https://github.com/pytorch/pytorch/issues/85607,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: rpc, oncall: distributed",scan when time allows,2022-10-04T20:51:17Z
issue,85377,CI fails for test_compare_cpu_nn_functional_embedding_cuda_float32 which is not reproducible locally,https://github.com/pytorch/pytorch/issues/85377,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: embedding, module: tests",scan when time allows,2022-10-04T19:24:36Z
issue,85889,High occupation on GPU 0 when converting Tensor to multi GPU,https://github.com/pytorch/pytorch/issues/85889,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: performance",scan when time allows,2022-10-04T17:07:24Z
issue,5565,[feature request] Different interpolation algos for 'grid_sample' function,https://github.com/pytorch/pytorch/issues/5565,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: interpolation",scan when time allows,2022-10-04T10:00:42Z
issue,85618,Build from source failed with error of different gpu architecture (compiler shows sm_30-related error but I use sm_86 GPU),https://github.com/pytorch/pytorch/issues/85618,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cuda",scan when time allows,2022-10-04T08:25:23Z
issue,85625,How to install pytorch with cuda 11.7 in anaconda envirment?,https://github.com/pytorch/pytorch/issues/85625,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-10-04T08:23:30Z
issue,47568,Failed to get generated_cpp list,https://github.com/pytorch/pytorch/issues/47568,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2022-10-04T04:56:29Z
issue,86055,partial view/reshaping,https://github.com/pytorch/pytorch/issues/86055,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python frontend,scan when time allows,2022-10-03T23:22:37Z
issue,24470,Benchmark cudnn version of grid sampler,https://github.com/pytorch/pytorch/issues/24470,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cudnn, module: performance",scan when time allows,2022-10-03T22:27:28Z
issue,86074,torch.remainder and torch.fmod produce wrong results,https://github.com/pytorch/pytorch/issues/86074,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: numerical-stability",scan when time allows,2022-10-03T22:22:47Z
issue,56489,[torchdeploy] ar: _bz2module.o: No such file or directory,https://github.com/pytorch/pytorch/issues/56489,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: deploy,scan when time allows,2022-10-03T17:48:52Z
issue,84316,Torch.fx tracing bug with dictionary.update calls on input,https://github.com/pytorch/pytorch/issues/84316,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-10-03T17:19:52Z
issue,83851,torch fx cannot trace assert for some cases,https://github.com/pytorch/pytorch/issues/83851,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-10-03T17:19:52Z
issue,84515,Torch.FX work with autograd.Function,https://github.com/pytorch/pytorch/issues/84515,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-10-03T17:18:24Z
issue,85815,string interning for dispatcher operator names,https://github.com/pytorch/pytorch/issues/85815,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dispatch,scan when time allows,2022-10-03T17:15:05Z
issue,59356,pytorch build failure on arm64,https://github.com/pytorch/pytorch/issues/59356,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: arm, module: build, module: cuda, module: jetson",scan when time allows,2022-10-03T15:51:47Z
issue,86097,[Activation Checkpointing] Investigate pin_memory for CPU offload,https://github.com/pytorch/pytorch/issues/86097,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: activation checkpointing, oncall: distributed",scan when time allows,2022-10-03T06:17:05Z
issue,85698,AMP consumes 30x gpu memory with bmm,https://github.com/pytorch/pytorch/issues/85698,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision),scan when time allows,2022-09-30T22:08:41Z
issue,82565,PyTorch for quantum mechanics,https://github.com/pytorch/pytorch/issues/82565,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: scientific computing,scan when time allows,2022-09-30T09:19:33Z
issue,80942,[LTC] OOM on mnist example,https://github.com/pytorch/pytorch/issues/80942,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: lazy,scan when time allows,2022-09-30T02:39:48Z
issue,85834,[primTorch] Need to update data-dependent check policy,https://github.com/pytorch/pytorch/issues/85834,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: primtorch,scan when time allows,2022-09-29T19:32:17Z
issue,85652,CUDA allocator feature requests,https://github.com/pytorch/pytorch/issues/85652,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: cudacachingallocator",scan when time allows,2022-09-29T17:37:56Z
issue,64254,[feature request] New copyseq_ method,https://github.com/pytorch/pytorch/issues/64254,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tensor creation,scan when time allows,2022-09-29T08:16:09Z
issue,84537,Minimal example for torch.optim.SparseAdam,https://github.com/pytorch/pytorch/issues/84537,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: sparse",scan when time allows,2022-09-28T16:11:18Z
issue,85642,Could not run 'aten::native_batch_norm' with arguments from the 'SparseCUDA' backend. using batch_norm,https://github.com/pytorch/pytorch/issues/85642,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2022-09-28T15:29:36Z
issue,85544,NvFuser single mode changes the output,https://github.com/pytorch/pytorch/issues/85544,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nvfuser,scan when time allows,2022-09-28T14:15:09Z
issue,56144,libtorch static linking results in undefined references to onnx_torch and caffe2::EmbeddingLookup,https://github.com/pytorch/pytorch/issues/56144,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: undefined reference,scan when time allows,2022-09-28T09:12:26Z
issue,60295,Optimize torch.einsum,https://github.com/pytorch/pytorch/issues/60295,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: linear algebra, module: performance",scan when time allows,2022-09-27T14:38:15Z
issue,14367,Error: DeviceGuardImpl for cpu is not available (static linking PyTorch),https://github.com/pytorch/pytorch/issues/14367,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: static linking",scan when time allows,2022-09-27T13:37:00Z
issue,85606,Creating NumPy array with `dtype=object` of PyTorch tensors fails,https://github.com/pytorch/pytorch/issues/85606,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: numpy,scan when time allows,2022-09-27T08:28:47Z
issue,85558,torch.Tensor.transpose().contiguous() on dimension of size 1 gives wrong stride,https://github.com/pytorch/pytorch/issues/85558,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2022-09-26T20:36:16Z
issue,85520,Implement `rand_like` ref and implement nvfuser_impl for `uniform` prim,https://github.com/pytorch/pytorch/issues/85520,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nvfuser,scan when time allows,2022-09-26T17:44:07Z
issue,85590,Solve default argument induced include cycles by not using defaults / moving the defaults to inl,https://github.com/pytorch/pytorch/issues/85590,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: internals",scan when time allows,2022-09-26T15:02:20Z
issue,52973,Add numerically stable methods to torch.distributions (e.g. logcdf),https://github.com/pytorch/pytorch/issues/52973,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distributions,scan when time allows,2022-09-26T09:16:00Z
issue,85201,performance between manually created graph and CUDAGraph.replay,https://github.com/pytorch/pytorch/issues/85201,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda graphs,scan when time allows,2022-09-26T02:01:23Z
issue,34839,[torch.jit.script] Allow `range` to index into Tensor,https://github.com/pytorch/pytorch/issues/34839,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2022-09-24T09:28:56Z
issue,85227,topk returns different results with the same input in cuda and cpu,https://github.com/pytorch/pytorch/issues/85227,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: sorting and selection",scan when time allows,2022-09-23T02:45:34Z
issue,31818,How to distinguish different layers in hook？,https://github.com/pytorch/pytorch/issues/31818,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2022-09-22T22:55:48Z
issue,85375,Inconsistency between geometric distributions,https://github.com/pytorch/pytorch/issues/85375,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: distributions, module: molly-guard",scan when time allows,2022-09-22T18:47:03Z
issue,85438,Prim Output Spec is Not Always Consistent With Eager,https://github.com/pytorch/pytorch/issues/85438,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: primtorch,scan when time allows,2022-09-22T15:24:49Z
issue,69991,Composite Compliance Problems Tracker,https://github.com/pytorch/pytorch/issues/69991,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: __torch_dispatch__, module: internals, module: linear algebra",scan when time allows,2022-09-22T00:22:45Z
issue,85397,torch.nn.utils.prune.remove reorders the parameters of a module unexpectedly,https://github.com/pytorch/pytorch/issues/85397,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: pruning,scan when time allows,2022-09-21T19:57:57Z
issue,85392,please report a bug to PyTorch. Expected Object but got PyObject,https://github.com/pytorch/pytorch/issues/85392,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: torchbind,scan when time allows,2022-09-21T19:56:32Z
issue,85390,Please put back missing rocm builds of Torch Vision.,https://github.com/pytorch/pytorch/issues/85390,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: rocm",scan when time allows,2022-09-21T19:55:16Z
issue,85387,very strange speed of torch.bmm with specific tensor shape,https://github.com/pytorch/pytorch/issues/85387,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: linear algebra, module: performance",scan when time allows,2022-09-21T19:54:24Z
issue,85351,Point community docs to master,https://github.com/pytorch/pytorch/issues/85351,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: doc infra,scan when time allows,2022-09-21T19:50:02Z
issue,85302,Remove `TypedStorage` and use only `UntypedStorage`,https://github.com/pytorch/pytorch/issues/85302,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python frontend,scan when time allows,2022-09-21T19:43:51Z
issue,85229,TracerWarning: Output nr 1. of the traced function does not match the corresponding output of the Python function.,https://github.com/pytorch/pytorch/issues/85229,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2022-09-21T19:34:33Z
issue,78050,RFC: [primTorch] Stride-agnostic Operator Semantics,https://github.com/pytorch/pytorch/issues/78050,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: primtorch, module: python frontend",scan when time allows,2022-09-21T16:31:43Z
issue,76288,Multi-GPU distributed training reports errors,https://github.com/pytorch/pytorch/issues/76288,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: c10d, oncall: distributed",scan when time allows,2022-09-21T02:30:04Z
issue,85258,"nvFuser support for {ceil,floor,round,trunc}(int)",https://github.com/pytorch/pytorch/issues/85258,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nvfuser,scan when time allows,2022-09-20T15:24:52Z
issue,81648,`sparse_coo.to_dense()` produces different results between CPU and CUDA backends for boolean non-coalesced inputs.,https://github.com/pytorch/pytorch/issues/81648,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2022-09-20T11:51:09Z
issue,84265,torch.Tensor.to.dtype_layout overload is not available in Python,https://github.com/pytorch/pytorch/issues/84265,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: codegen, module: python frontend",scan when time allows,2022-09-19T23:25:20Z
issue,83733,BCELoss results in autocast CUDA warning,https://github.com/pytorch/pytorch/issues/83733,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision),scan when time allows,2022-09-19T18:45:57Z
issue,85058,Selectively sync internal Meta discussions / posts to dev-discuss.pytorch.org,https://github.com/pytorch/pytorch/issues/85058,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-09-19T17:20:05Z
issue,32998,torch batchwise max with indices,https://github.com/pytorch/pytorch/issues/32998,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: batching, module: reductions",scan when time allows,2022-09-19T07:45:44Z
issue,84835,Batch multiplication for torch.sparse matrix multiplication,https://github.com/pytorch/pytorch/issues/84835,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: linear algebra, module: sparse",scan when time allows,2022-09-17T16:24:00Z
issue,4954,[docs] Docs website search finds duplicates and produces bad snippets,https://github.com/pytorch/pytorch/issues/4954,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: doc infra, module: docs",scan when time allows,2022-09-16T23:45:38Z
issue,85147,torch::quantile performance?,https://github.com/pytorch/pytorch/issues/85147,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: performance,scan when time allows,2022-09-16T20:56:48Z
issue,85122,"PT Dispatcher confusing error message ""There were no tensor arguments to this function""",https://github.com/pytorch/pytorch/issues/85122,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dispatch,scan when time allows,2022-09-16T20:55:01Z
issue,85105,Python dispatch for PyOps needs to respect tensor subclasses,https://github.com/pytorch/pytorch/issues/85105,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: __torch_dispatch__, module: dispatch",scan when time allows,2022-09-16T20:54:52Z
issue,85082,"[NNC] loop vectorization fails, `Ramp` and `Broadcast` undefined",https://github.com/pytorch/pytorch/issues/85082,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-09-16T20:30:09Z
issue,85078,"libtorch create a tensor is very slow, who can tell me why",https://github.com/pytorch/pytorch/issues/85078,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: performance",scan when time allows,2022-09-16T20:28:03Z
issue,46386,"GPU memory leak when registering a forward hook with ""self"" access",https://github.com/pytorch/pytorch/issues/46386,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: memory usage",scan when time allows,2022-09-16T19:40:23Z
issue,85121,make_traced() doesn't respect setting the seed,https://github.com/pytorch/pytorch/issues/85121,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: primtorch, module: tests",scan when time allows,2022-09-15T22:26:24Z
issue,85096,Update `use_deterministic_algorithms` documentation and tests to include `nn.functional` counterparts for all `nn` modules,https://github.com/pytorch/pytorch/issues/85096,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: determinism, module: docs, module: tests",scan when time allows,2022-09-15T18:47:18Z
issue,85081,primTorch/nvfuser: have a way to check that refs are added to __all__,https://github.com/pytorch/pytorch/issues/85081,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nvfuser, module: primtorch",scan when time allows,2022-09-15T18:35:36Z
issue,38240,RuntimeError: arg_types.size() == param_names.size() - (moduleSelf_ ? 1 : 0) INTERNAL ASSERT FAILED,https://github.com/pytorch/pytorch/issues/38240,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: typing, oncall: jit",scan when time allows,2022-09-15T11:11:57Z
issue,84813,MPS: allow selecting specific MTLDevice by registryID via environment variable,https://github.com/pytorch/pytorch/issues/84813,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: mps,scan when time allows,2022-09-14T21:22:12Z
issue,77514,`max_pool1d` can succeed when padding is negative for tensor requiring grad,https://github.com/pytorch/pytorch/issues/77514,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: edge cases, module: nn, module: pooling",scan when time allows,2022-09-14T21:03:45Z
issue,83817,[feature request] Add new device type works on CPU,https://github.com/pytorch/pytorch/issues/83817,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-09-14T07:10:57Z
issue,84717,Test aten decompositions match their alias information,https://github.com/pytorch/pytorch/issues/84717,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: primtorch,scan when time allows,2022-09-13T23:52:13Z
issue,84864,torch.utils.checkpoint (with use_reentrant=False) doesn't work with all PyTorch features that set TLS,https://github.com/pytorch/pytorch/issues/84864,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: __torch_dispatch__, module: activation checkpointing",scan when time allows,2022-09-13T23:45:08Z
issue,84805,compiling failed from source,https://github.com/pytorch/pytorch/issues/84805,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cuda, module: sparse",scan when time allows,2022-09-13T23:42:24Z
issue,84932,Separate doc and binaries build,https://github.com/pytorch/pytorch/issues/84932,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: doc infra,scan when time allows,2022-09-13T23:36:41Z
issue,84840,"TypeError: finfo(): argument 'type' (position 1) must be torch.dtype, not HFProxy",https://github.com/pytorch/pytorch/issues/84840,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-09-13T21:54:09Z
issue,84721,PyTorch-DirectML RFC,https://github.com/pytorch/pytorch/issues/84721,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-09-13T21:50:36Z
issue,84710,About the different ways to print models,https://github.com/pytorch/pytorch/issues/84710,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: printing,scan when time allows,2022-09-13T21:48:39Z
issue,79039,Add Autograd Support for Nested Tensor,https://github.com/pytorch/pytorch/issues/79039,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nestedtensor,scan when time allows,2022-09-13T18:10:48Z
issue,1927,[feature request] time-distributed layers for application of normal layers to sequence data,https://github.com/pytorch/pytorch/issues/1927,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2022-09-13T17:05:57Z
issue,72709,Feature: Use iterative refinement algorithm from cuSOLVER for the least-squares solver on CUDA,https://github.com/pytorch/pytorch/issues/72709,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: linear algebra",scan when time allows,2022-09-13T10:41:57Z
issue,84538,OpInfo tests should compare gpu to cpu implementations,https://github.com/pytorch/pytorch/issues/84538,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tests,scan when time allows,2022-09-12T18:36:11Z
issue,63921,More efficient colon backwards in advanced indexing,https://github.com/pytorch/pytorch/issues/63921,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: advanced indexing, module: performance",scan when time allows,2022-09-12T17:13:39Z
issue,84620,torch.hub.load local model,https://github.com/pytorch/pytorch/issues/84620,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: hub,scan when time allows,2022-09-12T15:10:40Z
issue,66162,Cuda Out of Memory - MMDMatic,https://github.com/pytorch/pytorch/issues/66162,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: memory usage,scan when time allows,2022-09-11T21:07:01Z
issue,84691,test_public_bindings is not robust to various build options,https://github.com/pytorch/pytorch/issues/84691,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-09-08T18:38:57Z
issue,84597,Accept SymInts and SymFloats For Scalar Inputs,https://github.com/pytorch/pytorch/issues/84597,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-09-08T18:35:11Z
issue,84565,Tensor Subclass that doesn't require grad may wrap a Tensor subclass that requires grad,https://github.com/pytorch/pytorch/issues/84565,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-09-07T23:35:19Z
issue,84630,"torch.nn.functional.interpolate fails on some degenerate shapes, but passes on others",https://github.com/pytorch/pytorch/issues/84630,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: interpolation,scan when time allows,2022-09-07T23:30:24Z
issue,25387,[DRAFT] Auto-casting in JIT - Automatic mixed precision,https://github.com/pytorch/pytorch/issues/25387,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2022-09-07T22:13:30Z
issue,77733,Adding Vulkan Support,https://github.com/pytorch/pytorch/issues/77733,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: vulkan,scan when time allows,2022-09-07T21:52:49Z
issue,84625,Beta distribution behaves incorrectly for small parameters,https://github.com/pytorch/pytorch/issues/84625,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: distributions, module: edge cases",scan when time allows,2022-09-07T21:08:11Z
issue,32166,"C++ randint returns float32, python returns int64",https://github.com/pytorch/pytorch/issues/32166,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp,scan when time allows,2022-09-07T17:59:29Z
issue,23233,"Unofficial ARMv6, ARMv7, ARMv8 builds",https://github.com/pytorch/pytorch/issues/23233,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: docs",scan when time allows,2022-09-07T12:54:11Z
issue,59243,Can't install pytorch =1.7 with python 3.8.5 on Raspberry Pi,https://github.com/pytorch/pytorch/issues/59243,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: arm,scan when time allows,2022-09-07T12:50:58Z
issue,80541,[Prims+NVFuser] Prims with missing NVFuser ops,https://github.com/pytorch/pytorch/issues/80541,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nvfuser, module: primtorch",scan when time allows,2022-09-07T10:44:28Z
issue,84468,[c10d] Support a public API to retrieve default process group,https://github.com/pytorch/pytorch/issues/84468,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2022-09-06T22:16:26Z
issue,83032,Support for CSR Tensor with NN layers,https://github.com/pytorch/pytorch/issues/83032,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: sparse",scan when time allows,2022-09-06T18:36:13Z
issue,84529,"test_prims.py:test_nvfuser_no_args_cuda, memory leak",https://github.com/pytorch/pytorch/issues/84529,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: primtorch,scan when time allows,2022-09-06T13:41:41Z
issue,84539,list of tensors can't be converted to a torch tensor while list of lists gets easily converted to a pytorch tensor,https://github.com/pytorch/pytorch/issues/84539,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: numpy,scan when time allows,2022-09-06T13:40:22Z
issue,84560,[optim] asgd : handling of complex params as real params (NaN vs inf),https://github.com/pytorch/pytorch/issues/84560,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: edge cases, module: optimizer",scan when time allows,2022-09-06T13:38:03Z
issue,83977,pytorch 1.12.1 doesn't build with ffmpeg 5.0,https://github.com/pytorch/pytorch/issues/83977,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2022-09-06T08:46:13Z
issue,75940,Add nesting of nested Tensor,https://github.com/pytorch/pytorch/issues/75940,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nestedtensor,scan when time allows,2022-09-05T18:43:22Z
issue,61781,Letting `_allgather_base` to support multiple tensors as inputs and outputs,https://github.com/pytorch/pytorch/issues/61781,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2022-09-05T08:41:36Z
issue,83902,"Bug in batch names with matmul (result tensor has names=('i', 'i', 'k')).",https://github.com/pytorch/pytorch/issues/83902,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: named tensor,scan when time allows,2022-09-05T07:05:49Z
issue,84335,Should enable skipped tests for `to` OpInfo,https://github.com/pytorch/pytorch/issues/84335,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: primtorch,scan when time allows,2022-09-03T19:46:23Z
issue,84445,Strange cuda illegal memory allocation error,https://github.com/pytorch/pytorch/issues/84445,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2022-09-02T20:06:45Z
issue,84473,Install LibTorch by Conan or other C++ package manager,https://github.com/pytorch/pytorch/issues/84473,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp,scan when time allows,2022-09-02T16:58:09Z
issue,84414,several questions about pytorch DDP,https://github.com/pytorch/pytorch/issues/84414,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ddp, module: nccl",scan when time allows,2022-09-02T16:52:37Z
issue,84412,Odd type-casting behaviour in prims.div,https://github.com/pytorch/pytorch/issues/84412,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: primtorch, module: type promotion",scan when time allows,2022-09-02T16:49:23Z
issue,84353,torch.Size should convert all elements to ints,https://github.com/pytorch/pytorch/issues/84353,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python frontend,scan when time allows,2022-09-01T15:00:55Z
issue,84311,DecompositionInterpreter creates invalid graph,https://github.com/pytorch/pytorch/issues/84311,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nvfuser,scan when time allows,2022-09-01T14:25:34Z
issue,84346,fix ATen tests that do not compile,https://github.com/pytorch/pytorch/issues/84346,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-09-01T14:24:24Z
issue,84259,would you like upload to the cpp libtorch to vcpkg package repo?,https://github.com/pytorch/pytorch/issues/84259,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ci, module: cpp",scan when time allows,2022-09-01T09:37:04Z
issue,71022,installation of pytorch `cpuonly` from conda with `nomkl` installs `mkl`,https://github.com/pytorch/pytorch/issues/71022,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: binaries,scan when time allows,2022-08-30T12:49:41Z
issue,83968,"Python3 Depletes 2021 M1 Mac Memory Running Training Ops For Model's M, L and X",https://github.com/pytorch/pytorch/issues/83968,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: macos,scan when time allows,2022-08-30T12:10:43Z
issue,84194,Pytorch gets small bias on the result of different types of divisors while doing floating point division.,https://github.com/pytorch/pytorch/issues/84194,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: numerical-stability",scan when time allows,2022-08-30T06:02:45Z
issue,84247,Ensure ops account for offsets and strides,https://github.com/pytorch/pytorch/issues/84247,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nestedtensor,scan when time allows,2022-08-30T03:32:01Z
issue,84202,Gradient value calculation error in MultiLabelMarginLoss,https://github.com/pytorch/pytorch/issues/84202,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: cuda, module: loss",scan when time allows,2022-08-29T23:32:33Z
issue,84140,Please include virtual/physical batch sizes in the tutorials,https://github.com/pytorch/pytorch/issues/84140,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2022-08-29T22:22:29Z
issue,80742,Add TorchDynamo as a submodule to Pytorch?,https://github.com/pytorch/pytorch/issues/80742,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2022-08-29T20:03:57Z
issue,84045,COREMLTOOLs/NNPACK Python Issue,https://github.com/pytorch/pytorch/issues/84045,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nnpack,scan when time allows,2022-08-29T13:56:00Z
issue,84050,I cannot install pytorch by Bad CRC-32 for file 'torch/lib/libtorch_cpu.so',https://github.com/pytorch/pytorch/issues/84050,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2022-08-29T13:43:02Z
issue,67590,Add `OptState.UPDATED` to `torch.cuda.amp.GradScaler`,https://github.com/pytorch/pytorch/issues/67590,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision),scan when time allows,2022-08-29T11:58:17Z
issue,42812,checkpoint_sequential breaks backpropagation,https://github.com/pytorch/pytorch/issues/42812,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: activation checkpointing, module: autograd",scan when time allows,2022-08-29T11:32:00Z
issue,80594,RuntimeError: DataLoader worker (pid 22822) is killed by signal: Aborted.,https://github.com/pytorch/pytorch/issues/80594,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2022-08-28T22:05:52Z
issue,44380,Support for NVIDIA UVM technology,https://github.com/pytorch/pytorch/issues/44380,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2022-08-28T21:53:19Z
issue,83775,[Nested Tensor] Move nested tensor specific ops to nested namespace,https://github.com/pytorch/pytorch/issues/83775,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nestedtensor,scan when time allows,2022-08-24T22:26:18Z
issue,81620,PyTorch 1.12 cu113 Illegal Memory Access or Internal Error instead of Out of Memory cases,https://github.com/pytorch/pytorch/issues/81620,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: cudnn",scan when time allows,2022-08-24T21:26:16Z
issue,83923,Support primtorch view ops in functionalization,https://github.com/pytorch/pytorch/issues/83923,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: functionalization, module: primtorch, module: viewing and reshaping",scan when time allows,2022-08-24T20:58:07Z
issue,83914,"RAM not free when deleting a model in CPU? worse after inference, is there some cache hidden?",https://github.com/pytorch/pytorch/issues/83914,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: memory usage,scan when time allows,2022-08-24T20:41:52Z
issue,84014,"fill_ OpInfo code not used, also, doesn't test the case where the second argument is a Tensor",https://github.com/pytorch/pytorch/issues/84014,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tests,scan when time allows,2022-08-24T20:40:39Z
issue,83931,primTorch: support refs and decompositions when ATen and Python disagree,https://github.com/pytorch/pytorch/issues/83931,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: primtorch,scan when time allows,2022-08-24T20:32:38Z
issue,83932,Hitting rate limits for pytorchbot token,https://github.com/pytorch/pytorch/issues/83932,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: infra,scan when time allows,2022-08-24T20:32:20Z
issue,83979,Undefined reference in libtorch_cpu.so `...std::__cxx11::basic_string...`,https://github.com/pytorch/pytorch/issues/83979,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2022-08-24T20:31:14Z
issue,83980,Scatter min/max reduce operation that returns the corresponding indices,https://github.com/pytorch/pytorch/issues/83980,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: scatter & gather ops,scan when time allows,2022-08-24T20:23:10Z
issue,84009,[Nested Tensor] Enable Nestedtensor to work with OpInfos,https://github.com/pytorch/pytorch/issues/84009,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nestedtensor,scan when time allows,2022-08-24T19:27:12Z
issue,83910,Tracking nested tensor functions with backward kernels registered in derivatives.yaml,https://github.com/pytorch/pytorch/issues/83910,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nestedtensor,scan when time allows,2022-08-24T17:59:44Z
issue,83947,Are PyTorch Android nightly builds getting automatically published,https://github.com/pytorch/pytorch/issues/83947,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: android, module: ci",scan when time allows,2022-08-24T01:17:45Z
issue,83941,empty_quantized should probably be new_empty_quantized,https://github.com/pytorch/pytorch/issues/83941,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2022-08-23T21:42:29Z
issue,83826,test_lazy spuriously fails if LAPACK is not installed,https://github.com/pytorch/pytorch/issues/83826,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: lazy, module: tests",scan when time allows,2022-08-23T14:39:39Z
issue,83714,Build from source failed on MacOS 10.6 with CUDA 10.1,https://github.com/pytorch/pytorch/issues/83714,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: macos",scan when time allows,2022-08-23T14:04:00Z
issue,83726,nvfuser + prim stack generated illegal PTX code on hardware with sm <= 70,https://github.com/pytorch/pytorch/issues/83726,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nvfuser, module: primtorch",scan when time allows,2022-08-23T13:29:15Z
issue,76128,test_license_for_wheel always fails on my local dev copy,https://github.com/pytorch/pytorch/issues/76128,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tests,scan when time allows,2022-08-22T15:45:32Z
issue,83320,Torch1.10.2 is slower than torch1.9.1,https://github.com/pytorch/pytorch/issues/83320,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ddp, oncall: distributed",scan when time allows,2022-08-22T02:48:45Z
issue,61653,Move `lobpcg` into `torch.linalg`,https://github.com/pytorch/pytorch/issues/61653,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: linear algebra,scan when time allows,2022-08-21T10:39:09Z
issue,72146,Also allow dicts as type of `params=` field in param groups of optimizers,https://github.com/pytorch/pytorch/issues/72146,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2022-08-20T12:42:20Z
issue,83749,[Nested Tensor] Update TestCase.AssertEqual,https://github.com/pytorch/pytorch/issues/83749,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nestedtensor, module: testing",scan when time allows,2022-08-20T09:45:31Z
issue,73171,libtorch need operator= in torch::Device,https://github.com/pytorch/pytorch/issues/73171,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp,scan when time allows,2022-08-19T15:15:10Z
issue,83657,[Discussion] Add custom device,https://github.com/pytorch/pytorch/issues/83657,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: backend, module: dispatch",scan when time allows,2022-08-19T14:37:17Z
issue,83148,Cannot call CUDAGeneratorImpl::current_seed during CUDA graph capture,https://github.com/pytorch/pytorch/issues/83148,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2022-08-19T07:32:20Z
issue,18414,Add layer-wise adaptive rate scaling (LARS) optimizer,https://github.com/pytorch/pytorch/issues/18414,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2022-08-19T04:17:41Z
issue,83624,Unhelpful error message from torch.linalg.ldl_factor,https://github.com/pytorch/pytorch/issues/83624,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: edge cases, module: linear algebra",scan when time allows,2022-08-18T15:23:38Z
issue,68055,Release official Python 3.9 images,https://github.com/pytorch/pytorch/issues/68055,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: binaries,scan when time allows,2022-08-18T14:39:45Z
issue,62640,add gamma to CosineAnnealingWarmRestarts so max lr can decrease in cycles,https://github.com/pytorch/pytorch/issues/62640,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: lrscheduler, module: optimizer",scan when time allows,2022-08-18T14:38:22Z
issue,63488,MultiStepLR with different gammas for each parameter group,https://github.com/pytorch/pytorch/issues/63488,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: lrscheduler, module: optimizer",scan when time allows,2022-08-18T14:37:33Z
issue,70352,Rprop Optimizer: UnboundLocalError: local variable 'step_size_min' referenced before assignment,https://github.com/pytorch/pytorch/issues/70352,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2022-08-18T14:31:15Z
issue,74022,Inconsistent implementation on SWA,https://github.com/pytorch/pytorch/issues/74022,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: optimizer",scan when time allows,2022-08-18T14:24:46Z
issue,81162,SparseAdam performance issue during optimizer step,https://github.com/pytorch/pytorch/issues/81162,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: optimizer, module: performance, module: sparse",scan when time allows,2022-08-18T14:08:34Z
issue,83579,Conv1d: NNPACK SpatialConvolution_updateOutput failed when batchsize or padding is too large,https://github.com/pytorch/pytorch/issues/83579,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: nn, module: nnpack",scan when time allows,2022-08-17T21:22:49Z
issue,83564,KL-divergence of two Generalized Dirichlet distributions,https://github.com/pytorch/pytorch/issues/83564,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distributions,scan when time allows,2022-08-17T14:18:26Z
issue,83583,Silent promotion of bool to int in the dispatcher,https://github.com/pytorch/pytorch/issues/83583,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: library, module: pybind, module: type promotion",scan when time allows,2022-08-17T13:13:04Z
issue,83493,Calling torch.linalg.cholesky on a CPU tensor requires compiling PyTorch with LAPACK.,https://github.com/pytorch/pytorch/issues/83493,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: linear algebra,scan when time allows,2022-08-17T06:54:24Z
issue,78018,Three memory copies of every dataloader cpu tensor,https://github.com/pytorch/pytorch/issues/78018,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: dataloader, module: multiprocessing",scan when time allows,2022-08-16T22:04:22Z
issue,83537,Add a new argument `check_inf=True` (by default) or check_pos_inf / check_neg_inf to anomaly mode,https://github.com/pytorch/pytorch/issues/83537,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2022-08-16T21:10:11Z
issue,83521,quantize_per_tensor/quantize_per_channel operators should honor the quant_min/quant_max from observer,https://github.com/pytorch/pytorch/issues/83521,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2022-08-16T16:44:10Z
issue,81317,Incorrect results for mean or sum kernels on aarch64 when building with gcc-7,https://github.com/pytorch/pytorch/issues/81317,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: arm,scan when time allows,2022-08-16T10:58:21Z
issue,50953,Dispatch-less structured wrapper / composite / alias kernels,https://github.com/pytorch/pytorch/issues/50953,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: internals, module: structured kernels",scan when time allows,2022-08-16T08:12:10Z
issue,63079,"PyTorch 1.9.0, test_optim fails on Nvidia A10.",https://github.com/pytorch/pytorch/issues/63079,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: optimizer, module: tests",scan when time allows,2022-08-15T14:52:21Z
issue,83392,"How to turn off determinism just for specific operations, e.g. upsampling through bilinear interpolation?",https://github.com/pytorch/pytorch/issues/83392,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: determinism",scan when time allows,2022-08-15T04:42:36Z
issue,83315,dataparallel function doesn't work,https://github.com/pytorch/pytorch/issues/83315,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: data parallel,scan when time allows,2022-08-15T02:23:58Z
issue,79016,Corner cases of ShardedTensor checkpoint when using TorchRec,https://github.com/pytorch/pytorch/issues/79016,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: activation checkpointing, oncall: distributed",scan when time allows,2022-08-12T19:59:47Z
issue,83292,TorchVision testing in CI + test_fx,https://github.com/pytorch/pytorch/issues/83292,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-08-12T17:38:40Z
issue,83271,cannot import name 'ProcessGroup' from 'torch.distributed',https://github.com/pytorch/pytorch/issues/83271,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: macos, oncall: distributed",scan when time allows,2022-08-12T13:03:44Z
issue,83272,bmm operator in bfloat16 has low TFLOPS for some tensor shapes with CUDA 11.6,https://github.com/pytorch/pytorch/issues/83272,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cublas,scan when time allows,2022-08-12T06:37:34Z
issue,83193,"module 'torch.distributed' has no attribute 'pipeline' - macOS, PyTorch 1.12.1",https://github.com/pytorch/pytorch/issues/83193,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2022-08-12T06:00:49Z
issue,83245,"During DDP training timm densenet121, mobilenetv2(v3) models do not save state_dict correctly.",https://github.com/pytorch/pytorch/issues/83245,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ddp, oncall: distributed",scan when time allows,2022-08-12T05:51:20Z
issue,83214,torchgen.model.FunctionSchema.parse fails with following ops' schema,https://github.com/pytorch/pytorch/issues/83214,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: codegen,scan when time allows,2022-08-11T23:19:26Z
issue,83019,TestCommon.test_dtypes error message is confusing,https://github.com/pytorch/pytorch/issues/83019,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: testing,scan when time allows,2022-08-11T21:56:54Z
issue,83024,New PR template suggests a pattern that does not close PR,https://github.com/pytorch/pytorch/issues/83024,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-08-11T21:56:36Z
issue,83152,"When padding is big int, torch.nn.functional.fold runs too long and can't return result",https://github.com/pytorch/pytorch/issues/83152,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2022-08-11T21:54:55Z
issue,83157,Fail to install torch for source,https://github.com/pytorch/pytorch/issues/83157,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2022-08-11T21:54:33Z
issue,83175,"torch.nn.GRU runs long time, when num_layers is large",https://github.com/pytorch/pytorch/issues/83175,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: edge cases, module: nn",scan when time allows,2022-08-11T21:53:02Z
issue,83169,torch.nn.functional.softplus / torch.nn.Softplus parameter beta can be set to zero,https://github.com/pytorch/pytorch/issues/83169,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2022-08-11T21:52:59Z
issue,83232,torch.nn.PixelShuffle error message wrong,https://github.com/pytorch/pytorch/issues/83232,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2022-08-11T21:44:35Z
issue,83238,The type of parameter 'p' in torch.nn.TripletMarginLoss wrong,https://github.com/pytorch/pytorch/issues/83238,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2022-08-11T21:44:29Z
issue,83229,torch.nn.MaxUnpool2d get negative size tensor,https://github.com/pytorch/pytorch/issues/83229,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2022-08-11T21:43:59Z
issue,83221,torch.nn.InstanceNorm{1|2|3}d doesn't verify the value type of parameter num_features,https://github.com/pytorch/pytorch/issues/83221,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2022-08-11T21:43:34Z
issue,83281,addcdiv_ (in and out of place) not implemented for torch.float16 and cpu,https://github.com/pytorch/pytorch/issues/83281,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: half,scan when time allows,2022-08-11T21:41:43Z
issue,83234,torch.nn.ReplicationPad{1|2}d supports more input dimension than are written on documentation,https://github.com/pytorch/pytorch/issues/83234,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: nn",scan when time allows,2022-08-11T21:40:55Z
issue,83257,PyYAML not listed as a dependency,https://github.com/pytorch/pytorch/issues/83257,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: releng,scan when time allows,2022-08-11T21:38:17Z
issue,79325,[NVFuser] hitting fallbacks on demucs (from torchbench + lazy tensor),https://github.com/pytorch/pytorch/issues/79325,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nvfuser,scan when time allows,2022-08-11T17:21:16Z
issue,71403,[discussion] torch.flatten to allow unsqueeze of inexisting dimension,https://github.com/pytorch/pytorch/issues/71403,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: batching,scan when time allows,2022-08-11T09:02:08Z
issue,83074,DPP training incompatibility with checkpoint and detach,https://github.com/pytorch/pytorch/issues/83074,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ddp, oncall: distributed",scan when time allows,2022-08-11T01:34:34Z
issue,82960,torch.bitwise_xor argument 'other' documentation wrong,https://github.com/pytorch/pytorch/issues/82960,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2022-08-10T13:23:41Z
issue,83060,torch.empty_strided argument 'size'and 'stride' documentation wrong,https://github.com/pytorch/pytorch/issues/83060,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2022-08-10T13:17:29Z
issue,83064,Updating the LTS version of the torch (1.8.2 -> 1.10.2\1.11.2?),https://github.com/pytorch/pytorch/issues/83064,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: binaries,scan when time allows,2022-08-10T13:07:21Z
issue,83082,[CI] Create periodic fuzzy testing for PyTorch build flags,https://github.com/pytorch/pytorch/issues/83082,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2022-08-10T13:02:55Z
issue,83135,torch.nn.functional.avg_pool{1|2|3}d error message does not match what is described in the documentation,https://github.com/pytorch/pytorch/issues/83135,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: nn",scan when time allows,2022-08-10T12:57:45Z
issue,7686,Only one thread is used on macOS (super slow on CPU),https://github.com/pytorch/pytorch/issues/7686,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: macos, module: multithreading",scan when time allows,2022-08-10T10:18:22Z
issue,82793,Profiling results on CPU is not reliable,https://github.com/pytorch/pytorch/issues/82793,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: performance,scan when time allows,2022-08-10T05:46:13Z
issue,82382,Problems in built-from-source pytorch with USE_DEPLOY=1 in Ubuntu,https://github.com/pytorch/pytorch/issues/82382,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: deploy,scan when time allows,2022-08-10T00:59:28Z
issue,82802,"[ROCm] build instruction is haphazard missing information unclear, build does not work",https://github.com/pytorch/pytorch/issues/82802,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: rocm",scan when time allows,2022-08-09T18:49:48Z
issue,75703,Let's host NVIDIA dependencies in our own S3,https://github.com/pytorch/pytorch/issues/75703,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2022-08-09T17:16:19Z
issue,10043,Sparse tensor use cases,https://github.com/pytorch/pytorch/issues/10043,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2022-08-09T16:05:28Z
issue,82577,RFC: Add flag for RNN decomposition to all RNN modules,https://github.com/pytorch/pytorch/issues/82577,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rnn,scan when time allows,2022-08-09T14:41:32Z
issue,82494,Implement torch.clamp() on sparse tensors with SparseCPU backend,https://github.com/pytorch/pytorch/issues/82494,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2022-08-08T19:46:39Z
issue,82727,DecompositionInterpreter creates invalid graphs for FX graph modules created with torch.fx.symbolic_trace,https://github.com/pytorch/pytorch/issues/82727,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-08-08T15:57:43Z
issue,82926,"Slice operation on ""ragged"" dimension in NestedTensor",https://github.com/pytorch/pytorch/issues/82926,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nestedtensor,scan when time allows,2022-08-08T15:57:16Z
issue,73160,Preprocessing function for backend coreml is not registered,https://github.com/pytorch/pytorch/issues/73160,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: mobile,scan when time allows,2022-08-08T09:26:22Z
issue,39917,build QT program use libtorch-cxx11-abi-shared-with-deps-1.5.0+cu101 ok with CPU but error with cuda GPU,https://github.com/pytorch/pytorch/issues/39917,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2022-08-08T05:36:42Z
issue,82871,GPU arch 8.6 is not covered by the `TORCH_CUDA_ARCH_LIST = All` option,https://github.com/pytorch/pytorch/issues/82871,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cuda",scan when time allows,2022-08-05T21:34:35Z
issue,81680,Provide an option to disable CUDA_GCC_VERSIONS,https://github.com/pytorch/pytorch/issues/81680,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2022-08-05T19:30:25Z
issue,82545,Incorrect CPU implementation of CTCLoss backward step,https://github.com/pytorch/pytorch/issues/82545,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: loss",scan when time allows,2022-08-05T18:56:58Z
issue,7857,Feature Request: Logistic Distribution,https://github.com/pytorch/pytorch/issues/7857,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distributions,scan when time allows,2022-08-05T17:19:53Z
issue,82669,unittest.subTest and way to selectively mark subTests as expected failures,https://github.com/pytorch/pytorch/issues/82669,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: testing,scan when time allows,2022-08-05T15:12:52Z
issue,81297,Files downloaded with torch.hub should respect umask,https://github.com/pytorch/pytorch/issues/81297,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: hub,scan when time allows,2022-08-05T14:50:17Z
issue,82823,getDLContext in DLConvertor.h cannot be found,https://github.com/pytorch/pytorch/issues/82823,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dlpack,scan when time allows,2022-08-05T13:30:49Z
issue,82789,[LibTorch] the C++ api needs detailed error reports like pytorch,https://github.com/pytorch/pytorch/issues/82789,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: logging,scan when time allows,2022-08-05T13:04:56Z
issue,82465,Does torch.utils.checkpoint compatible with torch.cuda.make_graphed_callables?,https://github.com/pytorch/pytorch/issues/82465,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: activation checkpointing, module: cuda graphs",scan when time allows,2022-08-04T13:53:46Z
issue,53438,Pipe should automatically build appropriate partitions.,https://github.com/pytorch/pytorch/issues/53438,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-08-04T02:03:20Z
issue,82764,Subclass of Tensor doesn't support __format__,https://github.com/pytorch/pytorch/issues/82764,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-08-03T23:24:28Z
issue,82761,torch.Tensor.bag() should automatically implement bagging,https://github.com/pytorch/pytorch/issues/82761,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-08-03T23:23:59Z
issue,82710,build fail when using lto with gcc,https://github.com/pytorch/pytorch/issues/82710,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2022-08-03T21:55:47Z
issue,82751,Refactor how errors decide whether to append C++ stacktrace,https://github.com/pytorch/pytorch/issues/82751,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-08-03T20:28:56Z
issue,82610,contiguous() not work for rank 1 length 1 tensor.,https://github.com/pytorch/pytorch/issues/82610,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dlpack,scan when time allows,2022-08-03T16:42:23Z
issue,82542,Is there Doc that explains how to call an extension op in another extension implementation?,https://github.com/pytorch/pytorch/issues/82542,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: docs",scan when time allows,2022-08-03T15:18:05Z
issue,38310,Compiling errors when trying to cross-compile the C++ API for RTOS (QNX),https://github.com/pytorch/pytorch/issues/38310,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cpp",scan when time allows,2022-08-02T20:39:53Z
issue,82583,DataLoader parameter pin_memory_device should accept torch.device type,https://github.com/pytorch/pytorch/issues/82583,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2022-08-02T18:38:44Z
issue,82303,Race condition between torch.tensor's view and /= (/= returns incorrect result),https://github.com/pytorch/pytorch/issues/82303,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: partial aliasing,scan when time allows,2022-08-02T17:42:20Z
issue,82510,Complex addition result in NaN when it shouldn't,https://github.com/pytorch/pytorch/issues/82510,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: complex, module: nans and infs",scan when time allows,2022-08-02T17:29:40Z
issue,78759,Add `inplace_view` tag for `resize_`,https://github.com/pytorch/pytorch/issues/78759,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: viewing and reshaping,scan when time allows,2022-08-02T17:06:26Z
issue,56500,Marking variable not required in backward calculation if it is not needed,https://github.com/pytorch/pytorch/issues/56500,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2022-08-02T15:45:19Z
issue,78482,RuntimeError: Event device type CUDA does not match blocking stream’s device type CPU,https://github.com/pytorch/pytorch/issues/78482,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: cuda, module: tests",scan when time allows,2022-08-02T01:34:14Z
issue,72749,"pytorchmergebot doesn't react to comments left from ""files"" tab",https://github.com/pytorch/pytorch/issues/72749,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2022-08-01T18:08:27Z
issue,81955,Investigate adding shell linter/checker to CI,https://github.com/pytorch/pytorch/issues/81955,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2022-08-01T18:05:45Z
issue,81945,Investigate adding Dockerfile linter hadolint to CI,https://github.com/pytorch/pytorch/issues/81945,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2022-08-01T18:01:18Z
issue,71896,CUDA Graph API Improvement,https://github.com/pytorch/pytorch/issues/71896,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda graphs,scan when time allows,2022-08-01T16:51:10Z
issue,82145,set_grad_enabled not respected when running on a web server,https://github.com/pytorch/pytorch/issues/82145,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: dependency bug",scan when time allows,2022-08-01T16:40:26Z
issue,39570,Dataloader._shutdown_workers hangs,https://github.com/pytorch/pytorch/issues/39570,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dataloader, module: multiprocessing",scan when time allows,2022-08-01T09:10:11Z
issue,82377,masked_scatter_ is very lacking,https://github.com/pytorch/pytorch/issues/82377,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: scatter & gather ops",scan when time allows,2022-07-30T17:11:05Z
issue,26957,Batched Dataloader,https://github.com/pytorch/pytorch/issues/26957,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2022-07-30T16:11:13Z
issue,36914,Problem with c10/utils/variant.h,https://github.com/pytorch/pytorch/issues/36914,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp,scan when time allows,2022-07-29T12:17:40Z
issue,50328,RTX3090 performs no better than 1080ti,https://github.com/pytorch/pytorch/issues/50328,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: performance,scan when time allows,2022-07-29T06:26:36Z
issue,69663,"After updated to pytorch1.10.0 cuda11.1, NCCL is not available",https://github.com/pytorch/pytorch/issues/69663,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2022-07-29T05:22:15Z
issue,82357,ufmt and flake8 lints race,https://github.com/pytorch/pytorch/issues/82357,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-07-28T20:47:00Z
issue,82419,The torch::deploy document is not updated,https://github.com/pytorch/pytorch/issues/82419,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: deploy,scan when time allows,2022-07-28T20:46:53Z
issue,82081,Could be clearer that Cross Entropy takes logits as input,https://github.com/pytorch/pytorch/issues/82081,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: loss, module: nn",scan when time allows,2022-07-28T16:11:16Z
issue,54408,It is strange that PyTorch is slow on RTX 3090,https://github.com/pytorch/pytorch/issues/54408,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: performance",scan when time allows,2022-07-28T07:40:13Z
issue,69025,[subgraph_rewriter] Support skipping matching for some ops,https://github.com/pytorch/pytorch/issues/69025,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-07-27T22:44:23Z
issue,69577,[subgraph_rewriter] Add match_filter for the matches,https://github.com/pytorch/pytorch/issues/69577,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-07-27T22:30:53Z
issue,81541,[vulkan]compiling VulkanOpContext.cpp with some errors,https://github.com/pytorch/pytorch/issues/81541,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: vulkan",scan when time allows,2022-07-27T20:47:31Z
issue,81669,Register refs for CompositeImplicitAutograd ops as decompositions,https://github.com/pytorch/pytorch/issues/81669,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: primtorch,scan when time allows,2022-07-27T20:34:13Z
issue,82185,DataLoader: `pin_memory` should respect object attributes before object collection type,https://github.com/pytorch/pytorch/issues/82185,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2022-07-27T15:48:00Z
issue,82276,DEBUG=1 env var doesn't actually set DEBUG preprocessor macro,https://github.com/pytorch/pytorch/issues/82276,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2022-07-27T13:33:20Z
issue,82242,logspace inconsistently casts inputs to int before performing computation,https://github.com/pytorch/pytorch/issues/82242,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tensor creation,scan when time allows,2022-07-26T23:56:51Z
issue,82230,logspace and linspace off by one on cuda for integer dtypes for some inputs,https://github.com/pytorch/pytorch/issues/82230,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tensor creation,scan when time allows,2022-07-26T23:56:34Z
issue,82239,primtorch refs should be composite compliant,https://github.com/pytorch/pytorch/issues/82239,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: __torch_dispatch__, module: primtorch",scan when time allows,2022-07-26T23:55:06Z
issue,81568,Improve interaction of PyTorch downstream libraries and torchdeploy,https://github.com/pytorch/pytorch/issues/81568,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: deploy,scan when time allows,2022-07-26T21:03:57Z
issue,82197,Provide error handling for ops that don't yet support Dynamic Shape,https://github.com/pytorch/pytorch/issues/82197,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-07-26T20:30:01Z
issue,81996,linspace cpu and sometimes cuda is wrong on integral types,https://github.com/pytorch/pytorch/issues/81996,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: edge cases, module: python frontend",scan when time allows,2022-07-26T17:54:25Z
issue,82212,"RuntimeError: ""reflection_pad2d"" not implemented for 'Half' in autocast enabled region",https://github.com/pytorch/pytorch/issues/82212,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision),scan when time allows,2022-07-26T16:48:41Z
issue,80061,pow CUDA tensor raised to CPU scalar tensor result can't backward properly,https://github.com/pytorch/pytorch/issues/80061,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2022-07-26T15:09:20Z
issue,82200,"model.to(device) takes time forever on A40-8Q, NVIDIA. cuda11.1, torch1.9.1.",https://github.com/pytorch/pytorch/issues/82200,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2022-07-26T14:07:22Z
issue,634,Feature Request: NegativeSampling and HierarchicalSoftmax loss functions,https://github.com/pytorch/pytorch/issues/634,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: loss, module: nn, module: primtorch",scan when time allows,2022-07-26T04:14:39Z
issue,35798,Dropout on sparse tensors,https://github.com/pytorch/pytorch/issues/35798,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: sparse",scan when time allows,2022-07-26T04:03:48Z
issue,76206,Update NCCL to 2.12,https://github.com/pytorch/pytorch/issues/76206,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, oncall: distributed",scan when time allows,2022-07-26T01:59:15Z
issue,82033,[PyTorch/XLA] Improve the XLA PR landing process,https://github.com/pytorch/pytorch/issues/82033,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ci, module: xla",scan when time allows,2022-07-25T23:19:53Z
issue,76785,"`addr, baddmm, dist, l1_loss` will backward fail when input tensors have different dtypes",https://github.com/pytorch/pytorch/issues/76785,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: complex",scan when time allows,2022-07-25T23:11:47Z
issue,80762,`atan2` will gradcheck fail when `other` is a tensor with `int8` dtype,https://github.com/pytorch/pytorch/issues/80762,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: edge cases",scan when time allows,2022-07-25T23:10:16Z
issue,68208,Make `torch.Tensor.view` support autograd for appropriate cases,https://github.com/pytorch/pytorch/issues/68208,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: viewing and reshaping",scan when time allows,2022-07-25T22:53:47Z
issue,74677,`check_batched_forward_grad` fails for `torch.norm` and related ops,https://github.com/pytorch/pytorch/issues/74677,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: batching, module: norms and normalization",scan when time allows,2022-07-25T22:44:31Z
issue,75586,Add ZeroTensor support for `mm`,https://github.com/pytorch/pytorch/issues/75586,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: performance",scan when time allows,2022-07-25T22:43:54Z
issue,63623,Expose a SavedTensorsHooks nn.Module for users to register saved tensors hooks,https://github.com/pytorch/pytorch/issues/63623,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: nn",scan when time allows,2022-07-25T22:22:35Z
issue,81417,Add a check to detect mutation of the inputs during backward,https://github.com/pytorch/pytorch/issues/81417,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: molly-guard",scan when time allows,2022-07-25T22:09:55Z
issue,82061,Documentation for torch.cuda.Event(blocking=True) is wrong,https://github.com/pytorch/pytorch/issues/82061,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: docs",scan when time allows,2022-07-25T21:10:13Z
issue,82132,External libraries cannot have a requirements.txt that needs to install a cpp_extension,https://github.com/pytorch/pytorch/issues/82132,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp-extensions,scan when time allows,2022-07-25T21:06:43Z
issue,82156,[Checkpoint] Support multiple unpack in saved tensor hooks,https://github.com/pytorch/pytorch/issues/82156,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: activation checkpointing,scan when time allows,2022-07-25T21:06:02Z
issue,82140,Stop manually binding sparse factory functions,https://github.com/pytorch/pytorch/issues/82140,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2022-07-25T20:57:57Z
issue,25310,Handling of packed_sequence by activation functions and linear layers,https://github.com/pytorch/pytorch/issues/25310,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: rnn",scan when time allows,2022-07-25T20:39:06Z
issue,82093,"When using libtorch v1.10.2, calling at::slow_conv_dilated3d directly returns wrong results on cpu backend",https://github.com/pytorch/pytorch/issues/82093,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: cpp",scan when time allows,2022-07-25T20:17:42Z
issue,59629,Potentially misleading note in documentation for PackedSequence,https://github.com/pytorch/pytorch/issues/59629,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: nestedtensor, module: nn, module: rnn",scan when time allows,2022-07-25T20:12:01Z
issue,81768,TestTagsCPU.test_tags__refs_constant_pad_nd_cpu_float32 flaky with dynamo & pytest,https://github.com/pytorch/pytorch/issues/81768,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2022-07-25T18:27:32Z
issue,38767,weight_decay in Adam is not an L2 Penalty,https://github.com/pytorch/pytorch/issues/38767,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2022-07-25T18:06:24Z
issue,81982,test_sparse_matmul_cpu_complex128 fails on my local copy,https://github.com/pytorch/pytorch/issues/81982,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: complex, module: sparse",scan when time allows,2022-07-25T16:20:12Z
issue,81979,test_sparse_spdiags_cpu_bool fails on my local working copy,https://github.com/pytorch/pytorch/issues/81979,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2022-07-25T16:20:10Z
issue,82095,torch.concat type hints fail for keyword argument,https://github.com/pytorch/pytorch/issues/82095,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: typing,scan when time allows,2022-07-25T15:30:12Z
issue,82088,linear.matrix_power is not composite compliant,https://github.com/pytorch/pytorch/issues/82088,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: linear algebra,scan when time allows,2022-07-25T14:12:12Z
issue,81935,implement sym_numel,https://github.com/pytorch/pytorch/issues/81935,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-07-25T12:39:43Z
issue,82041,[Misleading] The doc started using Tensorflow terminology in the document to explain how to use the Pytorch code.,https://github.com/pytorch/pytorch/issues/82041,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2022-07-24T16:35:11Z
issue,81988,Unify c10::Event and at::cuda::CUDAEvent,https://github.com/pytorch/pytorch/issues/81988,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2022-07-24T16:31:39Z
issue,36107,Allow grid_sample to accept pixel units (absolute coordinates),https://github.com/pytorch/pytorch/issues/36107,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: vision",scan when time allows,2022-07-23T13:51:57Z
issue,73141,libtorch: `DistributedRandomSampler` uses the same random order in every epoch,https://github.com/pytorch/pytorch/issues/73141,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: data,scan when time allows,2022-07-22T18:57:38Z
issue,81983,backwards compatibility ALLOWLIST is misused,https://github.com/pytorch/pytorch/issues/81983,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2022-07-22T17:17:34Z
issue,73469,torch.fx failed when tracing functions from other Libs.,https://github.com/pytorch/pytorch/issues/73469,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-07-22T09:00:24Z
issue,81959,Overloading multiple signatures for a single ref,https://github.com/pytorch/pytorch/issues/81959,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: primtorch,scan when time allows,2022-07-22T03:37:52Z
issue,81654,[packaging] Conda install missing python local version label (+cu123 or +cpu),https://github.com/pytorch/pytorch/issues/81654,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: releng,scan when time allows,2022-07-22T01:14:19Z
issue,81650,RFC: auto-generated plain Tensor argument only sparse primitives,https://github.com/pytorch/pytorch/issues/81650,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2022-07-22T01:12:02Z
issue,81801,"Rename DispatchKey Dense/Sparse/etc to DenseFunctionality/SparseFunctionality, use original name for alias",https://github.com/pytorch/pytorch/issues/81801,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: internals,scan when time allows,2022-07-22T01:10:31Z
issue,81938,"Devirtualize sym_sizes, virtualize sym_sizes_custom",https://github.com/pytorch/pytorch/issues/81938,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-07-22T01:07:04Z
issue,81943,Investigate if it's okay to throw a RuntimeError instead of TypeError here : https://github.com/pytorch/pytorch/pull/79560/files#diff-415017bcad4fa6cd6d3dfe5f6ea1caffcd7122b46b8c1e4825f7d889efc80a62R1816,https://github.com/pytorch/pytorch/issues/81943,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-07-22T01:06:52Z
issue,81937,Add more autograd tests with symints,https://github.com/pytorch/pytorch/issues/81937,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-07-22T01:05:09Z
issue,81625,[bug] the output shape from torch::mean and torch::var is different in libtorch,https://github.com/pytorch/pytorch/issues/81625,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp,scan when time allows,2022-07-22T01:02:36Z
issue,81649,Idiom for PrimTorch refs for Tensor methods,https://github.com/pytorch/pytorch/issues/81649,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: primtorch,scan when time allows,2022-07-22T00:52:24Z
issue,81932,Make sure we always redispatch through a dispatcher for all SymInt ops,https://github.com/pytorch/pytorch/issues/81932,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-07-21T22:37:03Z
issue,79138,[AUTOGRAD] support implicit reductions with SymInts in autograd.,https://github.com/pytorch/pytorch/issues/79138,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-07-21T22:32:45Z
issue,39820,I would like to install pytorch-nightly from requirements.txt,https://github.com/pytorch/pytorch/issues/39820,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-07-21T13:12:38Z
issue,8276,[Installation]: Support conda/pip install with ppc64le(power8),https://github.com/pytorch/pytorch/issues/8276,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: power",scan when time allows,2022-07-21T13:12:38Z
issue,81448,torch.utils.checkpoint optimization opportunity,https://github.com/pytorch/pytorch/issues/81448,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2022-07-21T10:10:26Z
issue,80765,Let torch.utils.tensorboard support multiprocessing,https://github.com/pytorch/pytorch/issues/80765,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: multiprocessing, module: tensorboard",scan when time allows,2022-07-21T02:09:04Z
issue,80966,[Prims+NVFuser] Aten2Aten decomp hurting performance,https://github.com/pytorch/pytorch/issues/80966,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nvfuser, module: primtorch",scan when time allows,2022-07-20T23:14:10Z
issue,81307,[Prims+NvFuser] Non-fusible ops Tracker,https://github.com/pytorch/pytorch/issues/81307,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nvfuser,scan when time allows,2022-07-20T22:53:51Z
issue,63076,Tracing models all the way down to basic functions,https://github.com/pytorch/pytorch/issues/63076,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-07-20T22:00:16Z
issue,68301,[FX] [BUG] Tensor.{inplace_method}_(.) is eliminated as dead code,https://github.com/pytorch/pytorch/issues/68301,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-07-20T21:58:46Z
issue,77332,"wrong overload resolved for `torch.mul(x, 4)` in `__torch_dispatch__`",https://github.com/pytorch/pytorch/issues/77332,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: __torch_dispatch__,scan when time allows,2022-07-20T21:52:22Z
issue,81244,CapabilityBasedPartitioner does not work correctly with mutating operations,https://github.com/pytorch/pytorch/issues/81244,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: capabilitybasedpartitioner, module: fx",scan when time allows,2022-07-20T21:48:35Z
issue,81539,CapabilityBasedPartitioner treats non-compute ops inconsistently,https://github.com/pytorch/pytorch/issues/81539,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: capabilitybasedpartitioner, module: fx, module: fx.passes",scan when time allows,2022-07-20T21:47:52Z
issue,81229,torch.fx.node.map_aggregate and torch.utils._pytree.tree_map do the same thing,https://github.com/pytorch/pytorch/issues/81229,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: fx, module: pytree",scan when time allows,2022-07-20T21:25:29Z
issue,80427,Torch fx print line number of each node,https://github.com/pytorch/pytorch/issues/80427,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-07-20T21:25:29Z
issue,80903,make_fx doesn't work with truly dynamic argument functions (e.g. fx.Interpreter),https://github.com/pytorch/pytorch/issues/80903,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-07-20T21:25:28Z
issue,80172,Torch.fx: add reporting of the name of a module not found during tracing,https://github.com/pytorch/pytorch/issues/80172,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-07-20T21:25:28Z
issue,78435,fx.Tracer with param_shapes_constant=True not working for RobertaForMaskedLM,https://github.com/pytorch/pytorch/issues/78435,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-07-20T21:25:26Z
issue,64997,FX failed to normalize op `block_diag` and `broadcast_tensors`,https://github.com/pytorch/pytorch/issues/64997,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-07-20T21:25:25Z
issue,78201,nn.Sequential causes fx.replace_pattern to not find any match.,https://github.com/pytorch/pytorch/issues/78201,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-07-20T21:25:24Z
issue,74058,torch.fx.symbolic_trace is non-deterministic,https://github.com/pytorch/pytorch/issues/74058,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-07-20T21:25:24Z
issue,76186,fx: cannot find module <built-in method matmul> when using apex.amp,https://github.com/pytorch/pytorch/issues/76186,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-07-20T21:25:21Z
issue,77016,Where is fx2trt fx to tensorrt tool?,https://github.com/pytorch/pytorch/issues/77016,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-07-20T21:25:20Z
issue,75342,`torch.fx.operator_schemas.normalize_function` is too permissive,https://github.com/pytorch/pytorch/issues/75342,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-07-20T21:25:20Z
issue,75090,torch.fx deepcopy bug,https://github.com/pytorch/pytorch/issues/75090,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-07-20T21:25:18Z
issue,74218,Cannot get op through FX when using nn.Sequential,https://github.com/pytorch/pytorch/issues/74218,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-07-20T21:25:18Z
issue,73992,[FX] Tensor constants are not lifted to attributes in direct Graph construction,https://github.com/pytorch/pytorch/issues/73992,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-07-20T21:25:18Z
issue,73304,"fx.symbolic_trace docs says that it ignores different values of concrete_args, but it doesn't",https://github.com/pytorch/pytorch/issues/73304,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-07-20T21:25:16Z
issue,71715,[FX] Support call_method in NormalizeArgs,https://github.com/pytorch/pytorch/issues/71715,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-07-20T21:25:15Z
issue,70892,Memory leak while training model generated by torch.fx.symbolic_trace() in data parallel mode,https://github.com/pytorch/pytorch/issues/70892,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-07-20T21:25:15Z
issue,70485,Cannot run FX tracer on a vision transformer model,https://github.com/pytorch/pytorch/issues/70485,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-07-20T21:25:15Z
issue,70449,[FX] Detect attribute mutation during tracing,https://github.com/pytorch/pytorch/issues/70449,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-07-20T21:25:14Z
issue,69095,[FX] `concrete_args` with unpacking breaks `fx.Interpreter`,https://github.com/pytorch/pytorch/issues/69095,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-07-20T21:25:14Z
issue,68513,torch.fx cannot trace torch.Size() properly,https://github.com/pytorch/pytorch/issues/68513,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-07-20T21:25:13Z
issue,68190,[FX][docs] Document gotcha about `training` flag,https://github.com/pytorch/pytorch/issues/68190,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-07-20T21:25:13Z
issue,67662,"""self.graph.owning_module not set for purity check"" error when trying to remove a node from torch.fx.graph",https://github.com/pytorch/pytorch/issues/67662,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-07-20T21:25:11Z
issue,67402,[Documentation] Incomplete FX module,https://github.com/pytorch/pytorch/issues/67402,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-07-20T21:25:11Z
issue,67543,[FX] Support `int` and similar variants in FX,https://github.com/pytorch/pytorch/issues/67543,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-07-20T21:25:04Z
issue,67364,[FX] FX torchvision tests fail with torchvision 0.10,https://github.com/pytorch/pytorch/issues/67364,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-07-20T21:25:03Z
issue,67030,fx graph mode quantizing error,https://github.com/pytorch/pytorch/issues/67030,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-07-20T21:25:02Z
issue,66335,`torch.fx.symbolic_trace()` loses module class information,https://github.com/pytorch/pytorch/issues/66335,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-07-20T21:25:02Z
issue,66197,`torch.fx.replace_pattern` doesn't work with untraceable wrapped functions,https://github.com/pytorch/pytorch/issues/66197,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-07-20T21:25:01Z
issue,65096,[FX] `passes.split_module` doesn't have documentation,https://github.com/pytorch/pytorch/issues/65096,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-07-20T21:25:01Z
issue,64497,FX tracing on Fairseq-based roberta fails.,https://github.com/pytorch/pytorch/issues/64497,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-07-20T21:25:00Z
issue,63815,[FX] Documentation for Graph.flatten_inps and Graph.unflatten_outs,https://github.com/pytorch/pytorch/issues/63815,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-07-20T21:24:58Z
issue,63490,[FX] Data type of target should be validated on Node construction and lint,https://github.com/pytorch/pytorch/issues/63490,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-07-20T21:24:58Z
issue,44851,fx: Python range function used on tensor shape is not symbolically traceable,https://github.com/pytorch/pytorch/issues/44851,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-07-20T21:24:31Z
issue,44049,[FX] Can't symbolically trace callsites to ScriptFunction,https://github.com/pytorch/pytorch/issues/44049,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-07-20T21:23:40Z
issue,43679,[FX] List unpacking is broken,https://github.com/pytorch/pytorch/issues/43679,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-07-20T21:23:40Z
issue,43659,Exploring supporting some cases of `__bool__` on Proxy,https://github.com/pytorch/pytorch/issues/43659,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-07-20T21:23:40Z
issue,44901,for loop can't be symbolically traced,https://github.com/pytorch/pytorch/issues/44901,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-07-20T21:23:39Z
issue,44316,[FX] Schema normalization tooling,https://github.com/pytorch/pytorch/issues/44316,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: fx, module: internals",scan when time allows,2022-07-20T21:23:39Z
issue,45534,"[FX] Make tracer return a Graph, not a GraphModule",https://github.com/pytorch/pytorch/issues/45534,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-07-20T21:23:38Z
issue,45220,FX should preserve type annotations and not break TorchScript-ability,https://github.com/pytorch/pytorch/issues/45220,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: fx, oncall: jit",scan when time allows,2022-07-20T21:23:38Z
issue,45683,[FX] Configurability of nested module representation,https://github.com/pytorch/pytorch/issues/45683,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-07-20T21:23:37Z
issue,45682,[FX] Allow customization of the behavior of Proxy,https://github.com/pytorch/pytorch/issues/45682,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-07-20T21:23:37Z
issue,47481,[fx] scripting a model with tensor list as input fails,https://github.com/pytorch/pytorch/issues/47481,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-07-20T21:23:36Z
issue,46210,[FX]Add support for len() on Proxy,https://github.com/pytorch/pytorch/issues/46210,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-07-20T21:23:36Z
issue,49265,[FX] Update `placeholder` docs to better reflect the role of `node` and `target`,https://github.com/pytorch/pytorch/issues/49265,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-07-20T21:23:35Z
issue,48813,inception_v3 is not symbolically traceable,https://github.com/pytorch/pytorch/issues/48813,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-07-20T21:23:35Z
issue,48209,[FX] Write an inference optimization pass,https://github.com/pytorch/pytorch/issues/48209,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-07-20T21:23:35Z
issue,50143,FX call_function does not work if target is a class member function,https://github.com/pytorch/pytorch/issues/50143,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-07-20T21:23:34Z
issue,49618,Support scope in fx Node,https://github.com/pytorch/pytorch/issues/49618,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-07-20T21:23:34Z
issue,50177,[FX] Static specialization of callsites (e.g. chunk),https://github.com/pytorch/pytorch/issues/50177,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-07-20T21:23:33Z
issue,52538,Helper methods not preserved during AST rewrite,https://github.com/pytorch/pytorch/issues/52538,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-07-20T21:23:32Z
issue,52022,[FX] Interpreter doesn't GC values,https://github.com/pytorch/pytorch/issues/52022,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-07-20T21:23:32Z
issue,52564,Submodules not rewritten during AST rewrite,https://github.com/pytorch/pytorch/issues/52564,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-07-20T21:23:31Z
issue,53678,[FX] Regression from 1.8: FX can no longer trace functions where the first element of an int list is a Proxy,https://github.com/pytorch/pytorch/issues/53678,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-07-20T21:23:30Z
issue,53121,[fx] Handle models with Optional inputs,https://github.com/pytorch/pytorch/issues/53121,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-07-20T21:23:30Z
issue,53794,[fx] GraphModule __init__ when provided a Module w/ inconsistent buffer handling,https://github.com/pytorch/pytorch/issues/53794,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-07-20T21:23:29Z
issue,53732,[FX] Allow customizing code generation for certain Nodes,https://github.com/pytorch/pytorch/issues/53732,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-07-20T21:23:29Z
issue,54071,[FX] Add guards in files in `test/fx` to make sure they're not run directly,https://github.com/pytorch/pytorch/issues/54071,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-07-20T21:23:28Z
issue,53935,SubgraphRewriter doesn't handle tensor constants in replacement graph,https://github.com/pytorch/pytorch/issues/53935,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-07-20T21:23:28Z
issue,54602,if cannot be symbolically traced,https://github.com/pytorch/pytorch/issues/54602,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-07-20T21:23:27Z
issue,54574,"Add OpInfo metadata for ""is_torch_functional"" and a skip for these ops in TestOperatorSignatures.test_get_torch_func_signature_exhaustive",https://github.com/pytorch/pytorch/issues/54574,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: fx, module: tests",scan when time allows,2022-07-20T21:23:27Z
issue,54503,torch.fx may have problem for annotations,https://github.com/pytorch/pytorch/issues/54503,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-07-20T21:23:27Z
issue,54899,[fx] forward references in annotations do not produce correct source code,https://github.com/pytorch/pytorch/issues/54899,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, module: fx",scan when time allows,2022-07-20T21:23:26Z
issue,54785,[FX] Cannot symbolically trace class method,https://github.com/pytorch/pytorch/issues/54785,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-07-20T21:23:26Z
issue,54913,FX tracer doesn't support returning a new instance of `object` subclass from module forward,https://github.com/pytorch/pytorch/issues/54913,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-07-20T21:23:25Z
issue,55395,[FX][testing] Test tracing into all the standard `torch.nn.functional` instances,https://github.com/pytorch/pytorch/issues/55395,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-07-20T21:23:24Z
issue,55394,[FX][testing] Test tracing into all the standard `nn.Module` instances,https://github.com/pytorch/pytorch/issues/55394,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-07-20T21:23:24Z
issue,55198,[FX] to_folder breaks with qualified type name,https://github.com/pytorch/pytorch/issues/55198,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, module: fx",scan when time allows,2022-07-20T21:23:24Z
issue,55400,[FX][testing] Add symbolic tracing tests for torchtext,https://github.com/pytorch/pytorch/issues/55400,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-07-20T21:23:23Z
issue,55397,[FX][testing] Run symbolic tracing tests over torch benchmark in CI,https://github.com/pytorch/pytorch/issues/55397,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-07-20T21:23:23Z
issue,55668,GraphModule.to_folder generates code with syntax error (imports in the middle of a class) [torch.fx],https://github.com/pytorch/pytorch/issues/55668,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-07-20T21:23:22Z
issue,55544,[FX] See if we can type-annotate attribute accesses on proxy,https://github.com/pytorch/pytorch/issues/55544,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-07-20T21:23:22Z
issue,55401,[FX][testing] Add symbolic tracing tests for torchaudio,https://github.com/pytorch/pytorch/issues/55401,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-07-20T21:23:22Z
issue,55694,symbolic tracing does not support patterns with multiple arguments?,https://github.com/pytorch/pytorch/issues/55694,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-07-20T21:23:21Z
issue,56778,can't find user's fuser method in quantization.fx.fusion_patterns.py when run fx graph qat.,https://github.com/pytorch/pytorch/issues/56778,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-07-20T21:23:20Z
issue,56688,Torch FX does not work with torchvision model,https://github.com/pytorch/pytorch/issues/56688,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-07-20T21:23:20Z
issue,56191,[FX] Issues with names of functions in user packages,https://github.com/pytorch/pytorch/issues/56191,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, module: fx",scan when time allows,2022-07-20T21:23:20Z
issue,56938,remove the dependence on node names in fx graph mode quant,https://github.com/pytorch/pytorch/issues/56938,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-07-20T21:23:19Z
issue,58051,quantize_per_tensor not symbolically traceable with FX if scale+zp are proxied,https://github.com/pytorch/pytorch/issues/58051,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-07-20T21:23:18Z
issue,57124,Allow `ScriptFunction`s to be symbolically traced,https://github.com/pytorch/pytorch/issues/57124,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-07-20T21:23:18Z
issue,58402,fx OpInfo tests don't work with nn.functional test names,https://github.com/pytorch/pytorch/issues/58402,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-07-20T21:23:17Z
issue,62021,[FX] Ability to provide autowrap function to Tracer init,https://github.com/pytorch/pytorch/issues/62021,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-07-20T21:23:14Z
issue,62554,`torch.fx.node.normalized_arguments` does not pass `normalize_to_only_use_kwargs`,https://github.com/pytorch/pytorch/issues/62554,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, module: fx",scan when time allows,2022-07-20T21:23:13Z
issue,71404,[feature request] [discussion] Generalize / recommend behavior of reset_parameters (and potentially rename),https://github.com/pytorch/pytorch/issues/71404,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: fx, module: nn",scan when time allows,2022-07-20T21:23:08Z
issue,69804,typing error in the signatures of `torch.unbind` when getting them using `torch.fx`.,https://github.com/pytorch/pytorch/issues/69804,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-07-20T21:23:08Z
issue,81682,Python operator registration API for subclasses,https://github.com/pytorch/pytorch/issues/81682,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: __torch_dispatch__, module: dispatch",scan when time allows,2022-07-20T13:03:59Z
issue,81722,Inconsistent naming convention for end of enum in DispatchKey,https://github.com/pytorch/pytorch/issues/81722,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dispatch,scan when time allows,2022-07-20T02:19:29Z
issue,80985,Nested tensor: Support Noncontiguous Buffer,https://github.com/pytorch/pytorch/issues/80985,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-07-20T00:07:11Z
issue,69685,Wrong PyTorch version in Docker image,https://github.com/pytorch/pytorch/issues/69685,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docker,scan when time allows,2022-07-19T21:36:25Z
issue,58036,[feature request] torch.as_tensor to support any object that NumPy's asarray or array can consume (consume __array_interface__),https://github.com/pytorch/pytorch/issues/58036,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: numpy,scan when time allows,2022-07-19T16:53:25Z
issue,42444,torch.random.randperm stuck in multiprocess,https://github.com/pytorch/pytorch/issues/42444,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: multiprocessing, module: random",scan when time allows,2022-07-19T12:00:12Z
issue,80986,"[Prims+NVFuser] nvFuser running into ""Tensors of type SparseTensorImpl do not have strides""",https://github.com/pytorch/pytorch/issues/80986,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nvfuser,scan when time allows,2022-07-19T04:02:56Z
issue,79164,PyTorch/XLA's DDP XLABackend is broken by upstream change,https://github.com/pytorch/pytorch/issues/79164,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ddp, module: xla, oncall: distributed",scan when time allows,2022-07-18T07:51:33Z
issue,81543,Linking pytorch libraries causes sstream behavior to be overridden globally,https://github.com/pytorch/pytorch/issues/81543,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2022-07-18T04:21:48Z
issue,81544,Torch does not build with Lazy TS disabled,https://github.com/pytorch/pytorch/issues/81544,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2022-07-15T12:15:30Z
issue,81531,CapabilityBasedPartitioner doesn't support horizontal (vertical?) fusion,https://github.com/pytorch/pytorch/issues/81531,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nvfuser,scan when time allows,2022-07-15T12:04:56Z
issue,81459,Missing corner case handling in ATen ctc_loss implementation,https://github.com/pytorch/pytorch/issues/81459,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: error checking, module: loss",scan when time allows,2022-07-15T11:58:33Z
issue,76743,Supporting torch.tensor.apply_ over GPU,https://github.com/pytorch/pytorch/issues/76743,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2022-07-14T22:06:49Z
issue,75625,[ONNX] Support unit tests in scripting that we already support in tracing,https://github.com/pytorch/pytorch/issues/75625,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2022-07-14T18:40:48Z
issue,80377,Modify _add_docstr to also set the correct module for the APIs,https://github.com/pytorch/pytorch/issues/80377,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python frontend,scan when time allows,2022-07-14T17:27:32Z
issue,34977,`conv2d` is slow with specific shapes of channels_last tensors,https://github.com/pytorch/pytorch/issues/34977,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cudnn, module: memory format, module: performance",scan when time allows,2022-07-14T13:50:00Z
issue,73537,Functionalization doesn't work when applied twice with wrapper tensor,https://github.com/pytorch/pytorch/issues/73537,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: functionalization,scan when time allows,2022-07-14T04:08:04Z
issue,81385,"""Attempted to resize a view tensor to a larger size. This is not allowed in the functionalization pass"" reported on non view tensor",https://github.com/pytorch/pytorch/issues/81385,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: functionalization,scan when time allows,2022-07-14T03:16:47Z
issue,81358,[Prims+NvFuser] Issue with aten.where.ScalarSelf,https://github.com/pytorch/pytorch/issues/81358,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nvfuser,scan when time allows,2022-07-14T03:16:09Z
issue,81428,torch.stft does not normalize non-rectangular windows correctly,https://github.com/pytorch/pytorch/issues/81428,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: complex, module: fft",scan when time allows,2022-07-14T03:09:40Z
issue,81413,torch.searchsorted error message and documentation is unclear,https://github.com/pytorch/pytorch/issues/81413,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: sorting and selection",scan when time allows,2022-07-14T03:09:11Z
issue,75147,Dataloader hangs. Potential deadlock with `set_num_threads` in worker processes?,https://github.com/pytorch/pytorch/issues/75147,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dataloader, module: deadlock",scan when time allows,2022-07-13T18:48:06Z
issue,42303,frame_length reference in pytorch.stft documentation,https://github.com/pytorch/pytorch/issues/42303,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2022-07-13T18:38:15Z
issue,81046,three typing inconsistencies on Tensor methods,https://github.com/pytorch/pytorch/issues/81046,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: typing,scan when time allows,2022-07-13T00:46:06Z
issue,52279,[docs] Official apex -> torch.cuda.amp migration guide,https://github.com/pytorch/pytorch/issues/52279,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision), module: docs",scan when time allows,2022-07-12T23:12:39Z
issue,81323,`torch.overrides.get_testing_overrides` does not function as intended for native tensor methods/operations,https://github.com/pytorch/pytorch/issues/81323,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: __torch_function__,scan when time allows,2022-07-12T20:07:28Z
issue,80921,[primTorch] `|` operator does not work with FakeTensor in _refs,https://github.com/pytorch/pytorch/issues/80921,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: meta tensors, module: primtorch",scan when time allows,2022-07-12T19:22:26Z
issue,81186,grad not preserved during copying or pickling,https://github.com/pytorch/pytorch/issues/81186,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python frontend,scan when time allows,2022-07-12T18:21:07Z
issue,80808,"The result of doing a dot product between two vectors, using einsum, depends on another unrelated vector",https://github.com/pytorch/pytorch/issues/80808,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: numerical-reproducibility,scan when time allows,2022-07-12T17:27:50Z
issue,80940,Wrong example of sliced computation in doc page Numerical Accuracy,https://github.com/pytorch/pytorch/issues/80940,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: numerical-reproducibility",scan when time allows,2022-07-12T17:23:59Z
issue,81167,move bazel files out of pytorch repo root,https://github.com/pytorch/pytorch/issues/81167,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: bazel,scan when time allows,2022-07-12T17:14:03Z
issue,81172,build libtorch with the same mkl as Matlab,https://github.com/pytorch/pytorch/issues/81172,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: mkl",scan when time allows,2022-07-12T17:13:54Z
issue,81259,Refactor linter adapters to avoid code duplication,https://github.com/pytorch/pytorch/issues/81259,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: lint,scan when time allows,2022-07-12T16:46:10Z
issue,81240,Functionalization and fake tensors failure in torture test,https://github.com/pytorch/pytorch/issues/81240,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: __torch_dispatch__, module: functionalization",scan when time allows,2022-07-12T16:44:50Z
issue,80929,"Interpolation artifacts when using nn.interpolate, trilinear mode for 3D label images",https://github.com/pytorch/pytorch/issues/80929,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: interpolation, module: nn",scan when time allows,2022-07-12T07:24:28Z
issue,74991,Tensor Subclass not preserved for Tensor subclasses created via inheritance of a TensorImpl subclass,https://github.com/pytorch/pytorch/issues/74991,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: __torch_dispatch__,scan when time allows,2022-07-11T14:10:12Z
issue,48626,Missing super().__init__() call in nn.Module,https://github.com/pytorch/pytorch/issues/48626,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2022-07-11T13:30:51Z
issue,81110,CI: Run cpu tests in parallel processes?,https://github.com/pytorch/pytorch/issues/81110,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2022-07-08T22:36:00Z
issue,51782,[discussion] [python] Generic way for generating module-wrappers for stateless parametrized functions,https://github.com/pytorch/pytorch/issues/51782,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2022-07-08T14:09:01Z
issue,80954,ExpandedWeights sometimes fail silently and doesn't compute .grad_sample attribute,https://github.com/pytorch/pytorch/issues/80954,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2022-07-07T22:10:50Z
issue,80951,ExpandedWeights can't handle modules with tied weights,https://github.com/pytorch/pytorch/issues/80951,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2022-07-07T22:09:32Z
issue,80932,Position embedding aware global circular convolution,https://github.com/pytorch/pytorch/issues/80932,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2022-07-07T21:58:01Z
issue,80187,[NVFuser] Investigate modules with bad performance relative to eager,https://github.com/pytorch/pytorch/issues/80187,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nvfuser, module: primtorch",scan when time allows,2022-07-07T20:16:01Z
issue,79202,[bug] Device dispatcher can choose CPU path for CUDA tensors.,https://github.com/pytorch/pytorch/issues/79202,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: codegen, module: dispatch",scan when time allows,2022-07-07T02:29:00Z
issue,75172,Runtime configuration to disable TORCH_WARN temporally?,https://github.com/pytorch/pytorch/issues/75172,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: internals,scan when time allows,2022-07-07T01:42:23Z
issue,79977,[CI] Do we run all cpp tests on CI?,https://github.com/pytorch/pytorch/issues/79977,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2022-07-06T23:51:47Z
issue,80230,[Prims+NVFuser] Aten2Prim refs tracking items,https://github.com/pytorch/pytorch/issues/80230,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nvfuser, module: primtorch",scan when time allows,2022-07-06T23:19:32Z
issue,79604,Some unit tests are failing,https://github.com/pytorch/pytorch/issues/79604,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docker, oncall: distributed",scan when time allows,2022-07-06T17:50:21Z
issue,79787,Improve clarity by making sharding a static nightly update,https://github.com/pytorch/pytorch/issues/79787,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2022-07-06T16:34:58Z
issue,80577,2-dimensional arange,https://github.com/pytorch/pytorch/issues/80577,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nestedtensor, module: tensor creation",scan when time allows,2022-07-06T15:39:03Z
issue,80804,`torch.renorm` gives wrong gradient for 0-valued input when `p` is even and `maxnorm=0`.,https://github.com/pytorch/pytorch/issues/80804,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: edge cases",scan when time allows,2022-07-05T15:53:14Z
issue,80803,`hardshrink` gives wrong gradient for 0 input when `lambd` is 0.,https://github.com/pytorch/pytorch/issues/80803,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: edge cases",scan when time allows,2022-07-05T15:53:02Z
issue,80551,NVFuser should extend caching to remove necessity for PrimTorch's executor to Provide Tensor Contiguity Info,https://github.com/pytorch/pytorch/issues/80551,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nvfuser, module: primtorch",scan when time allows,2022-07-05T08:59:24Z
issue,80549,Allow parameterization of Layouts,https://github.com/pytorch/pytorch/issues/80549,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: python frontend, module: sparse",scan when time allows,2022-07-04T20:33:35Z
issue,80776,`torch.inverse()` crash in cuda,https://github.com/pytorch/pytorch/issues/80776,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: linear algebra",scan when time allows,2022-07-03T03:12:21Z
issue,80605,pad_sequence and pack_sequence should support length zero tensors,https://github.com/pytorch/pytorch/issues/80605,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rnn,scan when time allows,2022-07-02T14:37:38Z
issue,80771,Anaconda is not a package manager,https://github.com/pytorch/pytorch/issues/80771,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2022-07-01T15:31:05Z
issue,80574,`bmm_sparse_cuda` kernel for `bfloat16`,https://github.com/pytorch/pytorch/issues/80574,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bfloat16, module: cuda, module: sparse",scan when time allows,2022-07-01T07:37:02Z
issue,79171,DistributedDataParallel `static_graph=True` fails to handle unused parameters,https://github.com/pytorch/pytorch/issues/79171,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ddp, oncall: distributed",scan when time allows,2022-07-01T03:56:02Z
issue,68394,"INTERNAL ASSERT FAILED at ""../aten/src/ATen/MapAllocator.cpp"":323, please report a bug to PyTorch. unable to mmap 68 bytes from file </torch_530808_7992>: Cannot allocate memory (12) ---------------------------------------------------------------------------",https://github.com/pytorch/pytorch/issues/68394,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: multiprocessing",scan when time allows,2022-06-30T03:02:41Z
issue,80439,scatter_reduce choosed indices,https://github.com/pytorch/pytorch/issues/80439,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: scatter & gather ops,scan when time allows,2022-06-29T22:45:09Z
issue,80458,Revisit OpInfo samples for nn.functional.max_poolNd,https://github.com/pytorch/pytorch/issues/80458,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: pooling, module: testing",scan when time allows,2022-06-29T22:26:07Z
issue,64692,Remove the dependency with CUDA_LIBRARIES in TorchConfig.cmake,https://github.com/pytorch/pytorch/issues/64692,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2022-06-28T22:58:41Z
issue,49406,Bitcode enable for iOS,https://github.com/pytorch/pytorch/issues/49406,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: ios",scan when time allows,2022-06-28T20:37:13Z
issue,80167,AMP step() enforce synchronization,https://github.com/pytorch/pytorch/issues/80167,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision),scan when time allows,2022-06-28T13:33:49Z
issue,80420,[DDP] output_device argument appears completely unused,https://github.com/pytorch/pytorch/issues/80420,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ddp, oncall: distributed",scan when time allows,2022-06-28T01:31:58Z
issue,80417,[c10d] Async object-based collectives,https://github.com/pytorch/pytorch/issues/80417,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: c10d, oncall: distributed",scan when time allows,2022-06-28T01:09:11Z
issue,79629,Missing the time unit in duration time of DDP logging,https://github.com/pytorch/pytorch/issues/79629,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ddp,scan when time allows,2022-06-27T23:16:10Z
issue,80204,`max_unpool` gives wrong gradient when `indices` has duplicate,https://github.com/pytorch/pytorch/issues/80204,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: edge cases, module: nn, module: pooling",scan when time allows,2022-06-27T19:21:53Z
issue,80154,[primTorch] No _refs support for torch.Tensor.requires_grad.__get__,https://github.com/pytorch/pytorch/issues/80154,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: primtorch,scan when time allows,2022-06-27T18:22:44Z
issue,78744,Unable to programmatically update models using references from model.named_modules()...requires additional parsing,https://github.com/pytorch/pytorch/issues/78744,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2022-06-27T17:32:17Z
issue,80238,test_conv_backend tests OOMing in 10.2 slow_gradcheck CI,https://github.com/pytorch/pytorch/issues/80238,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ci, module: convolution, module: nn",scan when time allows,2022-06-27T17:18:23Z
issue,80022,memory leaking when doing all_to_all_single communication,https://github.com/pytorch/pytorch/issues/80022,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: c10d, oncall: distributed",scan when time allows,2022-06-27T17:00:37Z
issue,80189,[NVFuser] Investigate models without any fusion groups found,https://github.com/pytorch/pytorch/issues/80189,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nvfuser,scan when time allows,2022-06-27T13:17:42Z
issue,80206,`soft_margin_loss` gives wrong gradient when `target` with dtype uint8,https://github.com/pytorch/pytorch/issues/80206,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: loss, module: nn",scan when time allows,2022-06-27T13:15:07Z
issue,80221,OpInfos for torch.ops.aten operations,https://github.com/pytorch/pytorch/issues/80221,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tests,scan when time allows,2022-06-27T13:14:31Z
issue,80226,Support tensor subclasses as `UninitializedParameter`s,https://github.com/pytorch/pytorch/issues/80226,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: lazy, module: nn",scan when time allows,2022-06-27T13:14:14Z
issue,80231,[Prims+NVFuser] Supports 0-sized inputs,https://github.com/pytorch/pytorch/issues/80231,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nvfuser, module: primtorch",scan when time allows,2022-06-27T13:00:11Z
issue,79351,[bazel] ability to run gpu tests on gpu machines in RBE,https://github.com/pytorch/pytorch/issues/79351,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: bazel,scan when time allows,2022-06-25T17:36:54Z
issue,80242,Write some torch.distributed.nn.* tests for the new dispatcher passable ops,https://github.com/pytorch/pytorch/issues/80242,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2022-06-24T20:34:53Z
issue,80241,Change c10d APIs in ProcessGroup to accept const std::vector<at::Tensor>&,https://github.com/pytorch/pytorch/issues/80241,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2022-06-24T20:32:25Z
issue,80134,[feature request] Add support for a custom DatasetFetcher in DataLoader,https://github.com/pytorch/pytorch/issues/80134,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: data, module: dataloader",scan when time allows,2022-06-24T15:37:24Z
issue,80208,F.binary_cross_entropy_with_logits unexpected behaviour,https://github.com/pytorch/pytorch/issues/80208,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: loss, module: nn",scan when time allows,2022-06-24T14:44:44Z
issue,80132,Expose more MAGMA backends for solve_triangular,https://github.com/pytorch/pytorch/issues/80132,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: linear algebra, module: magma",scan when time allows,2022-06-24T13:21:58Z
issue,79847,Multi-node training meets unknown error,https://github.com/pytorch/pytorch/issues/79847,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2022-06-23T23:54:36Z
issue,80157,Elliptic Functions and Integrals,https://github.com/pytorch/pytorch/issues/80157,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: special,scan when time allows,2022-06-23T18:51:29Z
issue,80118,"Allow a user provided ""test name - test time"" mapping file work with pytorch's test sharding mechanism",https://github.com/pytorch/pytorch/issues/80118,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2022-06-23T18:32:59Z
issue,80152,Orthogonal Polynomials,https://github.com/pytorch/pytorch/issues/80152,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: special,scan when time allows,2022-06-23T18:32:51Z
issue,80080,Complex support in DDP,https://github.com/pytorch/pytorch/issues/80080,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ddp, oncall: distributed",scan when time allows,2022-06-23T15:14:49Z
issue,76838,Windows CUDA TTS tracking task,https://github.com/pytorch/pytorch/issues/76838,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2022-06-22T23:29:08Z
issue,79464,Iteration # 1-offset in DDP logging,https://github.com/pytorch/pytorch/issues/79464,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ddp, oncall: distributed",scan when time allows,2022-06-22T17:29:35Z
issue,59592,Build failed with TBB,https://github.com/pytorch/pytorch/issues/59592,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: tbb",scan when time allows,2022-06-22T14:27:39Z
issue,42838,"RuntimeError: ""triangular_solve_cuda"" not implemented for 'Half'",https://github.com/pytorch/pytorch/issues/42838,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision), module: cuda, module: half",scan when time allows,2022-06-22T13:06:30Z
issue,50723,[PyTorch Mobile] Can't use Vulkan backend,https://github.com/pytorch/pytorch/issues/50723,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: android, module: vulkan, oncall: mobile",scan when time allows,2022-06-22T06:46:23Z
issue,79997,Comprehensive documentation for Tensor indexing?,https://github.com/pytorch/pytorch/issues/79997,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: advanced indexing, module: docs",scan when time allows,2022-06-22T06:26:53Z
issue,76656,Clarify dependency on NumPy (related to maskedtensor?),https://github.com/pytorch/pytorch/issues/76656,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-06-22T00:55:28Z
issue,77900,Floating point exception in slow_conv3d,https://github.com/pytorch/pytorch/issues/77900,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: edge cases",scan when time allows,2022-06-21T23:26:16Z
issue,77901,Heap corruption in slow_conv_transpose3d,https://github.com/pytorch/pytorch/issues/77901,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: edge cases",scan when time allows,2022-06-21T23:26:02Z
issue,54035,Web support?,https://github.com/pytorch/pytorch/issues/54035,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: language binding,scan when time allows,2022-06-21T17:33:13Z
issue,25091,JavaScript (Web Assembly) target for trained models,https://github.com/pytorch/pytorch/issues/25091,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-06-21T17:32:46Z
issue,79853,[bazel] [ci] `//:lazy_tests` Could not run 'aten::mul.Tensor' with arguments from the 'Lazy' backend,https://github.com/pytorch/pytorch/issues/79853,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bazel, module: cuda, module: lazy",scan when time allows,2022-06-21T13:41:15Z
issue,79867,All {view}_scatter variants should support all (or most) dtypes,https://github.com/pytorch/pytorch/issues/79867,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: functionalization,scan when time allows,2022-06-21T13:39:39Z
issue,79888,`CosineAnnealingWarmRestarts` does not update parameters added with `add_param_group`,https://github.com/pytorch/pytorch/issues/79888,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: lrscheduler,scan when time allows,2022-06-21T13:33:12Z
issue,79893,Write lint for isGreen,https://github.com/pytorch/pytorch/issues/79893,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2022-06-21T13:32:25Z
issue,79895,Modify update-viable-strict GHA to use internal version of checkout,https://github.com/pytorch/pytorch/issues/79895,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2022-06-21T13:32:02Z
issue,79848,Automatically calculate output_shape of sequential model (or any other fCNN),https://github.com/pytorch/pytorch/issues/79848,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: meta tensors,scan when time allows,2022-06-21T13:27:59Z
issue,79788,Parameter.__deepcopy__ doesn't preserve view relationships,https://github.com/pytorch/pytorch/issues/79788,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: nn",scan when time allows,2022-06-21T13:17:00Z
issue,79542,Unable to use a parameter with torch.sparse_coo layout with DDP,https://github.com/pytorch/pytorch/issues/79542,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ddp, module: sparse, oncall: distributed",scan when time allows,2022-06-21T00:19:09Z
issue,78367,Split up and reorganize RPC tests,https://github.com/pytorch/pytorch/issues/78367,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: rpc, oncall: distributed",scan when time allows,2022-06-20T23:49:06Z
issue,62094,"[FR] Support for non-Parameter, non-Buffer objects in state_dict",https://github.com/pytorch/pytorch/issues/62094,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2022-06-20T21:18:59Z
issue,79421,Mismatch in clang toolchain lead to binary incompatibilities on M1 between torch and torchvision,https://github.com/pytorch/pytorch/issues/79421,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: arm, module: binaries, module: ci, module: macos",scan when time allows,2022-06-20T17:41:36Z
issue,77538,Failed to run on iOS - Couldn't find an operator for `aten::conv1d`,https://github.com/pytorch/pytorch/issues/77538,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ios,scan when time allows,2022-06-20T15:47:47Z
issue,36560,Allow `__array__` to automatically detach and move to CPU,https://github.com/pytorch/pytorch/issues/36560,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: numpy",scan when time allows,2022-06-20T13:52:05Z
issue,77256,CrossEntropyLoss computes SoftMax always across the second dimension,https://github.com/pytorch/pytorch/issues/77256,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: loss, module: nn",scan when time allows,2022-06-20T09:31:01Z
issue,79705,gradgradcheck fails for torch.native_layer_norm,https://github.com/pytorch/pytorch/issues/79705,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: double backwards",scan when time allows,2022-06-19T17:27:45Z
issue,79703,Float and double tensors randomly initialized with the same seed get different values for size >= 16,https://github.com/pytorch/pytorch/issues/79703,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: random,scan when time allows,2022-06-19T17:27:37Z
issue,79647,tensorboard SummaryWriter.add_graph fails when model uses empty tuples,https://github.com/pytorch/pytorch/issues/79647,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tensorboard,scan when time allows,2022-06-19T17:26:09Z
issue,79336,Batches are being duplicated from go http call,https://github.com/pytorch/pytorch/issues/79336,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-06-17T20:56:51Z
issue,60028,RReLU and PReLU do not propagate input strides to outputs,https://github.com/pytorch/pytorch/issues/60028,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: memory format, module: nn",scan when time allows,2022-06-17T18:30:50Z
issue,76232,Observing a strange behavior - Row parallelism,https://github.com/pytorch/pytorch/issues/76232,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: nn, module: numerical-stability",scan when time allows,2022-06-17T14:34:23Z
issue,79359,[build] No documented way to install C++ binaries for pure-python development of pytorch,https://github.com/pytorch/pytorch/issues/79359,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: build",scan when time allows,2022-06-16T18:18:23Z
issue,79018,Random functions should infer device from user-specified Generator,https://github.com/pytorch/pytorch/issues/79018,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: random,scan when time allows,2022-06-16T14:59:53Z
issue,79592,[LTC] Introduce a `MetricsReport` python binding and allow backend to add their report as string,https://github.com/pytorch/pytorch/issues/79592,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-06-16T04:52:42Z
issue,79606,PyTorch Preview (Nightly) version number does not comply with Conda conventions,https://github.com/pytorch/pytorch/issues/79606,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: binaries,scan when time allows,2022-06-15T18:52:34Z
issue,79563,__torch__dispatch does not return new output in inplace function,https://github.com/pytorch/pytorch/issues/79563,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: __torch_dispatch__,scan when time allows,2022-06-15T14:28:28Z
issue,79425,API for accessing SymIntNode mandates refcount bump even when it is unnecessary,https://github.com/pytorch/pytorch/issues/79425,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-06-15T05:40:08Z
issue,79395,SymInt equality tests are unsound,https://github.com/pytorch/pytorch/issues/79395,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-06-15T05:40:01Z
issue,79387,caffe2_nvrtc is produced even when it won't be used,https://github.com/pytorch/pytorch/issues/79387,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: selective build",scan when time allows,2022-06-15T05:23:32Z
issue,79355,[bazel] build spams warnings,https://github.com/pytorch/pytorch/issues/79355,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: bazel,scan when time allows,2022-06-15T05:20:10Z
issue,79453,[LTC] Make `torch::lazy::BackendImplInterface::ExecuteComputation` takes `ComputationPtr` instead of `Computation`,https://github.com/pytorch/pytorch/issues/79453,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-06-14T18:33:45Z
issue,79349,PyTorch get positive log_prob of a multivariate normal distribution,https://github.com/pytorch/pytorch/issues/79349,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-06-14T18:07:43Z
issue,79407,Conda enviroment,https://github.com/pytorch/pytorch/issues/79407,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-06-14T18:07:24Z
issue,79518,"Display a ""reference"" link for ops that points to primTorch implementations",https://github.com/pytorch/pytorch/issues/79518,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: primtorch",scan when time allows,2022-06-14T18:03:52Z
issue,79469,Add a new _broadcast_coalesced op for DDP,https://github.com/pytorch/pytorch/issues/79469,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-06-13T22:44:22Z
issue,79468,Ensure the guards in distributed_c10d.py wrappers get executed in the replay of the graph,https://github.com/pytorch/pytorch/issues/79468,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2022-06-13T22:43:24Z
issue,79467,Add autograd support for dispatch passable c10d ops,https://github.com/pytorch/pytorch/issues/79467,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2022-06-13T22:42:09Z
issue,79222,scripted fft Convolutions are faster than nn.Conv1d with large kernels,https://github.com/pytorch/pytorch/issues/79222,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: cuda, module: performance",scan when time allows,2022-06-13T20:59:41Z
issue,78274,Memory allocation errors when attempting to initialize a large number of small feed-forward networks in RAM with shared memory despite having enough memory,https://github.com/pytorch/pytorch/issues/78274,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: memory usage,scan when time allows,2022-06-13T20:54:20Z
issue,79244,[NVFuser] bad performance on pyhpc_equation_of_state,https://github.com/pytorch/pytorch/issues/79244,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nvfuser,scan when time allows,2022-06-13T18:43:23Z
issue,62545,Incosistency with args for `nn.functional.max_poolNd` vs `nn.MaxPoolNd` functions,https://github.com/pytorch/pytorch/issues/62545,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: pooling",scan when time allows,2022-06-13T18:40:17Z
issue,79382,torch failure to open libcuda.so.1 on macOS,https://github.com/pytorch/pytorch/issues/79382,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: macos,scan when time allows,2022-06-13T18:19:39Z
issue,79083,CI workflow creates too many tags in RSS feed,https://github.com/pytorch/pytorch/issues/79083,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2022-06-13T17:47:28Z
issue,73113,PyTorch fails to compile on gcc 11.2 due to breakpad,https://github.com/pytorch/pytorch/issues/73113,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2022-06-12T12:15:56Z
issue,31945,"Function request: logerfc, logerfcx special functions",https://github.com/pytorch/pytorch/issues/31945,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: numerical-stability, module: numpy, module: special",scan when time allows,2022-06-11T23:54:47Z
issue,47916,Hash mismatch for METADATA file,https://github.com/pytorch/pytorch/issues/47916,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: binaries,scan when time allows,2022-06-11T16:42:58Z
issue,30532,Version 1.3 no longer supporting Tesla K40m?,https://github.com/pytorch/pytorch/issues/30532,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: cuda, module: docs",scan when time allows,2022-06-11T11:15:57Z
issue,78878,Improving clarity in the docs of different losses,https://github.com/pytorch/pytorch/issues/78878,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: nn, module: python frontend",scan when time allows,2022-06-10T21:10:22Z
issue,8921,[feature request] More methods for PackedSequence,https://github.com/pytorch/pytorch/issues/8921,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2022-06-10T15:51:51Z
issue,79261,[NVFuser] bad performance on pyhpc_isoneutral_mixing,https://github.com/pytorch/pytorch/issues/79261,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nvfuser,scan when time allows,2022-06-10T15:01:37Z
issue,79246,[NVFuser] bad performance on mobilenet_v2 and mobilenet_v3_large,https://github.com/pytorch/pytorch/issues/79246,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nvfuser,scan when time allows,2022-06-10T14:59:33Z
issue,56901,Static prebuilt libraries contain shared libraries,https://github.com/pytorch/pytorch/issues/56901,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: static linking",scan when time allows,2022-06-10T06:48:48Z
issue,79089,Improve PrimTorch testing for view consistency.,https://github.com/pytorch/pytorch/issues/79089,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: primtorch, module: tests, module: viewing and reshaping",scan when time allows,2022-06-09T23:37:32Z
issue,79177,_make_elementwise_unary_reference and other function factories in torch._refs don't set __name__ correctly,https://github.com/pytorch/pytorch/issues/79177,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: primtorch,scan when time allows,2022-06-09T23:36:45Z
issue,78987,DataLoader leaking resources?,https://github.com/pytorch/pytorch/issues/78987,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2022-06-09T23:08:27Z
issue,79032,Add a test that shows that lazy_ir reuse breaks SizeNodes,https://github.com/pytorch/pytorch/issues/79032,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-06-09T22:02:45Z
issue,79120,Hide or fuse TupleConstruct / TupleUnpack from tensorboard graph,https://github.com/pytorch/pytorch/issues/79120,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tensorboard,scan when time allows,2022-06-09T00:11:12Z
issue,79130,torchvision.models.mobilenetv3 can't save pre-trained model to custom dir?,https://github.com/pytorch/pytorch/issues/79130,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: vision,scan when time allows,2022-06-09T00:04:43Z
issue,79118,About the source code of,https://github.com/pytorch/pytorch/issues/79118,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-06-09T00:04:26Z
issue,52260,Unlucky batch->device split by DataParallel causes an exception,https://github.com/pytorch/pytorch/issues/52260,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: data parallel,scan when time allows,2022-06-08T18:21:25Z
issue,79137,[AUTOGRAD] codegen to use sym_sizes for ops w/ symint overloads in derivative formulas,https://github.com/pytorch/pytorch/issues/79137,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-06-08T18:05:14Z
issue,79021,Building PyTorch from Source with BUILD_LAZY_TS_BACKEND_ON,https://github.com/pytorch/pytorch/issues/79021,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2022-06-07T22:32:03Z
issue,79073,"Multi-node, Multi-GPU set up tutorial for Slurm cluster",https://github.com/pytorch/pytorch/issues/79073,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2022-06-07T22:31:58Z
issue,79053,[MetaIssue] Propagating SymInts through Autograd,https://github.com/pytorch/pytorch/issues/79053,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-06-07T20:19:29Z
issue,79050,[MetaIssue] Investigate if we should be reusing primtorch formulas for `is_dynamic`,https://github.com/pytorch/pytorch/issues/79050,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-06-07T20:17:14Z
issue,79052,Mirror and implement `SymbolicIntNode` API for `SymInt` so we can trace in C++,https://github.com/pytorch/pytorch/issues/79052,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-06-07T20:04:22Z
issue,79031,Implement SymbolicIntNode interface for lazy (i.e. lazy::SymbolicIntNode),https://github.com/pytorch/pytorch/issues/79031,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-06-07T17:41:02Z
issue,79030,Devirtualize `sym_sizes`. It still has to work for python tensor subclasses and LTC/Xla,https://github.com/pytorch/pytorch/issues/79030,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-06-07T17:31:53Z
issue,79020,"When setting sizes and strides on a tensor subclass in `THPVariable_make_wrapper_subclass`, also make offset symbolic",https://github.com/pytorch/pytorch/issues/79020,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: lazy,scan when time allows,2022-06-07T17:19:29Z
issue,79013,Multi30k can't be downloaded the destination domain can't be reached,https://github.com/pytorch/pytorch/issues/79013,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-06-07T13:47:09Z
issue,79004,Doc on index of CPU Device seems wrong,https://github.com/pytorch/pytorch/issues/79004,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp,scan when time allows,2022-06-07T13:30:36Z
issue,78961,[forwardAD] torch.no_grad has no effect under forward_ad,https://github.com/pytorch/pytorch/issues/78961,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: forward ad,scan when time allows,2022-06-07T13:27:52Z
issue,78876,Remove const from function return type if returning const value,https://github.com/pytorch/pytorch/issues/78876,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp,scan when time allows,2022-06-07T13:27:44Z
issue,78917,LambdaLR changes the learning rate in an undesired way,https://github.com/pytorch/pytorch/issues/78917,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: lrscheduler,scan when time allows,2022-06-07T03:14:56Z
issue,78921,Do we really need sampler for IterableDataset?,https://github.com/pytorch/pytorch/issues/78921,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dataloader, module: performance",scan when time allows,2022-06-06T21:41:52Z
issue,78884,Abnormal GPU memory usage when using CUDA tensors with multiprocessing,https://github.com/pytorch/pytorch/issues/78884,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: jiterator, module: memory usage, module: multiprocessing",scan when time allows,2022-06-06T19:04:22Z
issue,78882,Cannot build master on AWS cluster: error: ‘__fatDeviceText’ was not declared in this scope,https://github.com/pytorch/pytorch/issues/78882,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cuda",scan when time allows,2022-06-06T19:00:21Z
issue,78581,RFC: Improve the performance and usability of linear algebra on CUDA devices,https://github.com/pytorch/pytorch/issues/78581,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: linear algebra, module: magma",scan when time allows,2022-06-06T17:23:56Z
issue,78808,Feature request: Integer system decompositions,https://github.com/pytorch/pytorch/issues/78808,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: linear algebra,scan when time allows,2022-06-06T14:51:56Z
issue,5434,RuntimeError: $ Torch: not enough memory: you tried to allocate 72GB. Buy new RAM!,https://github.com/pytorch/pytorch/issues/5434,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: memory usage,scan when time allows,2022-06-06T08:01:09Z
issue,77200,[RFC] Allow device override during Tensor unpickling without torch.load,https://github.com/pytorch/pytorch/issues/77200,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: pickle, module: serialization",scan when time allows,2022-06-05T09:28:44Z
issue,78871,[RecordFunction] Hold a durable schema reference,https://github.com/pytorch/pytorch/issues/78871,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: profiler,scan when time allows,2022-06-04T03:48:01Z
issue,78845,"[Modes] no_dispatch is not the same as DisableTorchFunction, causing differences in modes",https://github.com/pytorch/pytorch/issues/78845,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: __torch_dispatch__, module: __torch_function__",scan when time allows,2022-06-03T23:22:45Z
issue,78737,New c10 constants,https://github.com/pytorch/pytorch/issues/78737,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: internals,scan when time allows,2022-06-03T22:49:02Z
issue,78800,Efficiency of unary operations on CPU for large tensors,https://github.com/pytorch/pytorch/issues/78800,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: performance,scan when time allows,2022-06-03T22:08:10Z
issue,78761,Extend tag testing for aliases,https://github.com/pytorch/pytorch/issues/78761,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: testing,scan when time allows,2022-06-03T22:04:40Z
issue,78742,Add a gallery of examples with sphinx-gallery,https://github.com/pytorch/pytorch/issues/78742,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2022-06-03T22:00:17Z
issue,78741,Test approximation and numerical stability of numerical operators,https://github.com/pytorch/pytorch/issues/78741,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: numerical-stability, module: testing",scan when time allows,2022-06-03T21:59:29Z
issue,78708,BuildExtension does not choose correct CUDA installation,https://github.com/pytorch/pytorch/issues/78708,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp-extensions, module: cuda",scan when time allows,2022-06-03T21:55:24Z
issue,63140,[documentation] torch.distributed.elastic: illustrate how to write load_checkpoint and save_checkpoint in Train Script,https://github.com/pytorch/pytorch/issues/63140,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: elastic, oncall: r2p",scan when time allows,2022-06-03T20:47:29Z
issue,78249,`max_unpool2d` is not deterministic,https://github.com/pytorch/pytorch/issues/78249,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: determinism, module: nn, module: numerical-stability, module: pooling",scan when time allows,2022-06-03T18:58:50Z
issue,78786,Deprecate hardtanh type promotion behavior.,https://github.com/pytorch/pytorch/issues/78786,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: primtorch",scan when time allows,2022-06-03T16:05:17Z
issue,73531,Mention docker build process in RELEASE.md and automate building those for release,https://github.com/pytorch/pytorch/issues/73531,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ci, module: docker",scan when time allows,2022-06-03T09:10:10Z
issue,78738,[primTorch] Sensible Error Messages,https://github.com/pytorch/pytorch/issues/78738,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: primtorch,scan when time allows,2022-06-03T03:13:12Z
issue,38138,"TorchScript to support == None, != None",https://github.com/pytorch/pytorch/issues/38138,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2022-06-03T01:55:37Z
issue,49846,torch init_process_group always hangs,https://github.com/pytorch/pytorch/issues/49846,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2022-06-02T19:52:13Z
issue,78729,[Better Engineering] Make OpInfo-based test failures easy to reproduce,https://github.com/pytorch/pytorch/issues/78729,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tests,scan when time allows,2022-06-02T19:11:41Z
issue,78743,Expose docs from the yaml for each torch.Tag in Python,https://github.com/pytorch/pytorch/issues/78743,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dispatch, module: docs, module: library",scan when time allows,2022-06-02T17:18:44Z
issue,78634,Debug job does not build in debug mode,https://github.com/pytorch/pytorch/issues/78634,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2022-06-02T00:16:56Z
issue,78624,linalg.pinv_singular tests are slow,https://github.com/pytorch/pytorch/issues/78624,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: linear algebra, module: tests",scan when time allows,2022-06-02T00:16:38Z
issue,77050,ENORMOUS OVERHEAD from mp.get_context('spawn'),https://github.com/pytorch/pytorch/issues/77050,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: multiprocessing, module: performance, module: power",scan when time allows,2022-06-01T21:40:00Z
issue,76920,Avoid Self-loops on Module Creation,https://github.com/pytorch/pytorch/issues/76920,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2022-06-01T18:21:20Z
issue,78519,test_python_dispatch fails on DEBUG=1,https://github.com/pytorch/pytorch/issues/78519,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dispatch,scan when time allows,2022-06-01T18:21:00Z
issue,77515,Inplace Bool API + `sum` will trigger INTERNAL ASSERT FAILED,https://github.com/pytorch/pytorch/issues/77515,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: edge cases",scan when time allows,2022-06-01T18:00:08Z
issue,77963,`logaddexp2` fails to backward,https://github.com/pytorch/pytorch/issues/77963,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: edge cases",scan when time allows,2022-06-01T17:35:15Z
issue,78422,Permutation of Sparse Tensor,https://github.com/pytorch/pytorch/issues/78422,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2022-06-01T17:20:44Z
issue,78413,torch.angle differs from np.angle for -0.,https://github.com/pytorch/pytorch/issues/78413,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: numpy, module: primtorch",scan when time allows,2022-06-01T08:11:22Z
issue,78530,LibTorch cannot be used without nvcc,https://github.com/pytorch/pytorch/issues/78530,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2022-06-01T04:29:01Z
issue,77052,Building from source results in broken __version__,https://github.com/pytorch/pytorch/issues/77052,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: build",scan when time allows,2022-06-01T02:44:51Z
issue,77808,`Could not start gRPC server` flakiness in XLA tests,https://github.com/pytorch/pytorch/issues/77808,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: xla,scan when time allows,2022-05-31T21:02:50Z
issue,78507,Parallel execution of multiple unrelated statements written sequentially,https://github.com/pytorch/pytorch/issues/78507,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-05-31T20:39:19Z
issue,78450,"Softmax, LogSoftmax are over parameterized",https://github.com/pytorch/pytorch/issues/78450,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2022-05-31T20:23:28Z
issue,78346,`gradcheck` fails for `torch.distribution.transform` APIs in forward mode,https://github.com/pytorch/pytorch/issues/78346,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: distributions, module: forward ad",scan when time allows,2022-05-31T19:06:17Z
issue,75926,Disable TracerWarnings on NVFuser opinfo tests,https://github.com/pytorch/pytorch/issues/75926,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, oncall: jit",scan when time allows,2022-05-31T17:39:56Z
issue,78047,DDP multi host with single GPU each.,https://github.com/pytorch/pytorch/issues/78047,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2022-05-31T17:26:59Z
issue,78513,clear input shape declaration on pytorch model inputs and outputs,https://github.com/pytorch/pytorch/issues/78513,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: python frontend, module: shape checking",scan when time allows,2022-05-31T16:14:33Z
issue,78261,"pytorch-android-lite use its own libfbjni.so, which is not compatible with any other version at all..",https://github.com/pytorch/pytorch/issues/78261,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: android, module: binaries",scan when time allows,2022-05-31T15:38:29Z
issue,78518,Exponentiating floating number with cuda tensor is slow,https://github.com/pytorch/pytorch/issues/78518,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2022-05-31T15:37:29Z
issue,74143,Tracing model parameter shapes without instantiating the model parameters,https://github.com/pytorch/pytorch/issues/74143,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2022-05-31T15:36:57Z
issue,78075,torch.multiprocessing.spawn raise PicklingError inside a decorator,https://github.com/pytorch/pytorch/issues/78075,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: multiprocessing, module: serialization",scan when time allows,2022-05-31T00:50:28Z
issue,19143,Support memoryview() method on torch.Tensor,https://github.com/pytorch/pytorch/issues/19143,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: numpy,scan when time allows,2022-05-30T14:35:08Z
issue,43546,"NCCL WARN Your program may be hanging, this may be caused by a collective mismatch around rank 2. Please check your collective calls at and around this rank. You can use NCCL_DEBUG=INFO and NCCL_DEBUG_SUBSYS=COLL to see the collective logs",https://github.com/pytorch/pytorch/issues/43546,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: nccl, oncall: distributed",scan when time allows,2022-05-30T12:33:22Z
issue,33385,Illegal instruction: 4 - OSX 10.13.6 install from source,https://github.com/pytorch/pytorch/issues/33385,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: macos",scan when time allows,2022-05-26T16:52:17Z
issue,78253,Any plan to add Noam scheduling?,https://github.com/pytorch/pytorch/issues/78253,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: lrscheduler,scan when time allows,2022-05-26T02:56:28Z
issue,46803,About 2 minor bug fixes on CUDA macOSX 10.13.6,https://github.com/pytorch/pytorch/issues/46803,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cuda, module: macos, module: nnpack",scan when time allows,2022-05-26T00:05:50Z
issue,78170,[chalf] reference_testing: low quality test for fast growing ops,https://github.com/pytorch/pytorch/issues/78170,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: complex, module: half",scan when time allows,2022-05-25T21:11:16Z
issue,78153,`pack_sequence` crash,https://github.com/pytorch/pytorch/issues/78153,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: edge cases,scan when time allows,2022-05-25T20:35:43Z
issue,78151,`ctc_loss` will backward crash,https://github.com/pytorch/pytorch/issues/78151,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: edge cases",scan when time allows,2022-05-25T20:35:19Z
issue,78157,[Optimizer Overlap] Custom optimizer registration,https://github.com/pytorch/pytorch/issues/78157,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ddp, oncall: distributed",scan when time allows,2022-05-25T19:06:45Z
issue,78159,[Optimizer Overlap] Parameter group support,https://github.com/pytorch/pytorch/issues/78159,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ddp, oncall: distributed",scan when time allows,2022-05-25T19:04:50Z
issue,78255,Floating point exception in _conv_depthwise2d,https://github.com/pytorch/pytorch/issues/78255,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: edge cases,scan when time allows,2022-05-25T18:43:16Z
issue,78122,Segfault in bincount,https://github.com/pytorch/pytorch/issues/78122,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: edge cases,scan when time allows,2022-05-25T18:37:39Z
issue,78185,test_to (__main__.TestTorch) fails with multiple gpus,https://github.com/pytorch/pytorch/issues/78185,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: multi-gpu,scan when time allows,2022-05-25T18:32:33Z
issue,78131,Segfault in _pad_packed_sequence,https://github.com/pytorch/pytorch/issues/78131,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: edge cases,scan when time allows,2022-05-25T18:30:40Z
issue,78129,Segfault in _embedding_bag_forward_only,https://github.com/pytorch/pytorch/issues/78129,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: edge cases,scan when time allows,2022-05-25T18:30:39Z
issue,78127,Segfault in torch._C._nn.reflection_pad2d,https://github.com/pytorch/pytorch/issues/78127,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: edge cases,scan when time allows,2022-05-25T18:30:39Z
issue,78126,Segfault in max_unpool3d,https://github.com/pytorch/pytorch/issues/78126,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: edge cases,scan when time allows,2022-05-25T18:30:38Z
issue,78125,Segfault in grid_sampler_3d,https://github.com/pytorch/pytorch/issues/78125,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: edge cases,scan when time allows,2022-05-25T18:30:38Z
issue,78130,Segfault in _grid_sampler_2d_cpu_fallback,https://github.com/pytorch/pytorch/issues/78130,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: edge cases,scan when time allows,2022-05-25T18:29:46Z
issue,78128,Segfault in torch._C._nn.thnn_conv2d,https://github.com/pytorch/pytorch/issues/78128,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: edge cases,scan when time allows,2022-05-25T18:29:25Z
issue,78260,[CI] Detect when tests are no longer running from CI,https://github.com/pytorch/pytorch/issues/78260,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2022-05-25T15:25:50Z
issue,77736,TimeSeriesDataset retrieve columns,https://github.com/pytorch/pytorch/issues/77736,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-05-25T09:44:59Z
issue,77463,conda CPU installation for LTS fails with UnsatisfiableError,https://github.com/pytorch/pytorch/issues/77463,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: lts,scan when time allows,2022-05-24T21:45:56Z
issue,77839,BUG: reference count leak when using `THPLayout_New` and `THPMemoryFormat_New` (static analyzer reports),https://github.com/pytorch/pytorch/issues/77839,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: memory usage, module: python frontend",scan when time allows,2022-05-23T17:21:32Z
issue,77946,[failing test] test_foreach::test_binary_op_scalarlist_fastpath,https://github.com/pytorch/pytorch/issues/77946,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-05-23T16:43:51Z
issue,78070,[primTorch] Meta function for item creates a dummy value,https://github.com/pytorch/pytorch/issues/78070,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: primtorch,scan when time allows,2022-05-23T14:31:01Z
issue,77552,Functional Jacobian does not work with Torchdiffeq,https://github.com/pytorch/pytorch/issues/77552,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2022-05-23T04:07:54Z
issue,78071,[primTorch] item prim can't return a bool properly,https://github.com/pytorch/pytorch/issues/78071,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: primtorch,scan when time allows,2022-05-23T04:01:36Z
issue,77951,`topk` returns different results with the same input twice in cuda,https://github.com/pytorch/pytorch/issues/77951,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: python frontend",scan when time allows,2022-05-23T03:41:53Z
issue,77801,`torch.utils.benchmark.examples.blas_compare` can not be parsed by Python-3.7 runtime,https://github.com/pytorch/pytorch/issues/77801,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: benchmark,scan when time allows,2022-05-23T03:39:10Z
issue,77899,Floating point exception in native_channel_shuffle,https://github.com/pytorch/pytorch/issues/77899,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-05-23T03:38:33Z
issue,77894,Floating point exception in channel_shuffle,https://github.com/pytorch/pytorch/issues/77894,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-05-23T03:37:01Z
issue,77838,Sporadic convolution error with dilation=0,https://github.com/pytorch/pytorch/issues/77838,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: convolution,scan when time allows,2022-05-23T03:27:19Z
issue,77844,Conv2D with large different number of input and output channels gives a CUDNN_STATUS_INTERNAL_ERROR,https://github.com/pytorch/pytorch/issues/77844,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cudnn,scan when time allows,2022-05-23T03:26:53Z
issue,77973,"forward-mode support for ""logically composite"" operators",https://github.com/pytorch/pytorch/issues/77973,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: derivatives, module: forward ad",scan when time allows,2022-05-22T22:13:19Z
issue,77981,Override sym_sizes to create LTC IR for SymIntNode,https://github.com/pytorch/pytorch/issues/77981,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-05-22T22:13:02Z
issue,78061,Python memory allocator called without holding the GIL when running torchrun under Python debug version,https://github.com/pytorch/pytorch/issues/78061,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: r2p,scan when time allows,2022-05-22T22:12:09Z
issue,78053,toleranceOverride should override atol and rtol even when explicitly specified in a test,https://github.com/pytorch/pytorch/issues/78053,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: primtorch, module: tests",scan when time allows,2022-05-22T00:39:47Z
issue,76783,`gradcheck` fails for `torch.trace`,https://github.com/pytorch/pytorch/issues/76783,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2022-05-21T13:58:01Z
issue,77614,PyTorch fails to build on gcc 12 due to gloo,https://github.com/pytorch/pytorch/issues/77614,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: third_party",scan when time allows,2022-05-20T15:50:31Z
issue,77967,Inference Tensors should not be allowed to hold `grad_fn`,https://github.com/pytorch/pytorch/issues/77967,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-05-20T15:16:13Z
issue,77869,Throw warning if python optimise flags are enabled,https://github.com/pytorch/pytorch/issues/77869,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python frontend,scan when time allows,2022-05-20T04:25:12Z
issue,47702,GPU acceleration for Apple's M1 chip?,https://github.com/pytorch/pytorch/issues/47702,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: performance,scan when time allows,2022-05-18T21:40:22Z
issue,77675,PrimTorch refs do not match argument naming with their PyTorch counterparts,https://github.com/pytorch/pytorch/issues/77675,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: primtorch,scan when time allows,2022-05-18T17:37:44Z
issue,77546,Build check for AVX512 fails with AMD CPU and march=native,https://github.com/pytorch/pytorch/issues/77546,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: vectorization",scan when time allows,2022-05-18T17:36:08Z
issue,77544,Modernize LoggingTensorMode,https://github.com/pytorch/pytorch/issues/77544,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: __torch_dispatch__,scan when time allows,2022-05-18T17:34:20Z
issue,77668,Extend BC test to test for __torch_function__ overridability,https://github.com/pytorch/pytorch/issues/77668,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: __torch_function__,scan when time allows,2022-05-18T17:30:54Z
issue,77731,complex abs strides are wrong on empty tensors and tensors with 1 dimension,https://github.com/pytorch/pytorch/issues/77731,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: complex, module: primtorch",scan when time allows,2022-05-18T16:42:41Z
issue,77738,net_observer_reporter_print.h missing,https://github.com/pytorch/pytorch/issues/77738,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2022-05-18T16:40:46Z
issue,28341,LinearOperator Abstraction / Structure-Exploiting LazyTensors for Linear Algebra,https://github.com/pytorch/pytorch/issues/28341,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: lazy, module: linear algebra",scan when time allows,2022-05-18T16:00:34Z
issue,77231,`torch.scatter_add` will succeed when the `index` is a complex tensor,https://github.com/pytorch/pytorch/issues/77231,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: scatter & gather ops,scan when time allows,2022-05-18T12:05:57Z
issue,62695,Use fast gpuAtomicAdd for FP16 data type in CUDA kernels,https://github.com/pytorch/pytorch/issues/62695,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: half",scan when time allows,2022-05-17T22:35:50Z
issue,63034,[bug] nn.functional.pad (circular) ubsan failure,https://github.com/pytorch/pytorch/issues/63034,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: padding",scan when time allows,2022-05-17T22:00:20Z
issue,74519,Lazy TS Test Failures,https://github.com/pytorch/pytorch/issues/74519,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-05-17T21:53:18Z
issue,24810,"Doesn't install the python module ""torch""",https://github.com/pytorch/pytorch/issues/24810,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2022-05-17T18:51:45Z
issue,77265,Subclasses with unwrapping `__torch_dispatch__` impls as parameters,https://github.com/pytorch/pytorch/issues/77265,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: __torch_dispatch__, module: nn",scan when time allows,2022-05-17T15:57:06Z
issue,77589,How to handle __module__ attribute for Public API bindings,https://github.com/pytorch/pytorch/issues/77589,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tests,scan when time allows,2022-05-17T14:37:45Z
issue,77548,lintrunner doesn't give good error message suggesting lintrunner init,https://github.com/pytorch/pytorch/issues/77548,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: lint,scan when time allows,2022-05-17T07:13:29Z
issue,77529,softmarginloss should use `log1p` and has an incorrect out= behaviour.,https://github.com/pytorch/pytorch/issues/77529,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: loss,scan when time allows,2022-05-17T07:05:33Z
issue,77454,More clarity in doc for `torch.cuda.Event.record`?,https://github.com/pytorch/pytorch/issues/77454,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2022-05-17T06:46:27Z
issue,58813,Single-Process Multi-GPU is not the recommended mode for DDP,https://github.com/pytorch/pytorch/issues/58813,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ddp, oncall: distributed",scan when time allows,2022-05-17T01:57:11Z
issue,76558,[jiterator] perf regression when jiterating few ops for complex dtype,https://github.com/pytorch/pytorch/issues/76558,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: complex, module: jiterator, module: performance",scan when time allows,2022-05-16T06:28:24Z
issue,52332,Reciprocals of complex tensors with infinities are different from NumPy.,https://github.com/pytorch/pytorch/issues/52332,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: complex, module: nans and infs, module: numpy",scan when time allows,2022-05-16T05:27:35Z
issue,57690,torch.cdist returns high diagonal values with CUDA,https://github.com/pytorch/pytorch/issues/57690,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: numerical-stability, module: tf32",scan when time allows,2022-05-15T15:27:09Z
issue,77415,Suboptimal error message - nn.Linear with double argument,https://github.com/pytorch/pytorch/issues/77415,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: error checking,scan when time allows,2022-05-15T14:35:03Z
issue,64145,F.conv2d: confusing error message when using uint8 input instead of float32,https://github.com/pytorch/pytorch/issues/64145,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: error checking",scan when time allows,2022-05-15T03:00:30Z
issue,31141,[JIT] tensor(device=...) and tensor.to(device = ...) does not work properly in traced functions and modules,https://github.com/pytorch/pytorch/issues/31141,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2022-05-15T02:04:53Z
issue,74748,[jiterator] Jiterate Complex Ops,https://github.com/pytorch/pytorch/issues/74748,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: complex, module: jiterator",scan when time allows,2022-05-14T17:00:59Z
issue,77397,Large numerical inconsistency for `torch.einsum` on RTX30 series GPU.,https://github.com/pytorch/pytorch/issues/77397,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: tf32",scan when time allows,2022-05-14T01:26:40Z
issue,77411,Allow force building with/without AVX,https://github.com/pytorch/pytorch/issues/77411,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2022-05-13T14:03:01Z
issue,77380,microbenchmark-style tests,https://github.com/pytorch/pytorch/issues/77380,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tests,scan when time allows,2022-05-13T13:55:45Z
issue,76354,NVFuser failing extremal opinfos,https://github.com/pytorch/pytorch/issues/76354,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nvfuser,scan when time allows,2022-05-12T22:40:53Z
issue,76780,`gradcheck` should support the comparison of NaN,https://github.com/pytorch/pytorch/issues/76780,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2022-05-12T20:36:16Z
issue,77340,Disable issue doesn't disable multiple dtypes correctly,https://github.com/pytorch/pytorch/issues/77340,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2022-05-12T14:36:03Z
issue,77223,torch.ops.aten.ceil(1.5) returns Tensor rather than scalar,https://github.com/pytorch/pytorch/issues/77223,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: type promotion,scan when time allows,2022-05-11T23:36:17Z
issue,62675,dynamic shapes,https://github.com/pytorch/pytorch/issues/62675,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: lazy,scan when time allows,2022-05-11T20:49:40Z
issue,7795,[feature request] Add cudaification API for distributions,https://github.com/pytorch/pytorch/issues/7795,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distributions,scan when time allows,2022-05-11T19:57:35Z
issue,77216,[primTorch] Reduction references don't return views consistent with their original operators,https://github.com/pytorch/pytorch/issues/77216,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: primtorch,scan when time allows,2022-05-11T19:35:36Z
issue,77241,libtorch1.8 torch::sigmoid is wrong,https://github.com/pytorch/pytorch/issues/77241,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: jetson",scan when time allows,2022-05-11T17:46:49Z
issue,77184,Undefined symbol error when compiling and loading C++ extension,https://github.com/pytorch/pytorch/issues/77184,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: cpp-extensions",scan when time allows,2022-05-11T15:46:41Z
issue,77251,The codegen unconditionaly generate code even when it is not going to be used,https://github.com/pytorch/pytorch/issues/77251,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: codegen,scan when time allows,2022-05-11T14:28:24Z
issue,77166,"torch.cholesky has been deprecated in favour of torch.linalg.cholesky. However, torch.cholesky_inverse remains as is. It should also be moved to torch.linalg",https://github.com/pytorch/pytorch/issues/77166,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: linear algebra,scan when time allows,2022-05-11T10:00:30Z
issue,75659,Misleading documentation for cholesky_inverse,https://github.com/pytorch/pytorch/issues/75659,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: linear algebra",scan when time allows,2022-05-10T23:07:03Z
issue,77140,"torch.randperm uses too much cpu, but not efficient.",https://github.com/pytorch/pytorch/issues/77140,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: multithreading, module: performance, module: random",scan when time allows,2022-05-10T18:37:05Z
issue,69080,Generalized matmul,https://github.com/pytorch/pytorch/issues/69080,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: linear algebra,scan when time allows,2022-05-10T16:44:48Z
issue,76709,`reshape` for distributions.,https://github.com/pytorch/pytorch/issues/76709,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distributions,scan when time allows,2022-05-10T16:40:33Z
issue,77161,SymInt shouldn't be in dynamic_type.h,https://github.com/pytorch/pytorch/issues/77161,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp,scan when time allows,2022-05-10T13:44:46Z
issue,77118,__name__ on OpOverload should not contain period,https://github.com/pytorch/pytorch/issues/77118,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: __torch_dispatch__,scan when time allows,2022-05-10T13:27:05Z
issue,77067,Unable to continue adding modules to `nn.Sequential` after using `del` method,https://github.com/pytorch/pytorch/issues/77067,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2022-05-09T21:34:40Z
issue,48845,memory_format kwarg doesn't work on most factory functions,https://github.com/pytorch/pytorch/issues/48845,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, module: memory format",scan when time allows,2022-05-09T20:45:54Z
issue,67565,Adding a new kwarg to a torch.nn.functional function breaks FC for JIT,https://github.com/pytorch/pytorch/issues/67565,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, oncall: jit",scan when time allows,2022-05-09T19:05:56Z
issue,76853,Inconsistent results between Pow and Float Pow with their numpy references for complex types,https://github.com/pytorch/pytorch/issues/76853,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: complex, module: nans and infs",scan when time allows,2022-05-09T18:29:22Z
issue,76779,"Strange warning from `matmul(..., out=...)`",https://github.com/pytorch/pytorch/issues/76779,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: linear algebra,scan when time allows,2022-05-09T17:09:46Z
issue,76887,Multiprocessing DataLoader hangs on exception inside iterator when using a simple queue and a producer thread,https://github.com/pytorch/pytorch/issues/76887,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dataloader, module: multiprocessing",scan when time allows,2022-05-09T13:57:04Z
issue,77047,Stop calling sizes/numel/dim/is_contiguous on undefined tensors,https://github.com/pytorch/pytorch/issues/77047,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: internals,scan when time allows,2022-05-09T04:33:35Z
issue,76972,Display EC2 information,https://github.com/pytorch/pytorch/issues/76972,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2022-05-09T04:29:49Z
issue,76966,[bug] `NATIVE` and `OMP` `parallel_for` implementations are inconsistent.,https://github.com/pytorch/pytorch/issues/76966,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: multithreading, module: openmp",scan when time allows,2022-05-09T04:29:21Z
issue,76936,"Private API for accessing all ""internal"" attributes on Tensors",https://github.com/pytorch/pytorch/issues/76936,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: internals,scan when time allows,2022-05-09T04:16:23Z
issue,76933,NVFuser opinfos - check for CudaFusionGroup in the graph,https://github.com/pytorch/pytorch/issues/76933,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nvfuser,scan when time allows,2022-05-09T04:16:03Z
issue,76906,torch.nn.functional.linear sometimes incorrectly accepts arguments of the different type,https://github.com/pytorch/pytorch/issues/76906,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: type promotion",scan when time allows,2022-05-09T04:14:46Z
issue,76891,Unwanted behavior with some in-place operations on CPU,https://github.com/pytorch/pytorch/issues/76891,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: viewing and reshaping,scan when time allows,2022-05-09T04:11:19Z
issue,76786,"`cumprod, prod` will backward fail if `dtype` argument is different than the dtype of input tensor",https://github.com/pytorch/pytorch/issues/76786,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: complex",scan when time allows,2022-05-09T04:07:42Z
issue,75599,kthvalue 20x slower than sort,https://github.com/pytorch/pytorch/issues/75599,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: performance, module: sorting and selection",scan when time allows,2022-05-08T20:39:01Z
issue,77046,torch.stack test_conj_view and test_neg_view are failing after 77043,https://github.com/pytorch/pytorch/issues/77046,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: complex, module: primtorch, module: viewing and reshaping",scan when time allows,2022-05-08T08:28:10Z
issue,42402,Reading data speed slower than tensorflow,https://github.com/pytorch/pytorch/issues/42402,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dataloader, module: performance",scan when time allows,2022-05-07T11:37:41Z
issue,70348,nan return by nn.CrossEntropyLoss when all the labels are ignore_index in torch 1.11,https://github.com/pytorch/pytorch/issues/70348,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2022-05-06T18:17:43Z
issue,76944,primTorch references don't handle scalar x scalar inputs correctly,https://github.com/pytorch/pytorch/issues/76944,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: primtorch, module: type promotion",scan when time allows,2022-05-06T06:22:48Z
issue,76441,[checkpoint] Switch away from pickle-base serialization,https://github.com/pytorch/pytorch/issues/76441,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-05-05T23:42:15Z
issue,76733,test_python_reference_meta_functions takes too long to run,https://github.com/pytorch/pytorch/issues/76733,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tests,scan when time allows,2022-05-05T21:10:51Z
issue,76871,PrimTorch binary refs do not handle CUDA + CPU scalar tensors correctly,https://github.com/pytorch/pytorch/issues/76871,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: primtorch,scan when time allows,2022-05-05T18:58:23Z
issue,76638,Design API for accessing sparse tensor indices,https://github.com/pytorch/pytorch/issues/76638,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2022-05-05T18:05:11Z
issue,15963,ProcessGroupGlooTest.test_scatter_stress_cuda is flaky,https://github.com/pytorch/pytorch/issues/15963,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: c10d, module: flaky-tests, oncall: distributed",scan when time allows,2022-05-05T15:55:25Z
issue,16295,ProcessGroupGlooTest.test_gather_stress is flaky,https://github.com/pytorch/pytorch/issues/16295,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: c10d, module: flaky-tests, oncall: distributed",scan when time allows,2022-05-05T15:52:10Z
issue,52230,Removing mutation on block input should print any meaning message.,https://github.com/pytorch/pytorch/issues/52230,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2022-05-04T23:23:54Z
issue,76483,Expand pow and float_pow sampling function for more coverage,https://github.com/pytorch/pytorch/issues/76483,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tests,scan when time allows,2022-05-04T22:25:15Z
issue,76844,Unify torch.ops argument parsing code with PythonArgParser,https://github.com/pytorch/pytorch/issues/76844,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: __torch_dispatch__, module: python frontend",scan when time allows,2022-05-04T21:41:55Z
issue,76827,TYPEIGNORE lint run locally disagrees with CI,https://github.com/pytorch/pytorch/issues/76827,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: lint, module: macos",scan when time allows,2022-05-04T21:32:37Z
issue,76659,Remove all docstrings when python is running in optimization mode,https://github.com/pytorch/pytorch/issues/76659,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: performance, module: python frontend",scan when time allows,2022-05-04T17:14:41Z
issue,76806,torch.Tensor.__rdiv__ long x scalar float type promotion is incorrect,https://github.com/pytorch/pytorch/issues/76806,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: primtorch, module: type promotion",scan when time allows,2022-05-04T10:03:33Z
issue,76804,"torch.add bool x bool allows integer alpha, inconsistent with other dtype type checking",https://github.com/pytorch/pytorch/issues/76804,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: primtorch,scan when time allows,2022-05-04T09:53:29Z
issue,75733,"`torch.sum, prod, cumsum, cumprod, sparse.sum` INTERNAL ASSERT FAIL",https://github.com/pytorch/pytorch/issues/75733,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: error checking, module: reductions",scan when time allows,2022-05-04T08:15:09Z
issue,76705,"Torch `x += y.bmm(z)` is faster than `x.baddbmm_(y, z)`",https://github.com/pytorch/pytorch/issues/76705,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: linear algebra, module: performance",scan when time allows,2022-05-03T21:24:19Z
issue,46164,Deprecate spmm and dsmm functions,https://github.com/pytorch/pytorch/issues/46164,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: deprecation, module: linear algebra, module: sparse",scan when time allows,2022-05-03T18:32:05Z
issue,76636,WeightNorm: Add reset_parameters Linear override,https://github.com/pytorch/pytorch/issues/76636,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: norms and normalization",scan when time allows,2022-05-03T18:27:31Z
issue,76747,[checkpoint] Stable file format for checkpoints,https://github.com/pytorch/pytorch/issues/76747,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2022-05-03T16:46:53Z
issue,76734,__torch_function__ callers should always pass kwargs,https://github.com/pytorch/pytorch/issues/76734,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: __torch_function__,scan when time allows,2022-05-03T16:45:42Z
issue,76745,[checkpoint] Handle overlapping storage during save and load,https://github.com/pytorch/pytorch/issues/76745,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2022-05-03T16:40:11Z
issue,76644,`torch.smm` backward fail with strange error message,https://github.com/pytorch/pytorch/issues/76644,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2022-05-03T16:18:22Z
issue,76732,Adding Polyloss to `torch`,https://github.com/pytorch/pytorch/issues/76732,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: loss, module: nn",scan when time allows,2022-05-03T16:06:03Z
issue,76338,torch.bucketize doc typo on the left boundary when 'right=True',https://github.com/pytorch/pytorch/issues/76338,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: sorting and selection",scan when time allows,2022-05-03T09:28:40Z
issue,76612,OpInfo incorrectly advertises lu_solve support on CUDA even when compiled without magma,https://github.com/pytorch/pytorch/issues/76612,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: linear algebra, module: tests",scan when time allows,2022-05-03T09:04:18Z
issue,76585,`torch.linalg.cond` has different results for tensor requiring autograd,https://github.com/pytorch/pytorch/issues/76585,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: linear algebra",scan when time allows,2022-05-03T09:02:18Z
issue,76618,pylint segfault,https://github.com/pytorch/pytorch/issues/76618,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: lint,scan when time allows,2022-05-03T08:57:35Z
issue,76611,OpInfo CUDA bfloat16 support detection is buggy,https://github.com/pytorch/pytorch/issues/76611,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tests,scan when time allows,2022-05-03T08:57:15Z
issue,76627,`Tensor.logit`'s signature in doc misses `eps` argument,https://github.com/pytorch/pytorch/issues/76627,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: python frontend",scan when time allows,2022-05-03T08:56:22Z
issue,76649,Add `ldl_unpack` functionality,https://github.com/pytorch/pytorch/issues/76649,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: linear algebra,scan when time allows,2022-05-03T08:50:59Z
issue,76655,[feature request] no-param sort to exploit parallelization,https://github.com/pytorch/pytorch/issues/76655,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: multithreading, module: performance, module: sorting and selection",scan when time allows,2022-05-03T08:50:18Z
issue,76654,`torch.sort` does not exploit parallelization when invoked without the `dim` parameter.,https://github.com/pytorch/pytorch/issues/76654,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: performance, module: sorting and selection",scan when time allows,2022-05-03T08:49:35Z
issue,76646,`torch.nn.HuberLoss` backwards unexpectedly fail,https://github.com/pytorch/pytorch/issues/76646,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: loss, module: nn, module: type promotion",scan when time allows,2022-05-03T00:19:42Z
issue,42407,torch.distributions.multinomial.Multinomial cannot be used in batch,https://github.com/pytorch/pytorch/issues/42407,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distributions,scan when time allows,2022-05-02T19:42:15Z
issue,76555,[numpy] Missing Tensor-Scalar support for multiple binary ops,https://github.com/pytorch/pytorch/issues/76555,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: numpy, module: python frontend",scan when time allows,2022-05-02T18:14:43Z
issue,71249,[docs] nn.Sequential docs should list member functions,https://github.com/pytorch/pytorch/issues/71249,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2022-05-02T14:51:12Z
issue,76519,Fix layout of masked output when all sparse dimensions are reduced,https://github.com/pytorch/pytorch/issues/76519,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2022-05-02T11:55:59Z
issue,17113,Distributed training jobs do not terminate properly if there is a crash,https://github.com/pytorch/pytorch/issues/17113,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2022-04-30T01:53:53Z
issue,76514,Clean up PyTorch's private operators,https://github.com/pytorch/pytorch/issues/76514,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: internals,scan when time allows,2022-04-30T00:47:48Z
issue,64041,torch.unique acting up for a binary tensor,https://github.com/pytorch/pytorch/issues/64041,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: boolean tensor, module: correctness (silent), module: numpy, module: sorting and selection",scan when time allows,2022-04-29T14:00:59Z
issue,76487,`F.interpolate` uses incorrect size when `align_corners=True`,https://github.com/pytorch/pytorch/issues/76487,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: interpolation,scan when time allows,2022-04-28T21:08:30Z
issue,76451,[checkpoint] Extension hooks to support logging and telemetry,https://github.com/pytorch/pytorch/issues/76451,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-04-27T22:51:24Z
issue,76491,[NVFuser] Automated generation of microbenchmarks,https://github.com/pytorch/pytorch/issues/76491,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nvfuser,scan when time allows,2022-04-27T22:49:42Z
issue,76465,Pyre type checking fails,https://github.com/pytorch/pytorch/issues/76465,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: typing,scan when time allows,2022-04-27T21:31:15Z
issue,76433,Remove _log_softmax/_softmax in favor of log_softmax and softmax respectively.,https://github.com/pytorch/pytorch/issues/76433,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: internals,scan when time allows,2022-04-27T21:25:38Z
issue,54135,torch.kron of a transposed input error,https://github.com/pytorch/pytorch/issues/54135,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: error checking,scan when time allows,2022-04-27T20:26:58Z
issue,76160,test_wishart_log_prob fails locally for me,https://github.com/pytorch/pytorch/issues/76160,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: distributions, module: tests",scan when time allows,2022-04-27T02:11:10Z
issue,76347,`index_select` allows negative `index` for sparse but not for strided `self`,https://github.com/pytorch/pytorch/issues/76347,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: advanced indexing, module: sparse",scan when time allows,2022-04-26T21:58:26Z
issue,76362,Whether 'targetSize' in inferExpandGeometryImpl needs to be checked when it is less than 0,https://github.com/pytorch/pytorch/issues/76362,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: error checking,scan when time allows,2022-04-26T21:58:20Z
issue,74554,Specifying left-right-padding as tuple for asymmetric padding,https://github.com/pytorch/pytorch/issues/74554,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: padding",scan when time allows,2022-04-26T21:26:43Z
issue,76225,"RuntimeError: bucket_count == per_bucket_sizes.size()INTERNAL ASSERT FAILED at ""/opt/conda/conda-bld/pytorch_1646755853042/work/torch/csrc/distributed/c10d/reducer.cpp"":980, please report a bug to PyTorch.",https://github.com/pytorch/pytorch/issues/76225,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2022-04-26T17:26:10Z
issue,76331,[checkpoint] Avoid loading whole tensor when resharding,https://github.com/pytorch/pytorch/issues/76331,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-04-26T17:02:18Z
issue,76327,[checkpoint] Add extension points to avoid the default serialization behavior,https://github.com/pytorch/pytorch/issues/76327,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-04-26T17:02:00Z
issue,76326,[checkpoint] Support models with different cross-rank metadata,https://github.com/pytorch/pytorch/issues/76326,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-04-26T17:01:55Z
issue,76323,[Checkpoint] Add module documentation,https://github.com/pytorch/pytorch/issues/76323,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-04-26T17:01:45Z
issue,76294,torch._remove_batch_dim is interceptable by __torch_function__ / batch tensors don't print correctly,https://github.com/pytorch/pytorch/issues/76294,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: __torch_function__, module: vmap",scan when time allows,2022-04-26T02:54:43Z
issue,76300,Unexpected _LinAlgError appeared only on my device,https://github.com/pytorch/pytorch/issues/76300,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: linear algebra,scan when time allows,2022-04-25T23:26:23Z
issue,76175,Many dispatch keys do not print to string correctly,https://github.com/pytorch/pytorch/issues/76175,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dispatch,scan when time allows,2022-04-25T17:19:01Z
issue,76166,at::real and at::imag as methods,https://github.com/pytorch/pytorch/issues/76166,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: complex,scan when time allows,2022-04-25T17:17:22Z
issue,76127,run_test.py option to write out failed tests,https://github.com/pytorch/pytorch/issues/76127,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tests,scan when time allows,2022-04-25T17:15:53Z
issue,76106,Allow any operation that takes a Storage to also take a contiguous Tensor instead,https://github.com/pytorch/pytorch/issues/76106,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python frontend,scan when time allows,2022-04-25T17:13:09Z
issue,75986,gql_mocks.json has really long lines,https://github.com/pytorch/pytorch/issues/75986,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-04-25T17:09:55Z
issue,75911,'python setup.py build' failed but succeed using 'pip install -v .' which calls 'python setup.py build'.,https://github.com/pytorch/pytorch/issues/75911,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2022-04-25T17:09:39Z
issue,70906,`full` should take an integer size,https://github.com/pytorch/pytorch/issues/70906,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: python array api, module: tensor creation",scan when time allows,2022-04-25T14:22:43Z
issue,42332,RuntimeError: ProcessGroupNCCL does not support recv (torch-nightly),https://github.com/pytorch/pytorch/issues/42332,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nccl, oncall: distributed",scan when time allows,2022-04-22T06:32:33Z
issue,75701,Einsum should have an `out=` parameter,https://github.com/pytorch/pytorch/issues/75701,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: linear algebra,scan when time allows,2022-04-22T00:39:29Z
issue,64864,Base parallelism on CPU count available to process rather than system total,https://github.com/pytorch/pytorch/issues/64864,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: multithreading,scan when time allows,2022-04-21T17:32:10Z
issue,72711,Feature request: Implement `gelsd` backend for `linalg.lstsq` via `linalg.svd`,https://github.com/pytorch/pytorch/issues/72711,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: linear algebra,scan when time allows,2022-04-21T07:12:08Z
issue,76012,torch.nn.LayerNorm is very slow on GPU (much slower than a custom LayerNorm version in the ConvNext model),https://github.com/pytorch/pytorch/issues/76012,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: nn",scan when time allows,2022-04-20T23:45:05Z
issue,75912,"interaction with psychopy during imports, script exits with: free(): invalid pointer. Aborted (core dumped)",https://github.com/pytorch/pytorch/issues/75912,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-04-20T23:12:06Z
issue,76138,Coverage test is only checking packages and not all submodules,https://github.com/pytorch/pytorch/issues/76138,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: python frontend",scan when time allows,2022-04-20T20:43:04Z
issue,76011,backcompat tests in test_nn.py are slow,https://github.com/pytorch/pytorch/issues/76011,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: tests",scan when time allows,2022-04-20T20:03:47Z
issue,75680,Addressing skips in OpInfo nn.functional.binary_cross_entropy_with_logits,https://github.com/pytorch/pytorch/issues/75680,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: tests",scan when time allows,2022-04-20T19:19:45Z
issue,70160,NewOperatorRegistrationTest.testImplNoDefGetsCaught failed.,https://github.com/pytorch/pytorch/issues/70160,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: dispatch, module: tests",scan when time allows,2022-04-20T13:02:46Z
issue,76070,NVFuser microbenchmark classifier - hash on memory formats,https://github.com/pytorch/pytorch/issues/76070,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nvfuser,scan when time allows,2022-04-20T12:24:33Z
issue,76030,Dirichlet with small concentration,https://github.com/pytorch/pytorch/issues/76030,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distributions,scan when time allows,2022-04-19T18:59:12Z
issue,75923,aten::_softmax.out doesn't work with non-contiguous Tensors,https://github.com/pytorch/pytorch/issues/75923,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2022-04-19T17:41:27Z
issue,76007,"Build a default NVFuser comparison callback, e.g. for use with torchbench",https://github.com/pytorch/pytorch/issues/76007,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nvfuser,scan when time allows,2022-04-19T17:40:22Z
issue,76025,Numerical instability: matrix multiplication got different results on cpu and gpu,https://github.com/pytorch/pytorch/issues/76025,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tf32,scan when time allows,2022-04-19T16:00:17Z
issue,76047,NNC failing opinfo accuracy tests,https://github.com/pytorch/pytorch/issues/76047,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-04-19T15:57:25Z
issue,76028,A bug in instructions for building PyTorch with ASAN,https://github.com/pytorch/pytorch/issues/76028,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ci, module: docs",scan when time allows,2022-04-19T15:57:19Z
issue,76037,__torch_function__ and generator input hazard,https://github.com/pytorch/pytorch/issues/76037,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: __torch_function__,scan when time allows,2022-04-19T15:56:41Z
issue,76043,RuntimeError: bucket_count == per_bucket_sizes.size() INTERNAL ASSERT FAILED,https://github.com/pytorch/pytorch/issues/76043,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: ddp",scan when time allows,2022-04-19T15:55:37Z
issue,76040,SSL certificate error: urlopen error [SSL: WRONG_VERSION_NUMBER] wrong version number (_ssl.c:1091)>,https://github.com/pytorch/pytorch/issues/76040,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-04-19T15:29:37Z
issue,75798,Fix workaround `__module__` used to appease public binding checks,https://github.com/pytorch/pytorch/issues/75798,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-04-18T18:29:44Z
issue,75652,Op segfaults with ForwardAD and Subclassed Tensor as Tangent,https://github.com/pytorch/pytorch/issues/75652,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: forward ad,scan when time allows,2022-04-18T15:34:04Z
issue,74616,"__rpow__(self, other) OpInfo should not test the case where `other` is a Tensor",https://github.com/pytorch/pytorch/issues/74616,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tests,scan when time allows,2022-04-18T02:34:55Z
issue,32881,CUDNN_STATUS_EXECUTION_FAILED,https://github.com/pytorch/pytorch/issues/32881,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: cudnn",scan when time allows,2022-04-17T05:16:42Z
issue,74788,[Structured] Make it possible to override create_out implementation on a per dispatch key basis (structured sparse),https://github.com/pytorch/pytorch/issues/74788,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: structured kernels,scan when time allows,2022-04-17T03:09:26Z
issue,47201,torch.save has a maximum size regardless of RAM,https://github.com/pytorch/pytorch/issues/47201,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: serialization",scan when time allows,2022-04-16T00:03:17Z
issue,75925,autogen-58 microbenchmark fails on NNC gpu fusion,https://github.com/pytorch/pytorch/issues/75925,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-04-15T22:51:08Z
issue,75904,Move _SKIP_PYTHON_BINDINGS to native_functions.yaml,https://github.com/pytorch/pytorch/issues/75904,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: codegen,scan when time allows,2022-04-15T21:55:10Z
issue,75909,Add batching rules for `{view}_copy` operators,https://github.com/pytorch/pytorch/issues/75909,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: batching,scan when time allows,2022-04-15T21:53:43Z
issue,75895,add -D_GLIBCXX_ASSERTIONS in debug mode,https://github.com/pytorch/pytorch/issues/75895,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2022-04-15T21:36:00Z
issue,75494,`torch.pinverse` on CUDA tensors produces non-optimal output for certain type of (invertible) matrix on torch > 1.7.1 but not on torch <= 1.7.1,https://github.com/pytorch/pytorch/issues/75494,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: linear algebra, module: numerical-stability",scan when time allows,2022-04-15T18:24:25Z
issue,66271,1.10.0-rc1 fails to compile with gcc 6.4.0,https://github.com/pytorch/pytorch/issues/66271,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2022-04-15T13:27:08Z
issue,74907,[Proposal] Use batched oprations to accelerate PowerSGD,https://github.com/pytorch/pytorch/issues/74907,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ddp, module: mta, module: optimizer, module: performance",scan when time allows,2022-04-13T22:31:22Z
issue,75582,Add `balance` flag to `random_split`,https://github.com/pytorch/pytorch/issues/75582,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: data,scan when time allows,2022-04-13T20:42:56Z
issue,75737,torch.device missing doctring,https://github.com/pytorch/pytorch/issues/75737,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2022-04-13T17:24:30Z
issue,72257,`torch.hasnan`,https://github.com/pytorch/pytorch/issues/72257,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, module: nans and infs",scan when time allows,2022-04-13T16:19:45Z
issue,75565,make lint should advertise make setup_lint,https://github.com/pytorch/pytorch/issues/75565,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: lint",scan when time allows,2022-04-12T16:28:26Z
issue,75562,"clang-tidy ""error: do not use const_cast"" cppcoreguidelines-pro-type-const-cast is counterproductive",https://github.com/pytorch/pytorch/issues/75562,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: lint",scan when time allows,2022-04-12T16:27:48Z
issue,75667,Implement histc for bfloat16 on CPU,https://github.com/pytorch/pytorch/issues/75667,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bfloat16, module: sorting and selection",scan when time allows,2022-04-12T15:43:51Z
issue,75474,[NVFuser] call kernels with informative names (e.g. pow_mul_add),https://github.com/pytorch/pytorch/issues/75474,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nvfuser,scan when time allows,2022-04-12T07:54:23Z
issue,75503,make_dual errors out when primal is a Tensor and tangent is a subclass Tensor,https://github.com/pytorch/pytorch/issues/75503,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: forward ad,scan when time allows,2022-04-12T07:08:23Z
issue,55277,Small model occupies too much GPU in CUDA11.1 + Torch1.8.1 but is normal in Torch 1.6 + CUDA10.1,https://github.com/pytorch/pytorch/issues/55277,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: memory usage, module: performance",scan when time allows,2022-04-12T04:54:00Z
issue,56396,Tests in CI are run from the test/ directory,https://github.com/pytorch/pytorch/issues/56396,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: tests",scan when time allows,2022-04-11T21:06:43Z
issue,75213,Improve wording in _store_based_barrier logging and identify ranks that have not joined,https://github.com/pytorch/pytorch/issues/75213,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2022-04-11T17:42:00Z
issue,65230,RuntimeError: Trying to create tensor with negative dimension -1741885395: [-1741885395],https://github.com/pytorch/pytorch/issues/65230,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2022-04-11T17:24:33Z
issue,74014,torch.utils.data.Dataset combined with pycuda issue,https://github.com/pytorch/pytorch/issues/74014,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: data,scan when time allows,2022-04-11T14:43:09Z
issue,75568,mypy typing strategy for Tensor-likes (`__torch_function__`),https://github.com/pytorch/pytorch/issues/75568,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: __torch_function__, module: typing",scan when time allows,2022-04-11T14:21:19Z
issue,75549,`torch.linalg.lstsq` raises `CUBLAS_STATUS_EXECUTION_FAILED` for large `B` in CUDA tensors,https://github.com/pytorch/pytorch/issues/75549,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cublas,scan when time allows,2022-04-11T13:41:00Z
issue,75554,torch.overrides testing is not catching people adding new kwargs and not passing on to handle_torch_function,https://github.com/pytorch/pytorch/issues/75554,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: __torch_function__, module: tests",scan when time allows,2022-04-11T13:40:16Z
issue,72345,Add ZeroTensor fastpath for torch.addmm,https://github.com/pytorch/pytorch/issues/72345,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-04-11T01:02:12Z
issue,72362,Add ZeroTensor fastpath for torch.bmm,https://github.com/pytorch/pytorch/issues/72362,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-04-11T00:59:00Z
issue,72359,Add ZeroTensor fastpath for torch.addmv,https://github.com/pytorch/pytorch/issues/72359,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-04-11T00:58:55Z
issue,73930,"User raised TypeError from __torch_dispatch__ gets turned into generic ""unsupported operand type"" message",https://github.com/pytorch/pytorch/issues/73930,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: __torch_dispatch__, module: bootcamp",scan when time allows,2022-04-10T23:13:48Z
issue,74876,Failure to set number of threads on AWS Lambda,https://github.com/pytorch/pytorch/issues/74876,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: multithreading",scan when time allows,2022-04-10T05:31:33Z
issue,75371,"NVFuser bad ""reshape"" performance",https://github.com/pytorch/pytorch/issues/75371,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nvfuser,scan when time allows,2022-04-08T21:47:20Z
issue,75025,torch.cuda.init() unstacks existing CUDA contexts,https://github.com/pytorch/pytorch/issues/75025,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2022-04-08T21:15:08Z
issue,75510,"[doc] view appears to mean different things, `view/reshape` vs `transpose/permute`.",https://github.com/pytorch/pytorch/issues/75510,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: numpy, module: partial aliasing, module: viewing and reshaping",scan when time allows,2022-04-08T19:24:20Z
issue,75419,Require PyTorch test suite to be warnings clean,https://github.com/pytorch/pytorch/issues/75419,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tests,scan when time allows,2022-04-08T17:47:51Z
issue,75427,`torch.cuda.is_bf16_supported()` seem to not work properly,https://github.com/pytorch/pytorch/issues/75427,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bfloat16, module: cuda",scan when time allows,2022-04-08T07:55:41Z
issue,75495,Forward AD convolution fails for the empty backend,https://github.com/pytorch/pytorch/issues/75495,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: forward ad",scan when time allows,2022-04-08T05:26:24Z
issue,75462,NotImplemented confusion between __torch_function__ and __rpow__ (and other dunder magic methods),https://github.com/pytorch/pytorch/issues/75462,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: __torch_function__,scan when time allows,2022-04-08T03:57:51Z
issue,75363,conv3d has numerical issue where same input produces output that are not bit-wise identical,https://github.com/pytorch/pytorch/issues/75363,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: determinism",scan when time allows,2022-04-07T18:46:29Z
issue,21567,torch.bernoulli's parameter generator not documented,https://github.com/pytorch/pytorch/issues/21567,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: random",scan when time allows,2022-04-07T11:06:14Z
issue,75251,Potential runtime optimization of Mish activation,https://github.com/pytorch/pytorch/issues/75251,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: python frontend",scan when time allows,2022-04-06T21:27:54Z
issue,75198,`torch.cuda.get_device_name` fails to identify RTX 3090 Ti,https://github.com/pytorch/pytorch/issues/75198,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2022-04-05T16:24:40Z
issue,10375,"Sending CUDA tensor to process, and then back, does not work",https://github.com/pytorch/pytorch/issues/10375,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, module: cuda, module: multiprocessing",scan when time allows,2022-04-05T13:33:11Z
issue,74185,[DDP] Parallelize initialization collectives,https://github.com/pytorch/pytorch/issues/74185,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ddp, oncall: distributed",scan when time allows,2022-04-04T19:26:48Z
issue,74485,dir(torch._VF) doesn't work,https://github.com/pytorch/pytorch/issues/74485,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-04-04T17:19:51Z
issue,75180,Add BUILD_LAZY_CUDA_SPARSE option,https://github.com/pytorch/pytorch/issues/75180,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cuda, module: sparse",scan when time allows,2022-04-04T15:56:17Z
issue,74052,assertEqual gives confusing error when comparing tuple with Tensor with Tensor,https://github.com/pytorch/pytorch/issues/74052,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tests,scan when time allows,2022-04-04T14:51:33Z
issue,74951,[feature request] `coalesce` to support `dim` argument.,https://github.com/pytorch/pytorch/issues/74951,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: performance, module: sparse",scan when time allows,2022-04-04T00:40:04Z
issue,71446,Discussion of TorchQuantum and QuantumNAS,https://github.com/pytorch/pytorch/issues/71446,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: complex,scan when time allows,2022-04-03T21:13:00Z
issue,71636,`torch.median` will return -2147483648 when input is an empty tensor,https://github.com/pytorch/pytorch/issues/71636,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: error checking,scan when time allows,2022-04-02T17:18:20Z
issue,74148,powersgd can not get linear growth due to extra 2 times allreduce,https://github.com/pytorch/pytorch/issues/74148,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ddp, oncall: distributed",scan when time allows,2022-04-02T17:14:56Z
issue,75129,[BE] add documentation for adjust learning rate when going to distributed training,https://github.com/pytorch/pytorch/issues/75129,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2022-04-01T19:35:53Z
issue,45125,Boolean indexing of an ndarray with a torch.tensor mask breaks for size=1,https://github.com/pytorch/pytorch/issues/45125,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: advanced indexing, module: numpy",scan when time allows,2022-04-01T14:10:44Z
issue,71629,"`torch.{max,min}` have strange error message when `input.numel()==0`",https://github.com/pytorch/pytorch/issues/71629,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: reductions,scan when time allows,2022-04-01T13:50:04Z
issue,74896,Install PyTorch from source on power machine,https://github.com/pytorch/pytorch/issues/74896,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: power",scan when time allows,2022-03-31T22:58:24Z
issue,65423,Enable is_contiguous slowpath for torch_dispatch,https://github.com/pytorch/pytorch/issues/65423,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: __torch_dispatch__,scan when time allows,2022-03-31T20:21:27Z
issue,32524,Pytorch 1.4 does not detect gpu,https://github.com/pytorch/pytorch/issues/32524,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2022-03-31T18:37:49Z
issue,73182,Segmentation fault in _sobol_engine_scramble_,https://github.com/pytorch/pytorch/issues/73182,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: edge cases, module: error checking, module: random",scan when time allows,2022-03-31T16:53:59Z
issue,74911,"Include finfo(dtype).[min, max, eps, tiny] in the extremal test case",https://github.com/pytorch/pytorch/issues/74911,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tests,scan when time allows,2022-03-30T04:58:30Z
issue,74942,Store SourceDataset in MapDataset using pointer,https://github.com/pytorch/pytorch/issues/74942,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: dataloader",scan when time allows,2022-03-30T03:46:43Z
issue,74937,[docs] RandomSampler has unrendered back-ticks,https://github.com/pytorch/pytorch/issues/74937,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2022-03-30T03:45:40Z
issue,64477,torch.onnx.export crash while export quantized model,https://github.com/pytorch/pytorch/issues/64477,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2022-03-30T01:08:14Z
issue,74222,Autograd API to get saved for backwards tensors for an autograd graph,https://github.com/pytorch/pytorch/issues/74222,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: double backwards",scan when time allows,2022-03-29T20:30:13Z
issue,74604,"Training, Forward / backward pass with _different_ batch-size, no speedup observed when backward pass has smaller batch-size",https://github.com/pytorch/pytorch/issues/74604,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2022-03-29T17:29:26Z
issue,74884,Problematic ASGD Optimizer,https://github.com/pytorch/pytorch/issues/74884,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: lrscheduler, module: optimizer",scan when time allows,2022-03-29T13:23:18Z
issue,74167,Slower performance of `torch.mm` method with sparse CSR tensor,https://github.com/pytorch/pytorch/issues/74167,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: mkl, module: sparse",scan when time allows,2022-03-29T09:51:27Z
issue,74491,__torch_function__ is hitting private functions in some cases,https://github.com/pytorch/pytorch/issues/74491,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: __torch_function__,scan when time allows,2022-03-28T23:10:02Z
issue,74500,One should be able to query if a DecorateInfo is a xfail or skip,https://github.com/pytorch/pytorch/issues/74500,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: testing,scan when time allows,2022-03-28T22:20:58Z
issue,74503,Compilation error on M1 Mac,https://github.com/pytorch/pytorch/issues/74503,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: arm, module: build, module: macos",scan when time allows,2022-03-28T22:19:47Z
issue,74514,Figure out a lint for directly returning a nullptr instead of raising python_error,https://github.com/pytorch/pytorch/issues/74514,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-03-28T22:19:19Z
issue,74855,Enable dtype keyword argument for to_dense method,https://github.com/pytorch/pytorch/issues/74855,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2022-03-28T22:16:45Z
issue,69033,Lazy Tensor Core Documentation Out-of-Date,https://github.com/pytorch/pytorch/issues/69033,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: lazy, module: xla",scan when time allows,2022-03-28T18:42:19Z
issue,73504,`ge` and `div` behaves differently when converting an overflow number,https://github.com/pytorch/pytorch/issues/73504,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: edge cases, module: type promotion",scan when time allows,2022-03-28T18:42:15Z
issue,74544,Many inplace operators are not being tested for variant consistency (test_variant_consistency_eager),https://github.com/pytorch/pytorch/issues/74544,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tests,scan when time allows,2022-03-28T13:10:17Z
issue,74679,Embedding isn't determinstic on linux-xenial-cuda11.3-py3.7-gcc7,https://github.com/pytorch/pytorch/issues/74679,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: embedding",scan when time allows,2022-03-27T02:53:07Z
issue,30896,"FATAL_ERROR ""Failed to determine the source files for the regular expression backend""",https://github.com/pytorch/pytorch/issues/30896,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: third_party",scan when time allows,2022-03-26T03:28:39Z
issue,74771,[bug] `torch.multinomial` should throw error as documented,https://github.com/pytorch/pytorch/issues/74771,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-03-25T22:33:19Z
issue,51062,test_profiler_with_remote_builtin (__main__.TensorPipeRpcTestWithSpawn) is flaky,https://github.com/pytorch/pytorch/issues/51062,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: flaky-tests, module: rpc, oncall: distributed",scan when time allows,2022-03-25T22:25:57Z
issue,64093,test_backward_accumulate_grads (__main__.TensorPipeDistAutogradTest) is flaky,https://github.com/pytorch/pytorch/issues/64093,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: flaky-tests, module: rpc, oncall: distributed",scan when time allows,2022-03-25T22:16:07Z
issue,74634,Sampling Issue With Distributions,https://github.com/pytorch/pytorch/issues/74634,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: half,scan when time allows,2022-03-25T21:22:49Z
issue,23547,model use dilated conv backward in v1.1.0 is ~3x slower than in v0.4.1 on 1080Ti,https://github.com/pytorch/pytorch/issues/23547,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: cuda, module: cudnn, module: performance",scan when time allows,2022-03-25T07:05:09Z
issue,71518,Compilation instructions are not exhaustive: <<parameter packs not expanded with ‘...’>> on Fedora 35/CUDA 11.6,https://github.com/pytorch/pytorch/issues/71518,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2022-03-25T02:06:20Z
issue,12435,`pstrf` on positive semi-definite matrices,https://github.com/pytorch/pytorch/issues/12435,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: linear algebra,scan when time allows,2022-03-25T00:21:47Z
issue,46157,Ability to disable cusolver,https://github.com/pytorch/pytorch/issues/46157,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cuda, module: linear algebra",scan when time allows,2022-03-25T00:00:09Z
issue,72448,Use Module `__getattr__` in `torch.ops` and friends,https://github.com/pytorch/pytorch/issues/72448,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-03-24T23:20:48Z
issue,74642,Some easy way to add xfails to OpInfos,https://github.com/pytorch/pytorch/issues/74642,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: testing,scan when time allows,2022-03-24T18:25:10Z
issue,69009,Mamba does not respect `cpuonly` when creating a conda environment from YAML file,https://github.com/pytorch/pytorch/issues/69009,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: releng,scan when time allows,2022-03-24T08:54:53Z
issue,74384,Package and Deploy: Generate External Python Registration files at runtime [WIP],https://github.com/pytorch/pytorch/issues/74384,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: deploy,scan when time allows,2022-03-23T20:45:11Z
issue,50345,torch.special tracking issue,https://github.com/pytorch/pytorch/issues/50345,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: numpy, module: special",scan when time allows,2022-03-23T20:40:32Z
issue,934,dataloader parallels over elements vs over batches,https://github.com/pytorch/pytorch/issues/934,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2022-03-23T20:05:35Z
issue,41292,"[RFC, Tracker] DataLoader improvements",https://github.com/pytorch/pytorch/issues/41292,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2022-03-23T18:25:11Z
issue,45111,(prototype) Graph Mode Dynamic Quantization on BERT failure on quantize_dynamic_jit(...) call,https://github.com/pytorch/pytorch/issues/45111,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: mobile,scan when time allows,2022-03-23T16:12:35Z
issue,36108,Add a flow_sample function to sample optical flows and displacement fields,https://github.com/pytorch/pytorch/issues/36108,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2022-03-23T12:58:02Z
issue,73180,Segmentation fault in _sobol_engine_ff_,https://github.com/pytorch/pytorch/issues/73180,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: edge cases, module: error checking, module: random",scan when time allows,2022-03-23T03:34:41Z
issue,73181,Segmentation fault in _sobol_engine_initialize_state_,https://github.com/pytorch/pytorch/issues/73181,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: edge cases, module: error checking, module: random",scan when time allows,2022-03-23T03:34:29Z
issue,73190,Segmentation fault in max_pool1d,https://github.com/pytorch/pytorch/issues/73190,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: edge cases, module: error checking, module: pooling",scan when time allows,2022-03-23T03:30:58Z
issue,74320,Make OpInfo repr more useful,https://github.com/pytorch/pytorch/issues/74320,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: testing,scan when time allows,2022-03-22T11:20:53Z
issue,74328,instantiate_device_type_tests is misnamed,https://github.com/pytorch/pytorch/issues/74328,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: testing,scan when time allows,2022-03-22T08:47:57Z
issue,74531,Feature Request: torch.testing.make_scalar (make_tensor for scalars),https://github.com/pytorch/pytorch/issues/74531,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: testing,scan when time allows,2022-03-22T08:41:46Z
issue,37332,Concerning default confiugration for distribution packages,https://github.com/pytorch/pytorch/issues/37332,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: build",scan when time allows,2022-03-21T20:53:31Z
issue,64025,RuntimeError: cuDNN error: CUDNN_STATUS_INTERNAL_ERROR,https://github.com/pytorch/pytorch/issues/64025,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cudnn,scan when time allows,2022-03-21T20:45:43Z
issue,74411,"UserWarning: size_average and reduce args will be deprecated, please use reduction='mean' instead.",https://github.com/pytorch/pytorch/issues/74411,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2022-03-21T16:35:50Z
issue,72960,Refactor/Cleanup lazy IR codegen,https://github.com/pytorch/pytorch/issues/72960,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: lazy,scan when time allows,2022-03-21T16:33:15Z
issue,74398,_run_ninja_build failing with clang,https://github.com/pytorch/pytorch/issues/74398,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2022-03-21T15:51:25Z
issue,74391,Can this cudaDeviceSynchronize call be removed?,https://github.com/pytorch/pytorch/issues/74391,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2022-03-21T15:50:11Z
issue,74421,"can not build pytorch, failing due to missing _ctypes module",https://github.com/pytorch/pytorch/issues/74421,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2022-03-21T15:06:21Z
issue,74337,None returned from data loader causes debugging difficulty from collate function,https://github.com/pytorch/pytorch/issues/74337,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2022-03-21T13:52:57Z
issue,74389,torch.package unpickling transforms: ModuleNotFoundError: No module named 'torch._C._linalg'; 'torch._C' is not a package,https://github.com/pytorch/pytorch/issues/74389,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: linear algebra, module: pickle, module: serialization",scan when time allows,2022-03-20T07:57:35Z
issue,61909,Support matrix operations between complex and real tensors,https://github.com/pytorch/pytorch/issues/61909,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: complex, module: linear algebra, module: type promotion",scan when time allows,2022-03-19T16:21:53Z
issue,41323,[discussion] Comparison operator chaining,https://github.com/pytorch/pytorch/issues/41323,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: boolean tensor,scan when time allows,2022-03-19T15:14:49Z
issue,74301,Error in lobpcg when using largest=False,https://github.com/pytorch/pytorch/issues/74301,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: masked operators, module: sparse",scan when time allows,2022-03-19T00:24:32Z
issue,74368,"Can't download torch vision models for older versions via torch.hub.load: ""multiple choices""",https://github.com/pytorch/pytorch/issues/74368,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: hub, module: models",scan when time allows,2022-03-18T21:52:07Z
issue,74329,CompositeImplicitAutograd addr implementation does not have correct behavior for bool (meta gives wrong dtype),https://github.com/pytorch/pytorch/issues/74329,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: meta tensors,scan when time allows,2022-03-17T16:22:10Z
issue,73710,Refactor/Cleanup Lazy Tensor Core,https://github.com/pytorch/pytorch/issues/73710,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: lazy,scan when time allows,2022-03-16T23:41:50Z
issue,62011,Enable clang-tidy on all of master,https://github.com/pytorch/pytorch/issues/62011,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-03-16T23:37:33Z
issue,74296,Time Gated Lstm,https://github.com/pytorch/pytorch/issues/74296,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: rnn",scan when time allows,2022-03-16T21:59:41Z
issue,74248,dir() on torch.ops.aten doesn't work,https://github.com/pytorch/pytorch/issues/74248,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-03-16T20:28:31Z
issue,74295,"TypeError: new_empty(): argument 'size' (position 1) must be tuple of ints, not list",https://github.com/pytorch/pytorch/issues/74295,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: error checking,scan when time allows,2022-03-16T20:23:36Z
issue,74300,repeat_interleave is not meta friendly; not transform friendly either,https://github.com/pytorch/pytorch/issues/74300,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-03-16T20:22:53Z
issue,74299,Gradients tests are very time consuming,https://github.com/pytorch/pytorch/issues/74299,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: ci",scan when time allows,2022-03-16T19:46:09Z
issue,74256,Create secure credential storage for metrics credentials and associated documentation on how to regenerate them if needed,https://github.com/pytorch/pytorch/issues/74256,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2022-03-16T17:30:02Z
issue,74281,Move torchbench workflow to `workflow_dispatch`?,https://github.com/pytorch/pytorch/issues/74281,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2022-03-16T05:03:41Z
issue,74187,[docker] test_corrcoef_cpu_complex64 fails on CPU build,https://github.com/pytorch/pytorch/issues/74187,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tests,scan when time allows,2022-03-15T07:49:03Z
issue,74138,`torch.set_printoptions` overwrites settings of its own previous calls,https://github.com/pytorch/pytorch/issues/74138,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: printing, module: ux",scan when time allows,2022-03-14T20:26:42Z
issue,73883,Possible issue with memory allocation.,https://github.com/pytorch/pytorch/issues/73883,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: memory usage, module: numpy",scan when time allows,2022-03-14T17:18:31Z
issue,74134,No module named 'pygame': soft_actor_critic,https://github.com/pytorch/pytorch/issues/74134,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: lazy,scan when time allows,2022-03-14T16:55:48Z
issue,74155,[Android Studio] DefaultCPUAllocator: not enough memory: you tried to allocate 280166432 bytes,https://github.com/pytorch/pytorch/issues/74155,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: android, module: memory usage",scan when time allows,2022-03-14T12:46:06Z
issue,74152,All-caps names for pages are confusing,https://github.com/pytorch/pytorch/issues/74152,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: doc infra,scan when time allows,2022-03-14T12:45:18Z
issue,74146,RuntimeError: Connection reset by peer when backened by NCCL,https://github.com/pytorch/pytorch/issues/74146,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: nccl",scan when time allows,2022-03-14T12:42:22Z
issue,73870,max_pool1d() returns when given invalid large `kernel_size` inputs,https://github.com/pytorch/pytorch/issues/73870,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: edge cases, module: nn",scan when time allows,2022-03-14T09:37:22Z
issue,73924,torch.cuda.amp: Remove SPMD DDP doc portion,https://github.com/pytorch/pytorch/issues/73924,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision), module: cuda, module: docs, oncall: distributed",scan when time allows,2022-03-14T09:28:34Z
issue,74101,C++ Context::setDeterministicAlgorithms default 2nd arg not defined in header,https://github.com/pytorch/pytorch/issues/74101,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: determinism",scan when time allows,2022-03-14T07:28:55Z
issue,72545,Split up torch.distributions docs into multiple pages,https://github.com/pytorch/pytorch/issues/72545,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: distributions, module: docs",scan when time allows,2022-03-13T23:38:30Z
issue,45943,Improve distributed documentation for NCCL_BLOCKING_WAIT,https://github.com/pytorch/pytorch/issues/45943,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ddp, oncall: distributed",scan when time allows,2022-03-12T03:50:09Z
issue,73916,Invalid code inrandom_ kernel,https://github.com/pytorch/pytorch/issues/73916,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distributions,scan when time allows,2022-03-11T15:28:32Z
issue,73989,Invalid example tensor: detectron2_maskrcnn,https://github.com/pytorch/pytorch/issues/73989,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: lazy,scan when time allows,2022-03-11T14:58:22Z
issue,73994,Lazy: add support for index ops,https://github.com/pytorch/pytorch/issues/73994,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: lazy,scan when time allows,2022-03-11T14:58:10Z
issue,73985,Release note bot is not sending messages on PRs anymore,https://github.com/pytorch/pytorch/issues/73985,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: releng,scan when time allows,2022-03-11T14:58:00Z
issue,6257,"[feature request] adding a nonzero element ""in-place"" in sparse tensor",https://github.com/pytorch/pytorch/issues/6257,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2022-03-11T09:10:47Z
issue,50097,Cannot print 32-bit complex tensors,https://github.com/pytorch/pytorch/issues/50097,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: complex, module: docs, module: half, module: printing",scan when time allows,2022-03-11T07:13:24Z
issue,66858,CUDAgraph error while capturing,https://github.com/pytorch/pytorch/issues/66858,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda graphs,scan when time allows,2022-03-09T20:03:29Z
issue,45760,Complex Number support for distributed,https://github.com/pytorch/pytorch/issues/45760,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, module: c10d, module: complex, oncall: distributed",scan when time allows,2022-03-09T18:59:59Z
issue,73860,tensor subclass and `__torch_function__` performance issues,https://github.com/pytorch/pytorch/issues/73860,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: __torch_function__,scan when time allows,2022-03-09T13:58:15Z
issue,73600,Add a section in DDP tutorial to explain why DDP sometimes is slower than local training and how to improve it,https://github.com/pytorch/pytorch/issues/73600,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ddp, oncall: distributed",scan when time allows,2022-03-08T22:03:17Z
issue,73683,test_broadcast_coalesced_nccl fails on A100 GPUs,https://github.com/pytorch/pytorch/issues/73683,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: c10d, oncall: distributed",scan when time allows,2022-03-08T22:03:14Z
issue,73721,Audit exception rewrapping to ensure stack traces are preserved,https://github.com/pytorch/pytorch/issues/73721,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: rpc, oncall: distributed",scan when time allows,2022-03-08T22:03:09Z
issue,73722,Why is distributed RPC using the default pickler?,https://github.com/pytorch/pytorch/issues/73722,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: rpc, oncall: distributed",scan when time allows,2022-03-08T22:03:04Z
issue,65022,"[AWS EC2 P3DN, EFA is enabled] Torch RPC tensorpipe/common/ibv.h:172 """": Operation not supported",https://github.com/pytorch/pytorch/issues/65022,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: rpc, module: tensorpipe, oncall: distributed",scan when time allows,2022-03-08T22:02:44Z
issue,73825,TypeError: cannot pickle 'torch._C._distributed_c10d._ProcessGroupGloo' object,https://github.com/pytorch/pytorch/issues/73825,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: c10d, module: ddp, oncall: distributed",scan when time allows,2022-03-08T21:59:42Z
issue,73070,`enumerate_support` for continuous distributions,https://github.com/pytorch/pytorch/issues/73070,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distributions,scan when time allows,2022-03-08T21:54:27Z
issue,67463,`test_forward_mode_AD` hangs for `nn.functional.cosine_embedding_loss`,https://github.com/pytorch/pytorch/issues/67463,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: deadlock, module: forward ad, module: nn",scan when time allows,2022-03-08T17:04:11Z
issue,73826,Better Error report in torch.distribution.*.sample (when passing a non-iterable),https://github.com/pytorch/pytorch/issues/73826,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: distributions, module: error checking",scan when time allows,2022-03-08T14:23:46Z
issue,73840,"max_pool1d(): argument 'dilation' must be tuple of ints, but found element of type int at pos 1",https://github.com/pytorch/pytorch/issues/73840,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: edge cases, module: pooling",scan when time allows,2022-03-07T20:34:48Z
issue,62973,Audit use of `C10_UNUSED`,https://github.com/pytorch/pytorch/issues/62973,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-03-07T18:19:44Z
issue,73841,max_pool1d(): `RuntimeError: [enforce fail at CPUAllocator.cpp:68] . DefaultCPUAllocator: can't allocate memory`,https://github.com/pytorch/pytorch/issues/73841,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: pooling,scan when time allows,2022-03-07T14:54:35Z
issue,73848,torch.distributions.multinomial.Multinomial (an example mistake of docs)?,https://github.com/pytorch/pytorch/issues/73848,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: distributions, module: docs",scan when time allows,2022-03-07T14:52:52Z
issue,73832,Legacy sparse tensor constructor (e.g. torch.cuda.sparse.FloatTensor) silently ignores device kwarg,https://github.com/pytorch/pytorch/issues/73832,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tensor creation,scan when time allows,2022-03-07T14:50:54Z
issue,73699,How to get tolerance override in OpInfo-based test?,https://github.com/pytorch/pytorch/issues/73699,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: testing",scan when time allows,2022-03-07T14:42:39Z
issue,46802,"DataLoader gives ""Broken pipe"" error on Linux platform",https://github.com/pytorch/pytorch/issues/46802,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dataloader, module: multiprocessing",scan when time allows,2022-03-07T02:54:02Z
issue,73176,"Cuda lacks checking of ""out of bound""",https://github.com/pytorch/pytorch/issues/73176,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: error checking, module: loss, module: nn",scan when time allows,2022-03-06T22:38:35Z
issue,73711,"Refactor/Cleanup LazyTensor, LTCTensorImpl, Data",https://github.com/pytorch/pytorch/issues/73711,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: lazy,scan when time allows,2022-03-05T00:12:08Z
issue,73755,CI: Bake as many dependencies as we can in the AMI (windows),https://github.com/pytorch/pytorch/issues/73755,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2022-03-04T19:23:55Z
issue,64048,"Be able to use pytest to run a ""core"" set of tests",https://github.com/pytorch/pytorch/issues/64048,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tests,scan when time allows,2022-03-04T19:09:52Z
issue,73083,Improve test_test_history.py,https://github.com/pytorch/pytorch/issues/73083,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2022-03-04T19:01:27Z
issue,72555,Enforce quotas on CI users,https://github.com/pytorch/pytorch/issues/72555,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2022-03-04T19:00:06Z
issue,45242,Functional interface for optimizers,https://github.com/pytorch/pytorch/issues/45242,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: optimizer",scan when time allows,2022-03-04T16:20:11Z
issue,73640,[Feature Request] Allow torch.Generator to be passed to torch.nn.init functions,https://github.com/pytorch/pytorch/issues/73640,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2022-03-03T21:45:06Z
issue,73709,DropIT: Dropping Intermediate Tensors for Memory-Efficient DNN Training,https://github.com/pytorch/pytorch/issues/73709,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: memory usage,scan when time allows,2022-03-03T18:36:41Z
issue,24205,deprecate cuda arch 3.5/3.7 in nightlies,https://github.com/pytorch/pytorch/issues/24205,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: binaries,scan when time allows,2022-03-03T17:37:36Z
issue,71951,Multivariate normal defined by eigendecomposition,https://github.com/pytorch/pytorch/issues/71951,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distributions,scan when time allows,2022-03-03T11:23:04Z
issue,60858,Sparse CSR layout CPU backend tracking issue,https://github.com/pytorch/pytorch/issues/60858,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: mkl, module: sparse",scan when time allows,2022-03-03T08:28:54Z
issue,37384,Error running trace on Pytorch Crowd Counting model,https://github.com/pytorch/pytorch/issues/37384,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2022-03-03T07:53:32Z
issue,21645,Batch Dataloader and Dataset,https://github.com/pytorch/pytorch/issues/21645,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2022-03-03T02:23:45Z
issue,73475,different results on each batch_size in torch==1.10.2+cu113 on RTX 3080,https://github.com/pytorch/pytorch/issues/73475,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: numerical-reproducibility,scan when time allows,2022-03-02T13:55:11Z
issue,69359,"[feature request] More masked reductions: amin/amax, argmin/argmax, quantile, mean/var/std/std_mean/var_mean",https://github.com/pytorch/pytorch/issues/69359,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: masked operators, module: sparse",scan when time allows,2022-03-02T11:46:37Z
issue,285,Keyword arguments passed to module's __call__ aren't forwarded to the hooks,https://github.com/pytorch/pytorch/issues/285,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2022-03-01T23:47:49Z
issue,1178,Batched sparse QR factorizations and solves with cusolver,https://github.com/pytorch/pytorch/issues/1178,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: linear algebra, module: sparse",scan when time allows,2022-03-01T21:02:45Z
issue,73585,Add optional log_scale argument for torch.distributions.Normal,https://github.com/pytorch/pytorch/issues/73585,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distributions,scan when time allows,2022-03-01T19:24:19Z
issue,73424,"""munmap_chunk(): invalid pointer"" interaction error with pytorch (< 1.10), pybind, and cv_bridge",https://github.com/pytorch/pytorch/issues/73424,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: lts,scan when time allows,2022-03-01T18:13:07Z
issue,842,BCELoss doesn't accept LongTensor targets,https://github.com/pytorch/pytorch/issues/842,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: loss,scan when time allows,2022-03-01T18:02:40Z
issue,711,Feature Request: Easier to extend base RNN implementation,https://github.com/pytorch/pytorch/issues/711,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-03-01T17:08:45Z
issue,630,Add Peephole connections for LSTMs?,https://github.com/pytorch/pytorch/issues/630,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-03-01T17:08:43Z
issue,88,expose backend selection and cudnn settings to the end user,https://github.com/pytorch/pytorch/issues/88,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cudnn,scan when time allows,2022-03-01T17:08:38Z
issue,73137,`create_graph=True` results in grad_fn error for differentiable functions,https://github.com/pytorch/pytorch/issues/73137,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2022-03-01T16:56:33Z
issue,65253,TestCppExtensionJit's test_crash_handler fails for force_on_cpu job on GHA,https://github.com/pytorch/pytorch/issues/65253,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2022-03-01T15:24:20Z
issue,73196,`torch.pow` errors out on specific input,https://github.com/pytorch/pytorch/issues/73196,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2022-03-01T14:33:59Z
issue,73150,How do we handle metadata-modifying in-place operators (like `squeeze_`) with `__torch_dispatch__`?,https://github.com/pytorch/pytorch/issues/73150,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: __torch_dispatch__,scan when time allows,2022-03-01T14:31:19Z
issue,73268,Add detection of interned submodule of externed module in PackageExporter,https://github.com/pytorch/pytorch/issues/73268,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: deploy,scan when time allows,2022-03-01T14:30:55Z
issue,73218,"This is not completely a bug, but something that will be amazing if can be taken care of regarding nn.DataParallel.",https://github.com/pytorch/pytorch/issues/73218,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: data parallel,scan when time allows,2022-03-01T14:27:29Z
issue,73425,ufunc codegen support for dtypes that are supported on CUDA but not CPU,https://github.com/pytorch/pytorch/issues/73425,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: codegen,scan when time allows,2022-03-01T14:22:07Z
issue,73505,Figure out what to do with functions that take both Tensor and TensorOptions,https://github.com/pytorch/pytorch/issues/73505,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: codegen,scan when time allows,2022-03-01T14:07:08Z
issue,73517,Teach tools.codegen.api.translate about IValues,https://github.com/pytorch/pytorch/issues/73517,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, module: codegen",scan when time allows,2022-03-01T14:05:51Z
issue,73506,Idiom for extensible string printing for TensorImpl subclasses,https://github.com/pytorch/pytorch/issues/73506,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: internals, module: printing",scan when time allows,2022-03-01T14:05:12Z
issue,44726,Pytorch Installation from source fails (GCC-8.4/CUDA-10.2/RHEL7) and (GCC-7.5/CUDA-10.2/RHEL7),https://github.com/pytorch/pytorch/issues/44726,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2022-03-01T09:05:41Z
issue,29042,Support clang+cuda builds,https://github.com/pytorch/pytorch/issues/29042,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2022-03-01T06:18:00Z
issue,64711,Improve Docs builds to not have to push to a repository to update,https://github.com/pytorch/pytorch/issues/64711,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: doc infra, module: docs",scan when time allows,2022-02-28T20:30:14Z
issue,67712,"Decorate tests with a ""deadline""",https://github.com/pytorch/pytorch/issues/67712,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2022-02-28T20:20:30Z
issue,68017,MacOS CI result is not in /README.md,https://github.com/pytorch/pytorch/issues/68017,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: macos",scan when time allows,2022-02-28T19:47:25Z
issue,70455,Make it possible to remove all hooks on a specified module without needing the hook handles,https://github.com/pytorch/pytorch/issues/70455,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2022-02-28T18:07:42Z
issue,73413,Conv3D consumes lots of memory on Mac with Apple Silicon,https://github.com/pytorch/pytorch/issues/73413,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: macos",scan when time allows,2022-02-28T16:27:58Z
issue,73484,Rename `keep_vars` in `nn.Module.state_dict`,https://github.com/pytorch/pytorch/issues/73484,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2022-02-28T15:48:14Z
issue,70419,LazyModules `cls_to_become` field exposes implementation detail,https://github.com/pytorch/pytorch/issues/70419,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: nn",scan when time allows,2022-02-28T09:03:00Z
issue,70500,Adding the new Phish activation function,https://github.com/pytorch/pytorch/issues/70500,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2022-02-28T08:03:06Z
issue,70447,The same code can not be reproduced on multiple GPUs with dataparallel,https://github.com/pytorch/pytorch/issues/70447,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: data parallel,scan when time allows,2022-02-27T22:08:16Z
issue,73161,native_batch_norm and native_layer_norm have strange epsilon behaviors,https://github.com/pytorch/pytorch/issues/73161,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: norms and normalization,scan when time allows,2022-02-27T21:29:17Z
issue,70490,Cuda sync mode input checking is wrong for non-string/int inputs.,https://github.com/pytorch/pytorch/issues/70490,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2022-02-27T14:03:19Z
issue,70413,"PyTorch crashes without an error message, when running this code snippet with torch.tensor subclassing & forward hooks (Not sure what the exact cause is, but the code snippet reliably causes it)",https://github.com/pytorch/pytorch/issues/70413,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-02-26T21:02:46Z
issue,70416,libtorch cuda use too much system memory,https://github.com/pytorch/pytorch/issues/70416,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp,scan when time allows,2022-02-26T21:02:45Z
issue,73461,`TestCommonCUDA.test_noncontiguous_samples_pca_lowrank_cuda_float32` fails on A100 due to TF32 operation in `svd_lowrank`,https://github.com/pytorch/pytorch/issues/73461,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: linear algebra, module: tests, module: tf32",scan when time allows,2022-02-26T19:07:30Z
issue,70386,AT_ASSERT fail with DataLoaderOptions().drop_last(),https://github.com/pytorch/pytorch/issues/70386,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2022-02-26T19:02:51Z
issue,70388,RuntimeError: tensor has too many (>25) dims when permuting tensor with GPU backend,https://github.com/pytorch/pytorch/issues/70388,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: assert failure,scan when time allows,2022-02-26T17:02:56Z
issue,70391,linalg.lstsq INTERNAL ASSERT FAILED,https://github.com/pytorch/pytorch/issues/70391,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: linear algebra, module: mkl",scan when time allows,2022-02-26T17:02:55Z
issue,65774,Non-persistent modules,https://github.com/pytorch/pytorch/issues/65774,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: serialization",scan when time allows,2022-02-26T15:59:50Z
issue,73417,MaybeEncodingError: Error sending result,https://github.com/pytorch/pytorch/issues/73417,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dataloader, module: multiprocessing",scan when time allows,2022-02-26T00:41:44Z
issue,72525,KL divergence between two Continuous Bernoulli is negative,https://github.com/pytorch/pytorch/issues/72525,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distributions,scan when time allows,2022-02-25T15:15:06Z
issue,73412,Build from source failed,https://github.com/pytorch/pytorch/issues/73412,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2022-02-25T11:10:08Z
issue,73355,Review and refactor the way libcublas static libraries are linked,https://github.com/pytorch/pytorch/issues/73355,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cublas, oncall: releng",scan when time allows,2022-02-25T04:38:15Z
issue,73380,elastic/rendezvous: _matches_machine_hostname doesn't resolve hostnames fully,https://github.com/pytorch/pytorch/issues/73380,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: elastic, oncall: r2p",scan when time allows,2022-02-25T04:36:10Z
issue,73394,Make it easier to figure out if packages need to be interned/mocked/externed,https://github.com/pytorch/pytorch/issues/73394,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2022-02-25T04:34:53Z
issue,72788,Docs bug: type annotations for linspace (and logspace) start and end arguments is wrong,https://github.com/pytorch/pytorch/issues/72788,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: tensor creation",scan when time allows,2022-02-24T10:14:09Z
issue,61758,`make_tensor` tracking issue,https://github.com/pytorch/pytorch/issues/61758,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: testing,scan when time allows,2022-02-24T09:58:38Z
issue,65640,[LTC] Memory growing up during training until OOM,https://github.com/pytorch/pytorch/issues/65640,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: lazy,scan when time allows,2022-02-23T18:57:23Z
issue,45321,Memory leak when using share_memory_ on cuda device,https://github.com/pytorch/pytorch/issues/45321,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: memory usage,scan when time allows,2022-02-23T17:18:11Z
issue,72194,`TestCase.assertEqual` has `equal_nan` default to `True`,https://github.com/pytorch/pytorch/issues/72194,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: testing, module: tests",scan when time allows,2022-02-23T17:16:39Z
issue,73051,Cannot install latest pytorch into Docker on Apple M1,https://github.com/pytorch/pytorch/issues/73051,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: arm, module: binaries, module: macos",scan when time allows,2022-02-22T20:20:33Z
issue,71698,[libtorch]can not save a vector<int> to AutogradContex->saved_data.,https://github.com/pytorch/pytorch/issues/71698,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: cpp",scan when time allows,2022-02-22T20:06:01Z
issue,73222,Having rrelu functional + module take a generator object to match native functions entry,https://github.com/pytorch/pytorch/issues/73222,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: random",scan when time allows,2022-02-22T18:58:08Z
issue,73134,Test functionalization doesn't run,https://github.com/pytorch/pytorch/issues/73134,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: internals,scan when time allows,2022-02-22T14:48:39Z
issue,72240,recurrent neural network module,https://github.com/pytorch/pytorch/issues/72240,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: nn, module: rnn",scan when time allows,2022-02-22T12:07:57Z
issue,72650,Some inplace ops don't raise on incompatible shapes and meta device,https://github.com/pytorch/pytorch/issues/72650,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: meta tensors, module: tests",scan when time allows,2022-02-22T11:53:32Z
issue,72537,Feature: Add tril_embed and triu_embed,https://github.com/pytorch/pytorch/issues/72537,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: linear algebra, module: tensor creation",scan when time allows,2022-02-22T11:49:56Z
issue,50098,Errors when coercing complex numbers of various sizes,https://github.com/pytorch/pytorch/issues/50098,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: complex, module: half",scan when time allows,2022-02-22T11:23:41Z
issue,70346,Add hints for gradient long time overflow when using torch.cuda.amp,https://github.com/pytorch/pytorch/issues/70346,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision), module: cuda",scan when time allows,2022-02-21T18:02:53Z
issue,56891,Batched SVD_LOWRANK being much slower than loop implementation (both CPU and GPU),https://github.com/pytorch/pytorch/issues/56891,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: linear algebra, module: performance",scan when time allows,2022-02-21T06:49:19Z
issue,43754,"Using Dataparallel with multi input error. Input and parameter tensors are not at the same device, found input tensor at cpu and parameter tensor at cuda:0",https://github.com/pytorch/pytorch/issues/43754,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: data parallel,scan when time allows,2022-02-20T23:52:44Z
issue,70171,Channels last performance problem,https://github.com/pytorch/pytorch/issues/70171,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: memory format, module: performance",scan when time allows,2022-02-20T07:02:51Z
issue,67457,"Reductions on zero-size dims: 1) by accepting a custom default value, 2) if tensor has another non-reduced zero-size dim",https://github.com/pytorch/pytorch/issues/67457,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: reductions,scan when time allows,2022-02-19T14:26:33Z
issue,68553,[distributed] elastic/agent/server/api.py could react faster on exit / Ctrl+D / Ctrl+C,https://github.com/pytorch/pytorch/issues/68553,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: r2p,scan when time allows,2022-02-19T11:26:54Z
issue,73034,Is it intentional that PyTorch linux binaries aren't manylinux1 compliant?,https://github.com/pytorch/pytorch/issues/73034,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: binaries,scan when time allows,2022-02-18T23:39:18Z
issue,69665,No dtype check for zero sparse tensor!,https://github.com/pytorch/pytorch/issues/69665,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2022-02-18T23:14:54Z
issue,73121,make c++ logger preamble meaningful,https://github.com/pytorch/pytorch/issues/73121,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: c10d,scan when time allows,2022-02-18T23:06:23Z
issue,70199,Missing Docker image for 1.10.1,https://github.com/pytorch/pytorch/issues/70199,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: docker",scan when time allows,2022-02-18T23:02:50Z
issue,70158,pin_memory *still* destroys custom containers,https://github.com/pytorch/pytorch/issues/70158,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2022-02-18T20:03:04Z
issue,73062,"Clang Compilation Error: more than one constructor applies to convert from ""ptrdiff_t"" to ""c10::Scalar""",https://github.com/pytorch/pytorch/issues/73062,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cuda, module: macos",scan when time allows,2022-02-18T19:48:38Z
issue,73065,`index_fill_` accepts wrong dtype for meta tensors,https://github.com/pytorch/pytorch/issues/73065,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: meta tensors, module: tests",scan when time allows,2022-02-18T19:22:37Z
issue,71843,Migrate master to main: https://github.com/pytorch/tutorials,https://github.com/pytorch/pytorch/issues/71843,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: infra,scan when time allows,2022-02-18T15:47:01Z
issue,17614,Document whether it is possible to train TorchScript modules,https://github.com/pytorch/pytorch/issues/17614,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: docs, oncall: jit",scan when time allows,2022-02-18T08:47:18Z
issue,72580,"Setting a list of Modules as an attribute does not work like setting a Module as an attribute, and throws no warning",https://github.com/pytorch/pytorch/issues/72580,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2022-02-17T19:44:44Z
issue,73007,Feature: support better rendering for ..deprecated Sphinx directive,https://github.com/pytorch/pytorch/issues/73007,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2022-02-17T19:26:10Z
issue,72933,Why there are 8 flavors of iOS build jobs for every commit,https://github.com/pytorch/pytorch/issues/72933,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ci, module: ios",scan when time allows,2022-02-17T03:19:34Z
issue,72909,"`torch.svd_lowrank` should set the default value of `q` as `min(6, m, n)`",https://github.com/pytorch/pytorch/issues/72909,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: linear algebra,scan when time allows,2022-02-17T03:18:05Z
issue,72911,Vectorized Jacobian and Hessian errors with ffts,https://github.com/pytorch/pytorch/issues/72911,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: vmap",scan when time allows,2022-02-16T22:27:47Z
issue,72915,Is the current behavior with addcmul and integer dtypes intended?,https://github.com/pytorch/pytorch/issues/72915,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tests,scan when time allows,2022-02-16T22:22:04Z
issue,72824,Some loss functions support `dtype` broadcast but some do not,https://github.com/pytorch/pytorch/issues/72824,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: loss, module: nn, module: type promotion",scan when time allows,2022-02-16T14:45:38Z
issue,71473,pytorch installation error with cuda!!,https://github.com/pytorch/pytorch/issues/71473,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: binaries,scan when time allows,2022-02-16T08:39:27Z
issue,70138,build_ios.sh prevents iOS.cmake from configuring ios deployment target correctly,https://github.com/pytorch/pytorch/issues/70138,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: ios",scan when time allows,2022-02-16T02:16:28Z
issue,66565,Bazel target all_tests improperly reports failures on CPU-only (non-CUDA) build,https://github.com/pytorch/pytorch/issues/66565,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bazel, module: bootcamp, module: tests",scan when time allows,2022-02-15T20:33:26Z
issue,63972,bazel build warning: Artifact 'torch/csrc/api/include/torch/version.h' is duplicated,https://github.com/pytorch/pytorch/issues/63972,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bazel, module: build, module: build warnings",scan when time allows,2022-02-15T20:33:26Z
issue,46631,error in bazel build //...,https://github.com/pytorch/pytorch/issues/46631,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bazel, module: build",scan when time allows,2022-02-15T20:33:25Z
issue,70100,Feature Request: dim parameter of torch.nn.functional.normalize should accept tuples,https://github.com/pytorch/pytorch/issues/70100,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: norms and normalization",scan when time allows,2022-02-15T20:03:04Z
issue,57121,"einsum ""jk,ijkl->il"" is ~16x slower than numpy",https://github.com/pytorch/pytorch/issues/57121,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: linear algebra, module: numpy, module: performance",scan when time allows,2022-02-15T18:43:27Z
issue,72850,N-dimensional Convolutions,https://github.com/pytorch/pytorch/issues/72850,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: convolution,scan when time allows,2022-02-15T16:56:18Z
issue,70095,Add support for a `default` arg in `ModuleDict.pop`,https://github.com/pytorch/pytorch/issues/70095,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2022-02-15T16:02:59Z
issue,71078,"`torch.nn.{Constant,Zero}Pad` unexpectedly fail",https://github.com/pytorch/pytorch/issues/71078,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: error checking, module: padding",scan when time allows,2022-02-15T02:17:40Z
issue,71059,`torch.scatter` will return random value when `input` is empty tensor,https://github.com/pytorch/pytorch/issues/71059,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: error checking, module: scatter & gather ops",scan when time allows,2022-02-15T02:17:15Z
issue,69348,`torch.sspaddmm` should broadcast the input tensor,https://github.com/pytorch/pytorch/issues/69348,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2022-02-14T23:11:20Z
issue,60234,Numerical-reproducibility issue in torch.matmul,https://github.com/pytorch/pytorch/issues/60234,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: numerical-reproducibility",scan when time allows,2022-02-14T20:39:46Z
issue,72714,Add `pct_end` parameter to `OneCycleLR`,https://github.com/pytorch/pytorch/issues/72714,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: lrscheduler, module: optimizer",scan when time allows,2022-02-14T16:49:37Z
issue,69984,Docs for torch.nn.MSELoss are confusing,https://github.com/pytorch/pytorch/issues/69984,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: loss, module: nn",scan when time allows,2022-02-14T15:02:55Z
issue,41162,Advanced indexing gradient is extremely slow when there are many duplicate indices,https://github.com/pytorch/pytorch/issues/41162,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: performance,scan when time allows,2022-02-14T14:46:20Z
issue,69408,`torch.hstack` should raise an error when tensor is 0 dimensional,https://github.com/pytorch/pytorch/issues/69408,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2022-02-14T13:43:11Z
issue,72784,Torch version in docker container does not match tag,https://github.com/pytorch/pytorch/issues/72784,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docker,scan when time allows,2022-02-14T05:19:14Z
issue,72768,Add NCCL and MPI version printing to torch.utils.collect_env,https://github.com/pytorch/pytorch/issues/72768,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: collect_env.py, module: mpi, module: nccl",scan when time allows,2022-02-14T04:44:34Z
issue,72452,Re-raise the exception when the `forward` of a parametrization raises,https://github.com/pytorch/pytorch/issues/72452,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: nn.utils.parametrize",scan when time allows,2022-02-14T01:03:12Z
issue,69960,"FeatureAlphaDropout doesn't drop channels for (C, D, H, W)",https://github.com/pytorch/pytorch/issues/69960,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: batching, module: nn",scan when time allows,2022-02-13T23:02:56Z
issue,69912,"torch.nn.functional.ctc_loss with invalid input produce NaN or infinity gradient, while the batch entries are fine",https://github.com/pytorch/pytorch/issues/69912,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: error checking, module: loss, module: nn",scan when time allows,2022-02-13T22:02:41Z
issue,72672,nn.functional No-batch-dim support should have OpInfo examples,https://github.com/pytorch/pytorch/issues/72672,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: tests",scan when time allows,2022-02-13T21:15:11Z
issue,69786,Implement aten::equal for sparse tensors,https://github.com/pytorch/pytorch/issues/69786,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2022-02-13T20:02:50Z
issue,41499,Helping test example code blocks in the docs,https://github.com/pytorch/pytorch/issues/41499,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: tests",scan when time allows,2022-02-13T19:30:49Z
issue,25661,finfo operator not bound into JIT,https://github.com/pytorch/pytorch/issues/25661,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2022-02-13T18:49:13Z
issue,71211,Lack of type check in `nn.functional` APIs,https://github.com/pytorch/pytorch/issues/71211,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2022-02-13T12:13:52Z
issue,69800,Incomplete error message at tensor indexing (when indexing with set),https://github.com/pytorch/pytorch/issues/69800,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: error checking,scan when time allows,2022-02-12T00:06:01Z
issue,69753,last_epoch parameter of CyclicLR and OneCycleLR is not the number of epochs,https://github.com/pytorch/pytorch/issues/69753,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: lrscheduler, module: optimizer",scan when time allows,2022-02-11T16:03:08Z
issue,72179,"Mechanism for Tensor subclasses to ""disable autograd""",https://github.com/pytorch/pytorch/issues/72179,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: __torch_dispatch__, module: autograd",scan when time allows,2022-02-10T21:29:17Z
issue,70678,Decouple `TensorIteratorBase` output from structured kernel outputs.,https://github.com/pytorch/pytorch/issues/70678,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: structured kernels,scan when time allows,2022-02-10T21:15:09Z
issue,6107,"Well documented, safe method to deserialize model parameters from untrusted sources",https://github.com/pytorch/pytorch/issues/6107,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: pickle, module: serialization",scan when time allows,2022-02-10T15:52:49Z
issue,43120,"When modified the model python file, the pytorch will raise the KeyError of this file",https://github.com/pytorch/pytorch/issues/43120,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: data parallel,scan when time allows,2022-02-10T04:37:16Z
issue,71364,make pytorch support different hardware acceleratioin configuration,https://github.com/pytorch/pytorch/issues/71364,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: cuda",scan when time allows,2022-02-09T21:32:53Z
issue,72258,NotImplementedError in torch.distributions,https://github.com/pytorch/pytorch/issues/72258,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distributions,scan when time allows,2022-02-09T14:14:22Z
issue,72558,Pin dependencies + expand the current linter,https://github.com/pytorch/pytorch/issues/72558,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, module: ci",scan when time allows,2022-02-08T23:14:30Z
issue,69714,Possibly out of date error in autograd codegen,https://github.com/pytorch/pytorch/issues/69714,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2022-02-08T22:03:16Z
issue,43567,`torch.svd()` CUDA gives incorrect results when input contains `nan`,https://github.com/pytorch/pytorch/issues/43567,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: linear algebra, module: nans and infs",scan when time allows,2022-02-08T18:17:01Z
issue,22848,[feature request] Log-determinant for symmetric positive definite matrices,https://github.com/pytorch/pytorch/issues/22848,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: linear algebra,scan when time allows,2022-02-08T15:31:55Z
issue,4358,Invoking MKL in multiprocessing with importing torch causes blocking,https://github.com/pytorch/pytorch/issues/4358,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: mkl, module: mkldnn, module: multiprocessing",scan when time allows,2022-02-08T15:11:28Z
issue,69744,Add ability to ignore arguments/outputs in `torch.autograd.functional.jacobian`,https://github.com/pytorch/pytorch/issues/69744,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2022-02-08T15:03:02Z
issue,69756,Introduction of nn.LazyRNNs,https://github.com/pytorch/pytorch/issues/69756,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: rnn",scan when time allows,2022-02-08T15:03:01Z
issue,72041,RuntimeError: Trying to backward through the graph a second time (or directly access saved tensors after they have already been freed). Saved intermediate values of the graph are freed when you call .backward() or autograd.grad(). Specify retain_graph=True if you need to backward through the graph a second time or if you need to access saved tensors after calling backward,https://github.com/pytorch/pytorch/issues/72041,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: nn.utils.parametrize",scan when time allows,2022-02-08T14:36:07Z
issue,68331,Casting real parameter to complex during forward produces warning on backward,https://github.com/pytorch/pytorch/issues/68331,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: complex",scan when time allows,2022-02-08T12:54:24Z
issue,58181,torch.split_with_sizes is not documented,https://github.com/pytorch/pytorch/issues/58181,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2022-02-08T10:38:46Z
issue,71575,Better Engineering: test_..._mem_overlap in test_torch.py should be ported to ErrorInputs,https://github.com/pytorch/pytorch/issues/71575,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tests,scan when time allows,2022-02-08T04:43:25Z
issue,72421,Feature Request: A simpler decorator for disabling mixed precision,https://github.com/pytorch/pytorch/issues/72421,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision),scan when time allows,2022-02-08T03:57:21Z
issue,71472,Feature request: Add complex support to `torch.nanmean`,https://github.com/pytorch/pytorch/issues/71472,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: complex, module: nans and infs, module: reductions",scan when time allows,2022-02-08T03:34:43Z
issue,66924,[BUG] Inconsistent initialization on different machines (tensor.uniform_()),https://github.com/pytorch/pytorch/issues/66924,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: random,scan when time allows,2022-02-08T03:01:07Z
issue,29844,Fold DispatchStub into c10 dispatcher,https://github.com/pytorch/pytorch/issues/29844,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dispatch,scan when time allows,2022-02-08T02:53:19Z
issue,69676,"[LazyTensor] model after to(device), traced IR changed and lose type info",https://github.com/pytorch/pytorch/issues/69676,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-02-08T00:06:58Z
issue,16737,torch.save overwrite,https://github.com/pytorch/pytorch/issues/16737,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2022-02-07T19:38:24Z
issue,16717,Performance regression on CPU from 0.4.1 to 1.0.0 on ResNet inference,https://github.com/pytorch/pytorch/issues/16717,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: performance, module: single threaded",scan when time allows,2022-02-07T19:36:36Z
issue,4930,[feature request] Type-1 Multi-layer bidirectional RNN,https://github.com/pytorch/pytorch/issues/4930,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cudnn, module: rnn",scan when time allows,2022-02-07T19:34:23Z
issue,10298,[feature request] Runtime warning for inappropriate labels (among others),https://github.com/pytorch/pytorch/issues/10298,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: error checking, module: loss",scan when time allows,2022-02-07T18:35:11Z
issue,3667,Exposing CuDNN benchmark strategy selection,https://github.com/pytorch/pytorch/issues/3667,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, module: cudnn",scan when time allows,2022-02-07T18:29:56Z
issue,72341,InstanceNorm doesn't preserve memory format,https://github.com/pytorch/pytorch/issues/72341,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: memory format, module: norms and normalization",scan when time allows,2022-02-07T18:29:23Z
issue,69530,Extend composite compliant testing to backward formulas,https://github.com/pytorch/pytorch/issues/69530,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: __torch_dispatch__,scan when time allows,2022-02-06T22:02:51Z
issue,29372,torch.std() returns nan for single item tensors.,https://github.com/pytorch/pytorch/issues/29372,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: numpy,scan when time allows,2022-02-06T21:40:12Z
issue,69360,1.10.0 failed to build due to missing TensorBody.h,https://github.com/pytorch/pytorch/issues/69360,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2022-02-06T14:45:03Z
issue,69531,"`(svd|pca)_lowrank`: backward is unstable when for a matrix `A`, the parameter `q` is set to a value `q > rank(A)`.",https://github.com/pytorch/pytorch/issues/69531,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: linear algebra, module: numerical-stability",scan when time allows,2022-02-05T22:03:30Z
issue,69516,`torch.sparse.softmax` and `torch.sparse.log_softmax` do not support negative dim.,https://github.com/pytorch/pytorch/issues/69516,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2022-02-05T20:02:43Z
issue,69442,The formula for KL-divloss is wrong in the document,https://github.com/pytorch/pytorch/issues/69442,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: nn",scan when time allows,2022-02-04T22:02:54Z
issue,72361,Add ZeroTensor fastpath for torch.baddbmm,https://github.com/pytorch/pytorch/issues/72361,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-02-04T20:31:23Z
issue,72358,Add ZeroTensor fastpath for torch.addbmm,https://github.com/pytorch/pytorch/issues/72358,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-02-04T20:28:14Z
issue,72347,Add ZeroTensor fastpath for torch.addcmul,https://github.com/pytorch/pytorch/issues/72347,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-02-04T18:37:39Z
issue,72332,Ability to assign to `tensor.require_grad` might lead to bugs,https://github.com/pytorch/pytorch/issues/72332,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: error checking, module: ux",scan when time allows,2022-02-04T16:47:29Z
issue,69921,Context manager to enable/disabled TensorFloat32 on demand,https://github.com/pytorch/pytorch/issues/69921,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tf32,scan when time allows,2022-02-03T19:08:27Z
issue,62812,Support `torch.linalg.einsum`,https://github.com/pytorch/pytorch/issues/62812,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: linear algebra,scan when time allows,2022-02-02T19:00:26Z
issue,71359,"Torchvision Installation Logic Python vs C++ ""Mismatch""",https://github.com/pytorch/pytorch/issues/71359,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2022-02-02T17:15:13Z
issue,71697,Libtorch dlls delayed loading,https://github.com/pytorch/pytorch/issues/71697,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: static linking",scan when time allows,2022-02-02T16:12:24Z
issue,32030,Dockerfile for people to quick-start contributing,https://github.com/pytorch/pytorch/issues/32030,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docker,scan when time allows,2022-02-02T14:07:53Z
issue,71918,Support mixed python scalar/tensor types for torch.clamp's min/max args,https://github.com/pytorch/pytorch/issues/71918,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: type promotion,scan when time allows,2022-02-02T14:06:34Z
issue,64645,Compilation from source fails,https://github.com/pytorch/pytorch/issues/64645,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: abi, module: build, module: protobuf",scan when time allows,2022-02-01T22:25:32Z
issue,71495,Memory Leak in PyTorch 1.10.1,https://github.com/pytorch/pytorch/issues/71495,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: memory usage",scan when time allows,2022-02-01T19:27:20Z
issue,71340,locally installed PyTorch upgraded or superseded on windows/macos conda tests,https://github.com/pytorch/pytorch/issues/71340,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: lts,scan when time allows,2022-02-01T19:03:32Z
issue,71877,Add `output_size` argument to `Upsample` forward method (just like for `ConvTranspose` Modules),https://github.com/pytorch/pytorch/issues/71877,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2022-02-01T17:59:49Z
issue,69325,`nn.functional.fractional_max_pool2d` and `nn.functional.fractional_max_pool3d` produce incorrect output on non-contiguous inputs,https://github.com/pytorch/pytorch/issues/69325,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2022-02-01T16:03:13Z
issue,14907,Missing dilation from several pooling modules (AvgPool),https://github.com/pytorch/pytorch/issues/14907,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: pooling,scan when time allows,2022-02-01T15:03:11Z
issue,69286,"subprocess.CalledProcessError: Command '['cmake', '--build', '.', '--target', 'install', '--config', 'Release', '--', '-j', '32']' returned non-zero exit status 1.",https://github.com/pytorch/pytorch/issues/69286,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2022-02-01T02:14:34Z
issue,72046,Building PyTorch with Vulkan backend don't work,https://github.com/pytorch/pytorch/issues/72046,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: vulkan,scan when time allows,2022-02-01T01:24:27Z
issue,72055,index_add : Inconsistent between CPU and CUDA,https://github.com/pytorch/pytorch/issues/72055,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: cuda",scan when time allows,2022-02-01T01:20:50Z
issue,71671,Many APIs `INTERNAL ASSERT FAILED` when promoting `complex32` dtype,https://github.com/pytorch/pytorch/issues/71671,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: complex,scan when time allows,2022-02-01T01:16:51Z
issue,71635,"`torch.nn.functional.{instance, batch}_norm` trigger INTERNAL ASSERT FAILED when input is empty tensor with `complex32`",https://github.com/pytorch/pytorch/issues/71635,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: complex, module: half",scan when time allows,2022-01-31T20:43:29Z
issue,69287,`out=` variant for `conj` unary operator,https://github.com/pytorch/pytorch/issues/69287,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: complex,scan when time allows,2022-01-31T20:02:57Z
issue,70654,[RFC] UCC integration in ProcessGroupNCCL,https://github.com/pytorch/pytorch/issues/70654,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2022-01-31T19:29:40Z
issue,69281,Implement reparameterized sampling for LKJCholesky distribution w.r.t. concentration parameter,https://github.com/pytorch/pytorch/issues/69281,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distributions,scan when time allows,2022-01-31T18:02:48Z
issue,69290,Deprecation warnings generated when including header files in C++17 code,https://github.com/pytorch/pytorch/issues/69290,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build warnings,scan when time allows,2022-01-31T18:02:46Z
issue,69288,Inplace and `out` variants for `positive` operator in PyTorch,https://github.com/pytorch/pytorch/issues/69288,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ux,scan when time allows,2022-01-31T16:03:54Z
issue,72061,"`svd_backward`: does not handle inputs of rank `r < min(m, n)`.",https://github.com/pytorch/pytorch/issues/72061,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: linear algebra",scan when time allows,2022-01-31T15:49:56Z
issue,54047,RuntimeError: cuDNN error: CUDNN_STATUS_NOT_INITIALIZED in pytorch lightning,https://github.com/pytorch/pytorch/issues/54047,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: cudnn",scan when time allows,2022-01-31T05:11:18Z
issue,71911,Functions depending on SVD are broken for inputs with non-finite values with MKL 2022+ and OpenBLAS 0.3.15+,https://github.com/pytorch/pytorch/issues/71911,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: linear algebra, module: mkl, module: openblas",scan when time allows,2022-01-31T03:43:21Z
issue,72022,[PT-D] To make the ShardedTensor Reshard more generic,https://github.com/pytorch/pytorch/issues/72022,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2022-01-31T03:34:18Z
issue,71842,Migrate master to main: https://github.com/pytorch/ignite,https://github.com/pytorch/pytorch/issues/71842,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: infra,scan when time allows,2022-01-31T03:32:33Z
issue,71841,Migrate master to main: https://github.com/pytorch/ELF,https://github.com/pytorch/pytorch/issues/71841,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: infra,scan when time allows,2022-01-31T03:32:33Z
issue,71815,Migrate master to main: https://github.com/pytorch/add-annotations-github-action,https://github.com/pytorch/pytorch/issues/71815,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: infra,scan when time allows,2022-01-31T03:32:03Z
issue,71814,Migrate master to main: https://github.com/pytorch/ossci-job-dsl,https://github.com/pytorch/pytorch/issues/71814,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: infra,scan when time allows,2022-01-31T03:32:02Z
issue,71813,Migrate master to main: https://github.com/pytorch/pytorch-integration-testing,https://github.com/pytorch/pytorch/issues/71813,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: infra,scan when time allows,2022-01-31T03:32:02Z
issue,71812,Migrate master to main: https://github.com/pytorch/pytorch-ci-dockerfiles,https://github.com/pytorch/pytorch/issues/71812,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: infra,scan when time allows,2022-01-31T03:32:01Z
issue,71810,Migrate master to main: https://github.com/pytorch/labeler-github-action,https://github.com/pytorch/pytorch/issues/71810,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: infra,scan when time allows,2022-01-31T03:32:00Z
issue,71840,Migrate master to main: https://github.com/pytorch/captum,https://github.com/pytorch/pytorch/issues/71840,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: infra,scan when time allows,2022-01-31T03:31:35Z
issue,71839,Migrate master to main: https://github.com/pytorch/glow,https://github.com/pytorch/pytorch/issues/71839,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: infra,scan when time allows,2022-01-31T03:31:35Z
issue,71838,Migrate master to main: https://github.com/pytorch/serve,https://github.com/pytorch/pytorch/issues/71838,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: infra,scan when time allows,2022-01-31T03:31:34Z
issue,71837,Migrate master to main: https://github.com/pytorch/xla,https://github.com/pytorch/pytorch/issues/71837,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: infra,scan when time allows,2022-01-31T03:31:34Z
issue,71836,Migrate master to main: https://github.com/pytorch/QNNPACK,https://github.com/pytorch/pytorch/issues/71836,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: infra,scan when time allows,2022-01-31T03:31:34Z
issue,71835,Migrate master to main: https://github.com/pytorch/tnt,https://github.com/pytorch/pytorch/issues/71835,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: infra,scan when time allows,2022-01-31T03:31:33Z
issue,71834,Migrate master to main: https://github.com/pytorch/hub,https://github.com/pytorch/pytorch/issues/71834,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: infra,scan when time allows,2022-01-31T03:31:33Z
issue,71833,Migrate master to main: https://github.com/pytorch/extension-cpp,https://github.com/pytorch/pytorch/issues/71833,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: infra,scan when time allows,2022-01-31T03:31:32Z
issue,71832,Migrate master to main: https://github.com/pytorch/android-demo-app,https://github.com/pytorch/pytorch/issues/71832,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: infra,scan when time allows,2022-01-31T03:31:32Z
issue,71831,Migrate master to main: https://github.com/pytorch/translate,https://github.com/pytorch/pytorch/issues/71831,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: infra,scan when time allows,2022-01-31T03:31:32Z
issue,71830,Migrate master to main: https://github.com/pytorch/elastic,https://github.com/pytorch/pytorch/issues/71830,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: infra,scan when time allows,2022-01-31T03:31:31Z
issue,71828,Migrate master to main: https://github.com/pytorch/tvm,https://github.com/pytorch/pytorch/issues/71828,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: infra,scan when time allows,2022-01-31T03:31:31Z
issue,71826,Migrate master to main: https://github.com/pytorch/ios-demo-app,https://github.com/pytorch/pytorch/issues/71826,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: infra,scan when time allows,2022-01-31T03:31:30Z
issue,71825,Migrate master to main: https://github.com/pytorch/accimage,https://github.com/pytorch/pytorch/issues/71825,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: infra,scan when time allows,2022-01-31T03:31:29Z
issue,71824,Migrate master to main: https://github.com/pytorch/extension-ffi,https://github.com/pytorch/pytorch/issues/71824,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: infra,scan when time allows,2022-01-31T03:31:29Z
issue,71823,Migrate master to main: https://github.com/pytorch/nestedtensor,https://github.com/pytorch/pytorch/issues/71823,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: infra,scan when time allows,2022-01-31T03:31:29Z
issue,71822,Migrate master to main: https://github.com/pytorch/cppdocs,https://github.com/pytorch/pytorch/issues/71822,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: infra,scan when time allows,2022-01-31T03:31:28Z
issue,71821,Migrate master to main: https://github.com/pytorch/workshops,https://github.com/pytorch/pytorch/issues/71821,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: infra,scan when time allows,2022-01-31T03:31:28Z
issue,71820,Migrate master to main: https://github.com/pytorch/extension-script,https://github.com/pytorch/pytorch/issues/71820,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: infra,scan when time allows,2022-01-31T03:31:27Z
issue,71819,Migrate master to main: https://github.com/pytorch/java-demo,https://github.com/pytorch/pytorch/issues/71819,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: infra,scan when time allows,2022-01-31T03:31:27Z
issue,71818,Migrate master to main: https://github.com/pytorch/csprng,https://github.com/pytorch/pytorch/issues/71818,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: infra,scan when time allows,2022-01-31T03:31:27Z
issue,71817,Migrate master to main: https://github.com/pytorch/pytorch_sphinx_theme,https://github.com/pytorch/pytorch/issues/71817,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: infra,scan when time allows,2022-01-31T03:31:26Z
issue,71816,Migrate master to main: https://github.com/pytorch/rfcs,https://github.com/pytorch/pytorch/issues/71816,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: infra,scan when time allows,2022-01-31T03:31:26Z
issue,69207,Function(s) expecting a tuple argument don't accept generators,https://github.com/pytorch/pytorch/issues/69207,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2022-01-30T21:02:51Z
issue,69233,[ROCM] elementwise kernel launch,https://github.com/pytorch/pytorch/issues/69233,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rocm,scan when time allows,2022-01-30T21:02:48Z
issue,69072,Libtorch dll versioning,https://github.com/pytorch/pytorch/issues/69072,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: binaries,scan when time allows,2022-01-30T11:02:49Z
issue,69086,Confusing value of `TORCH_CUDA_ARCH_LIST`,https://github.com/pytorch/pytorch/issues/69086,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2022-01-29T22:02:56Z
issue,54309,[discussion] Support torch.matmul: strided x sparse in addition to sparse x strided,https://github.com/pytorch/pytorch/issues/54309,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2022-01-29T21:54:27Z
issue,56126,AttributeError: module 'torch._C' has no attribute 'ComplexDoubleStorageBase',https://github.com/pytorch/pytorch/issues/56126,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2022-01-29T10:13:07Z
issue,69013,libtorch_cuda links against wrong libnccl.so,https://github.com/pytorch/pytorch/issues/69013,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: abi, module: build, module: third_party",scan when time allows,2022-01-29T00:05:27Z
issue,71633,Negative Exponents of Int tensors result in output of zero,https://github.com/pytorch/pytorch/issues/71633,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2022-01-28T01:41:58Z
issue,63992,Direct inversion and linear systems solutions for small matrices,https://github.com/pytorch/pytorch/issues/63992,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: linear algebra, module: performance",scan when time allows,2022-01-27T13:07:39Z
issue,71377,is_alias_of support for storageless tensors,https://github.com/pytorch/pytorch/issues/71377,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ddp, module: lazy, module: partial aliasing",scan when time allows,2022-01-26T22:39:55Z
issue,65216,Add function that retrieves a batch from a DataLoader,https://github.com/pytorch/pytorch/issues/65216,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2022-01-26T18:56:33Z
issue,70559,Behavior of torch.nn.functional.interpolate with unchanged output size and recompute_scale_factor=False,https://github.com/pytorch/pytorch/issues/70559,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: interpolation, module: nn",scan when time allows,2022-01-26T17:14:42Z
issue,66066,Finishing OpInfos: test_autograd.py,https://github.com/pytorch/pytorch/issues/66066,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: tests",scan when time allows,2022-01-26T15:07:45Z
issue,64502,[numpy] torch.nonzero is similar to np.argwhere not np.nonzero,https://github.com/pytorch/pytorch/issues/64502,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: numpy,scan when time allows,2022-01-26T06:23:10Z
issue,70954,Training grouped Conv2D is slow,https://github.com/pytorch/pytorch/issues/70954,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: cudnn, module: performance",scan when time allows,2022-01-26T02:11:38Z
issue,71760,Adapting the citation style according to GitHub's CFF,https://github.com/pytorch/pytorch/issues/71760,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: doc infra,scan when time allows,2022-01-25T20:00:18Z
issue,42861,Can not get pytorch working with tensorboard,https://github.com/pytorch/pytorch/issues/42861,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tensorboard,scan when time allows,2022-01-25T18:00:42Z
issue,64635,'LSTM' object has no attribute '_flat_weights_names',https://github.com/pytorch/pytorch/issues/64635,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rnn,scan when time allows,2022-01-25T15:16:28Z
issue,71204,`torch.diag` unexpectedly fails,https://github.com/pytorch/pytorch/issues/71204,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: numpy, module: viewing and reshaping",scan when time allows,2022-01-25T05:56:51Z
issue,71660,Build release binaries with USE_GLOG=ON by default,https://github.com/pytorch/pytorch/issues/71660,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: infra",scan when time allows,2022-01-25T03:33:54Z
issue,32021,[discussion] Relax optimizer constructor constraints for simplicity,https://github.com/pytorch/pytorch/issues/32021,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2022-01-25T00:34:16Z
issue,71682,_GLIBCXX_USE_CXX11_ABI=0 does not work when building from source code,https://github.com/pytorch/pytorch/issues/71682,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2022-01-24T23:57:31Z
issue,71689,torch.distributions.categorical.Categorical does not work with 0 batch size,https://github.com/pytorch/pytorch/issues/71689,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distributions,scan when time allows,2022-01-24T23:47:31Z
issue,71396,a lot nightly builds are canceled due to VM errors since Jan14,https://github.com/pytorch/pytorch/issues/71396,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: ci",scan when time allows,2022-01-24T23:42:34Z
issue,64124,ciflow tracking meta issues,https://github.com/pytorch/pytorch/issues/64124,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-01-24T20:44:33Z
issue,52851,"`embedding_bag(..., include_last_offset=True)` should always error if `offsets[-1] != input.size()`",https://github.com/pytorch/pytorch/issues/52851,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: embedding, module: nn",scan when time allows,2022-01-24T20:13:03Z
issue,55549,`embedding_bag` has unexpected behavior when given `offsets` that are not monotonically increasing,https://github.com/pytorch/pytorch/issues/55549,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: embedding, module: nn",scan when time allows,2022-01-24T20:12:29Z
issue,68803,[DDP] Verify ignored parameter names in debug mode during init,https://github.com/pytorch/pytorch/issues/68803,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: c10d, oncall: distributed",scan when time allows,2022-01-22T10:02:39Z
issue,68701,Tensor precision manifests differently between CPU and GPU.,https://github.com/pytorch/pytorch/issues/68701,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: arm, module: cuda, module: jetson, module: numerical-stability",scan when time allows,2022-01-22T09:02:47Z
issue,68696,Is it possible to use torch.linalg.cholesky for cholesky decomposition of a huge matrix that doesn't fit in memory?,https://github.com/pytorch/pytorch/issues/68696,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: linear algebra,scan when time allows,2022-01-22T08:02:57Z
issue,71084,"`torch.{inverse,cholesky}` have wrong shape check of square matrices",https://github.com/pytorch/pytorch/issues/71084,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: error checking, module: linear algebra",scan when time allows,2022-01-22T06:17:36Z
issue,70916,uint8 scalar tensors cannot be used for integer indexing,https://github.com/pytorch/pytorch/issues/70916,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python array api,scan when time allows,2022-01-22T01:55:37Z
issue,68781,[docs] Clarify DDP activation checkpointing support,https://github.com/pytorch/pytorch/issues/68781,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ddp, module: docs, oncall: distributed",scan when time allows,2022-01-22T00:05:44Z
issue,68655,[ux] F.binary_cross_entropy (and maybe other losses) to auto-cast bool category mask to float,https://github.com/pytorch/pytorch/issues/68655,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: ux",scan when time allows,2022-01-21T22:03:06Z
issue,68746,Add kl_divergence between Normal and Laplace distribution.,https://github.com/pytorch/pytorch/issues/68746,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distributions,scan when time allows,2022-01-21T22:03:02Z
issue,71630,More than 4 Dimensions for Channel Last Memory Format,https://github.com/pytorch/pytorch/issues/71630,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: memory format,scan when time allows,2022-01-21T17:06:38Z
issue,69491,[RFC] Support MemoryView for Tensors,https://github.com/pytorch/pytorch/issues/69491,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: internals, module: numpy, module: viewing and reshaping",scan when time allows,2022-01-21T15:15:43Z
issue,71541,"`torch.sub` behaves differently with `add`, `mul`, `div`",https://github.com/pytorch/pytorch/issues/71541,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: error checking,scan when time allows,2022-01-21T14:16:57Z
issue,71471,"[pytorch1.5.0] subprocess.CalledProcessError: Command '['cmake', '--build', '.', '--target', 'install', '--config', 'Release', '--', '-j', '6']' returned non-zero exit status 2.",https://github.com/pytorch/pytorch/issues/71471,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2022-01-21T11:52:43Z
issue,71595,Optimizer Overlap: Follow up features,https://github.com/pytorch/pytorch/issues/71595,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ddp, oncall: distributed",scan when time allows,2022-01-20T22:39:51Z
issue,71209,support setting `keepdim` without setting `dim`,https://github.com/pytorch/pytorch/issues/71209,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python array api,scan when time allows,2022-01-20T22:06:39Z
issue,70454,Building docs locally fails,https://github.com/pytorch/pytorch/issues/70454,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2022-01-20T21:57:11Z
issue,71543,Initialize DataLoader workers in parallel,https://github.com/pytorch/pytorch/issues/71543,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: data,scan when time allows,2022-01-20T17:51:24Z
issue,69524,[feature request] Accepting python scalar inputs for torch.minimum/torch.maximum and friends,https://github.com/pytorch/pytorch/issues/69524,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distance functions,scan when time allows,2022-01-20T09:43:20Z
issue,34951,"[feature request] Sparse (hybrid sparse-dense) output option for topk, min, max",https://github.com/pytorch/pytorch/issues/34951,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2022-01-20T06:24:16Z
issue,71475,The signature of `torch.nanmedian` in the doc is wrong,https://github.com/pytorch/pytorch/issues/71475,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nans and infs,scan when time allows,2022-01-20T03:25:39Z
issue,64744,Consolidate ProcessGroup allreduce_coalesced and allreduce,https://github.com/pytorch/pytorch/issues/64744,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, module: c10d, oncall: distributed",scan when time allows,2022-01-19T18:35:16Z
issue,71330,[RFC] Gossip SGD (as a DDP Communication Hook),https://github.com/pytorch/pytorch/issues/71330,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ddp, module: optimizer, oncall: distributed",scan when time allows,2022-01-19T05:02:31Z
issue,49795,ninja: build stopped: subcommand failed.,https://github.com/pytorch/pytorch/issues/49795,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2022-01-19T01:49:09Z
issue,71414,Gamma distribution returns some wrong extreme values,https://github.com/pytorch/pytorch/issues/71414,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: distributions, module: numerical-stability",scan when time allows,2022-01-19T01:00:13Z
issue,71383,Allow to shutdown persistent workers,https://github.com/pytorch/pytorch/issues/71383,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: multithreading,scan when time allows,2022-01-18T16:53:42Z
issue,71407,"Push to fork failed with cryptic ""refusing to allow a Personal Access Token to create or update workflow `.github/workflows/run_torchbench.yml` without `workflow` scope""",https://github.com/pytorch/pytorch/issues/71407,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2022-01-18T15:44:58Z
issue,71366,nn.Batchnorm1d input shape notation inconsistency,https://github.com/pytorch/pytorch/issues/71366,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: norms and normalization",scan when time allows,2022-01-18T15:13:55Z
issue,71331,Change the order of checks for tensor indexing errors,https://github.com/pytorch/pytorch/issues/71331,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: advanced indexing,scan when time allows,2022-01-18T14:59:57Z
issue,71322,Allow users to pass use_reentrant=False to checkpoint_sequential,https://github.com/pytorch/pytorch/issues/71322,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: activation checkpointing,scan when time allows,2022-01-18T14:57:39Z
issue,71305,Convert a tensor with type caffe2::TypeMeta to std::vector,https://github.com/pytorch/pytorch/issues/71305,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: meta tensors,scan when time allows,2022-01-18T14:55:47Z
issue,71357,[feature request] Multidim support for softmax/log_softmax/softmin,https://github.com/pytorch/pytorch/issues/71357,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: norms and normalization,scan when time allows,2022-01-18T14:53:57Z
issue,70735,named tensor doesn't work with deepcopy,https://github.com/pytorch/pytorch/issues/70735,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: named tensor,scan when time allows,2022-01-18T06:21:29Z
issue,50343,RFC: identify analogous NumPy operators when documenting PyTorch operators,https://github.com/pytorch/pytorch/issues/50343,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: numpy",scan when time allows,2022-01-17T19:25:29Z
issue,71029,"remote failure INTERNAL ASSERT FAILED at ""../torch/csrc/distributed/rpc/rref_context.cpp"":389",https://github.com/pytorch/pytorch/issues/71029,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: rpc, oncall: distributed",scan when time allows,2022-01-16T04:37:25Z
issue,62141,Dedupe code in functional optim classes,https://github.com/pytorch/pytorch/issues/62141,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, module: ddp",scan when time allows,2022-01-16T03:32:46Z
issue,71317,[torch.deploy] Using zipped torch modules in torch.deploy interpreter,https://github.com/pytorch/pytorch/issues/71317,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: package/deploy,scan when time allows,2022-01-14T18:47:48Z
issue,63085,Split up test_nn.py,https://github.com/pytorch/pytorch/issues/63085,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2022-01-14T15:35:35Z
issue,71266,"""Memory Leak"" when creating an iterator from a tensor",https://github.com/pytorch/pytorch/issues/71266,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: memory usage,scan when time allows,2022-01-14T13:39:45Z
issue,31546,[Feature Request] Make torch.solve output NaN for singular matrix,https://github.com/pytorch/pytorch/issues/31546,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: linear algebra,scan when time allows,2022-01-13T17:20:04Z
issue,43114,"out kwargs are sometimes inconsistent with returned named tuple field name (and in some cases, cannot be made consistent)",https://github.com/pytorch/pytorch/issues/43114,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: codegen,scan when time allows,2022-01-13T17:01:49Z
issue,71205,Possible security issue of `torch.hub.load`,https://github.com/pytorch/pytorch/issues/71205,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: hub,scan when time allows,2022-01-13T13:10:00Z
issue,49630,Bug in CosineAnnealingWarmRestarts,https://github.com/pytorch/pytorch/issues/49630,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2022-01-13T03:36:06Z
issue,71228,1.10.11 fails to compile libtorch_cpu with -fopenmp,https://github.com/pytorch/pytorch/issues/71228,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: undefined reference",scan when time allows,2022-01-12T19:06:26Z
issue,17525,cmake fails with -DBUILD_PYTHON=OFF -DUSE_NNPACK=OFF,https://github.com/pytorch/pytorch/issues/17525,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, oncall: releng",scan when time allows,2022-01-12T11:05:09Z
issue,53712,Poor support of `Optimizer.add_param_group`,https://github.com/pytorch/pytorch/issues/53712,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2022-01-12T08:08:36Z
issue,71152,clang format hash mismatched for linux64,https://github.com/pytorch/pytorch/issues/71152,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ci, module: lint",scan when time allows,2022-01-12T03:59:15Z
issue,65576,[LTC] Code-gen IRs and TorchScript lowerings needed by running full TorchBench,https://github.com/pytorch/pytorch/issues/65576,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: lazy, oncall: jit",scan when time allows,2022-01-11T20:46:10Z
issue,70995,Rewrite tests in test_nn to not depend on LAPACK,https://github.com/pytorch/pytorch/issues/70995,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: tests",scan when time allows,2022-01-11T19:23:16Z
issue,70608,Cannot compile C++ documentation: Sphynx assertion,https://github.com/pytorch/pytorch/issues/70608,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2022-01-11T17:00:03Z
issue,25137,Serialization does not work for quantized modules,https://github.com/pytorch/pytorch/issues/25137,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: serialization,scan when time allows,2022-01-11T16:52:13Z
issue,39411,Allow to_here to interrupt blocking wait in the case of rpc.remote timeout,https://github.com/pytorch/pytorch/issues/39411,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rpc,scan when time allows,2022-01-11T16:24:50Z
issue,70970,argmin/argmax incorrect doc for the first form,https://github.com/pytorch/pytorch/issues/70970,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: reductions",scan when time allows,2022-01-11T15:52:23Z
issue,71058,`torch.Tensor.where` cannot work when `y` is float,https://github.com/pytorch/pytorch/issues/71058,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: type promotion,scan when time allows,2022-01-11T15:36:54Z
issue,71076,Error in `torch.Tensor.logit` documentation,https://github.com/pytorch/pytorch/issues/71076,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2022-01-11T15:25:16Z
issue,70901,Accuracy problem of `torch.batch_norm_gather_stats_with_counts` when `running_mean` is half tensor,https://github.com/pytorch/pytorch/issues/70901,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: half, module: numerical-stability",scan when time allows,2022-01-11T14:56:56Z
issue,71134,Slowdown in torch.distributed.new_group when scaling to large clusters.,https://github.com/pytorch/pytorch/issues/71134,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: c10d, oncall: distributed",scan when time allows,2022-01-11T02:36:12Z
issue,71071,Feature Request: torch.special.ellipe,https://github.com/pytorch/pytorch/issues/71071,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: special,scan when time allows,2022-01-11T02:24:45Z
issue,70191,[libtorch] Loading in Java two differente libtorch_cpu.so from different versions fails,https://github.com/pytorch/pytorch/issues/70191,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, oncall: java",scan when time allows,2022-01-10T22:58:06Z
issue,46076,NCCL watchdog thread should log warnings about long-running GPU operations instead of silently hanging,https://github.com/pytorch/pytorch/issues/46076,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, module: c10d",scan when time allows,2022-01-10T21:50:16Z
issue,67571,[discussion] Have default size/shape equal to a scalar,https://github.com/pytorch/pytorch/issues/67571,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: random,scan when time allows,2022-01-10T19:59:28Z
issue,70397,`torch.empty_strided` works when the stride is negative!,https://github.com/pytorch/pytorch/issues/70397,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: tensor creation",scan when time allows,2022-01-10T18:15:49Z
issue,70398,`torch.broadcast_to` can create tensor with negative dimension.,https://github.com/pytorch/pytorch/issues/70398,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: shape checking",scan when time allows,2022-01-10T18:14:49Z
issue,71069,C++ torch::nn::Sequential clone() method overwrites child module names,https://github.com/pytorch/pytorch/issues/71069,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: nn",scan when time allows,2022-01-10T16:08:30Z
issue,70931,Redefinition of `cub` namespace misses Debug,https://github.com/pytorch/pytorch/issues/70931,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cuda",scan when time allows,2022-01-08T22:31:53Z
issue,30987,Remove `.data`,https://github.com/pytorch/pytorch/issues/30987,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2022-01-08T04:06:41Z
issue,70940,"ThreadLocalState::setThreadLocalState is not setting the ""enabled"" flag of SavedTensorDefaultHooks",https://github.com/pytorch/pytorch/issues/70940,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: multithreading,scan when time allows,2022-01-07T04:25:48Z
issue,59566,[discussion] Make docs consistent for torch.nn Modules,https://github.com/pytorch/pytorch/issues/59566,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: nn",scan when time allows,2022-01-06T23:25:35Z
issue,11578,Request to import pytest in test/*.py,https://github.com/pytorch/pytorch/issues/11578,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tests,scan when time allows,2022-01-06T23:22:34Z
issue,67653,"TSAN issue in autograd ""set_next_edges""",https://github.com/pytorch/pytorch/issues/67653,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: sanitizers",scan when time allows,2022-01-06T14:42:45Z
issue,6850,[feature request] More `index_*_` functionality and/or lambda functionality,https://github.com/pytorch/pytorch/issues/6850,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: scatter & gather ops,scan when time allows,2022-01-06T13:00:19Z
issue,8853,Todo functions and autograd supports for Sparse Tensor,https://github.com/pytorch/pytorch/issues/8853,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2022-01-06T12:44:49Z
issue,56836,Remove single device constraint from ProcessGroupNCCL profiling,https://github.com/pytorch/pytorch/issues/56836,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: c10d, module: nccl",scan when time allows,2022-01-06T09:16:56Z
issue,68179,torch.nn.Linear wrong type annotation for bias,https://github.com/pytorch/pytorch/issues/68179,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: typing",scan when time allows,2022-01-06T02:49:57Z
issue,70521,tmpxft_00008487_00000000-6_THCStorage.compute_86.cudafe1.cpp:(.text+0x60b): additional relocation overflows omitted from the output when build PyTorch 1.8.2 from source,https://github.com/pytorch/pytorch/issues/70521,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cuda",scan when time allows,2022-01-06T02:21:26Z
issue,42919,Legacy Python2 and early Python3 leftovers,https://github.com/pytorch/pytorch/issues/42919,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-01-06T02:20:13Z
issue,16014,[docs] Missing argument description (value) in scatter_ function documentation,https://github.com/pytorch/pytorch/issues/16014,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: scatter & gather ops",scan when time allows,2022-01-05T20:08:41Z
issue,69688,`torch.cuda.set_per_process_memory_fraction()` does not perform VRAM isolation,https://github.com/pytorch/pytorch/issues/69688,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: docs, module: memory usage",scan when time allows,2022-01-05T17:52:26Z
issue,55296,Build errors with USE_VULKAN=ON when cross-compiling for Android,https://github.com/pytorch/pytorch/issues/55296,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: android, module: build",scan when time allows,2022-01-05T16:55:16Z
issue,70242,`TensorIterator`: provide a two-argument version of `set_output`,https://github.com/pytorch/pytorch/issues/70242,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: structured kernels, module: tensoriterator",scan when time allows,2022-01-05T16:20:53Z
issue,64023,A modest proposal: delete arithmetic overloads from c10::Half,https://github.com/pytorch/pytorch/issues/64023,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: half,scan when time allows,2022-01-05T04:40:43Z
issue,55655,Performance debugging / warning mode,https://github.com/pytorch/pytorch/issues/55655,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: performance,scan when time allows,2022-01-05T04:39:27Z
issue,70487,Fusion of Convolution and BatchNorm,https://github.com/pytorch/pytorch/issues/70487,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cudnn,scan when time allows,2022-01-05T03:08:01Z
issue,69772,RuntimeError: t == DeviceType::CUDAINTERNAL ASSERT FAILED when trying to calculate gradients,https://github.com/pytorch/pytorch/issues/69772,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2022-01-04T14:58:32Z
issue,58741,`torch.size()` diverges from array API specification,https://github.com/pytorch/pytorch/issues/58741,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: deprecation, module: python array api",scan when time allows,2022-01-04T10:39:11Z
issue,70498,Multiple invalid summaries in torch.nn documentation page,https://github.com/pytorch/pytorch/issues/70498,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: nn",scan when time allows,2022-01-03T12:43:55Z
issue,70392,third_party/breakpad/ compilation failure,https://github.com/pytorch/pytorch/issues/70392,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: third_party,scan when time allows,2022-01-01T15:34:59Z
issue,39670,Force JIT to do type inference even when mypy annotated,https://github.com/pytorch/pytorch/issues/39670,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: typing, oncall: jit",scan when time allows,2021-12-30T15:08:12Z
issue,70342,Empty or NaN data pollute gradient even if they are not involved during backward,https://github.com/pytorch/pytorch/issues/70342,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: nans and infs",scan when time allows,2021-12-29T12:35:48Z
issue,32268,Numpy array functionality in torchscript.,https://github.com/pytorch/pytorch/issues/32268,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2021-12-28T09:08:19Z
issue,70265,"Shape parameter inconsistency in torch.Tensor.view, torch.reshape, torch.Tensor.reshape",https://github.com/pytorch/pytorch/issues/70265,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: viewing and reshaping",scan when time allows,2021-12-27T22:47:17Z
issue,42336,Audit quantization functions to ensure proper argument sizes,https://github.com/pytorch/pytorch/issues/42336,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2021-12-27T16:44:20Z
issue,39495,Misannotation of layer_norm parameters causes internal assert failure,https://github.com/pytorch/pytorch/issues/39495,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2021-12-27T16:42:29Z
issue,36426,Any reference to LPPool2d,https://github.com/pytorch/pytorch/issues/36426,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2021-12-27T12:37:22Z
issue,33800,Can't get module gradient in autograd.Function's custom backward when DataParallel is used,https://github.com/pytorch/pytorch/issues/33800,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: data parallel,scan when time allows,2021-12-26T22:23:38Z
issue,67136,addition of loss function RMSE in the torch.nn,https://github.com/pytorch/pytorch/issues/67136,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: loss, module: nn",scan when time allows,2021-12-24T15:15:15Z
issue,70241,CPU parallelization across batch has random faulty behavior on backward,https://github.com/pytorch/pytorch/issues/70241,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: multithreading,scan when time allows,2021-12-23T23:00:31Z
issue,70184,Optimization: convolution_backward doesn't always need to call .contiguous on certain inputs,https://github.com/pytorch/pytorch/issues/70184,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: performance",scan when time allows,2021-12-23T16:55:57Z
issue,70180,Can you provide the torch.trt module to directly convert the pytorch weights to tensorrt?,https://github.com/pytorch/pytorch/issues/70180,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: unknown,scan when time allows,2021-12-23T15:39:44Z
issue,70222,Grad strides do not match bucket view strides,https://github.com/pytorch/pytorch/issues/70222,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ddp,scan when time allows,2021-12-23T14:40:55Z
issue,70170,EmbeddingBag allows out of index ranges.!,https://github.com/pytorch/pytorch/issues/70170,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dependency bug,scan when time allows,2021-12-23T14:12:14Z
issue,56333,Unable to pip install PyTorch on M1 Mac - Errors,https://github.com/pytorch/pytorch/issues/56333,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: arm, module: build, module: macos",scan when time allows,2021-12-22T15:41:25Z
issue,69893,subclassing torch.Tensor,https://github.com/pytorch/pytorch/issues/69893,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-12-21T15:28:41Z
issue,69932,"Would pytorch like some free multi-factor authentication (MFA) tokens from Google & GitHub, via the OpenSSF?",https://github.com/pytorch/pytorch/issues/69932,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-12-20T18:27:31Z
issue,66073,[feature request] BatchNorm frozen mode in core,https://github.com/pytorch/pytorch/issues/66073,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: norms and normalization",scan when time allows,2021-12-20T16:23:08Z
issue,69822,Implement torch.*_like tensor creation functions on sparse inputs,https://github.com/pytorch/pytorch/issues/69822,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: sparse, module: tensor creation",scan when time allows,2021-12-19T18:28:31Z
issue,69892,[feature request] Support `like=` argument in tensor factory methods,https://github.com/pytorch/pytorch/issues/69892,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: numpy, module: tensor creation",scan when time allows,2021-12-19T11:33:49Z
issue,19953,Performance difference between 0.4.1 and 1.1.0,https://github.com/pytorch/pytorch/issues/19953,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: performance,scan when time allows,2021-12-18T18:46:20Z
issue,69831,Incorrect error for integer `out=` dtypes when a float is expected.,https://github.com/pytorch/pytorch/issues/69831,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: structured kernels, module: type promotion",scan when time allows,2021-12-18T02:43:06Z
issue,69433,`torch.transpose` should raise an error when indexing 0 for 0 dimensional tensor.,https://github.com/pytorch/pytorch/issues/69433,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: viewing and reshaping,scan when time allows,2021-12-17T15:24:48Z
issue,70040,Error in `torch.cdist` documentation,https://github.com/pytorch/pytorch/issues/70040,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: distance functions, module: docs",scan when time allows,2021-12-17T14:43:25Z
issue,69386,Port `normal` to structured kernel,https://github.com/pytorch/pytorch/issues/69386,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: structured kernels,scan when time allows,2021-12-17T05:00:49Z
issue,16954,torch.multiprocessing.pool.Pool broken,https://github.com/pytorch/pytorch/issues/16954,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: multiprocessing,scan when time allows,2021-12-15T16:04:00Z
issue,41081,Add a function to convert SyncBatchNorm layers back to BatchNorm Layers,https://github.com/pytorch/pytorch/issues/41081,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, module: nn",scan when time allows,2021-12-15T12:59:09Z
issue,55912,Support float_qparams in quantized_clone,https://github.com/pytorch/pytorch/issues/55912,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2021-12-15T05:17:23Z
issue,55230,Profile Optionals and Optional[Tensor] specifically,https://github.com/pytorch/pytorch/issues/55230,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2021-12-15T01:42:41Z
issue,60156,PEP-585 type annotations not supported for class-level annotations,https://github.com/pytorch/pytorch/issues/60156,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2021-12-14T15:52:50Z
issue,69213,torch.stack should have a pin_memory parameter,https://github.com/pytorch/pytorch/issues/69213,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: memory format,scan when time allows,2021-12-14T11:26:01Z
issue,53696,Support GPU/CPU communication in RPC,https://github.com/pytorch/pytorch/issues/53696,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: rpc, module: tensorpipe, oncall: distributed",scan when time allows,2021-12-14T08:52:53Z
issue,69721,Alternative to keras StringLookup in Pytorch,https://github.com/pytorch/pytorch/issues/69721,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2021-12-13T18:22:40Z
issue,69654,Fused AlphaFold 2 modules,https://github.com/pytorch/pytorch/issues/69654,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-12-13T17:43:01Z
issue,67761,Allow LRScheduler to take in param_groups directly without an optimizer,https://github.com/pytorch/pytorch/issues/67761,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: lrscheduler, module: optimizer",scan when time allows,2021-12-13T15:24:48Z
issue,68979,Support for arbitrary schedulers in SequentialLR,https://github.com/pytorch/pytorch/issues/68979,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: lrscheduler, module: optimizer",scan when time allows,2021-12-13T15:23:32Z
issue,59137,[autograd] `fill_` gradgradcheck raises RuntimeError,https://github.com/pytorch/pytorch/issues/59137,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2021-12-13T14:44:37Z
issue,51575,[doc] being able to switch version while keeping the same doc,https://github.com/pytorch/pytorch/issues/51575,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: doc infra, module: docs",scan when time allows,2021-12-10T20:51:12Z
issue,69627,c2r fft input generation,https://github.com/pytorch/pytorch/issues/69627,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: fft, module: tests",scan when time allows,2021-12-10T20:27:34Z
issue,42843,Documentation mistake of Adam in v1.6.0?,https://github.com/pytorch/pytorch/issues/42843,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: optimizer",scan when time allows,2021-12-10T16:09:26Z
issue,42353,Error while trying to install pytorch on pycharm,https://github.com/pytorch/pytorch/issues/42353,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2021-12-10T03:18:22Z
issue,59309,CUDA memory error when using torch.device('cpu'),https://github.com/pytorch/pytorch/issues/59309,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: memory usage",scan when time allows,2021-12-09T22:03:49Z
issue,68909,Bug about Dropout CUDA Kernel,https://github.com/pytorch/pytorch/issues/68909,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: nn",scan when time allows,2021-12-09T20:30:36Z
issue,33388,Sobol engine generates out-of-bounds samples after drawing too many samples,https://github.com/pytorch/pytorch/issues/33388,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: random,scan when time allows,2021-12-08T21:24:04Z
issue,33086,F.max_pool*d/F.min_pool*d should support integer dtypes and bool tensors,https://github.com/pytorch/pytorch/issues/33086,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: pooling,scan when time allows,2021-12-08T11:15:12Z
issue,69453,Jacobian elliptic functions,https://github.com/pytorch/pytorch/issues/69453,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: special,scan when time allows,2021-12-08T05:21:33Z
issue,69452,Complete elliptic integral of the first kind,https://github.com/pytorch/pytorch/issues/69452,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: special,scan when time allows,2021-12-08T05:21:09Z
issue,69451,Exponentially scalable modified Bessel function of the second kind,https://github.com/pytorch/pytorch/issues/69451,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: special,scan when time allows,2021-12-08T05:20:17Z
issue,69363,"Citation request for ""probabilities for each class"" in the doc's description on cross-entropy",https://github.com/pytorch/pytorch/issues/69363,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: nn",scan when time allows,2021-12-07T22:18:37Z
issue,69467,Distribution `cross_entropy` method,https://github.com/pytorch/pytorch/issues/69467,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distributions,scan when time allows,2021-12-07T19:36:27Z
issue,69469,Distribution `covariance` property,https://github.com/pytorch/pytorch/issues/69469,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distributions,scan when time allows,2021-12-07T19:32:45Z
issue,46024,Reproducibility breaks down with weighted Cross Entropy loss,https://github.com/pytorch/pytorch/issues/46024,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: determinism, module: loss, module: numerical-reproducibility",scan when time allows,2021-12-07T18:13:38Z
issue,55464,Support quantized functional linear/conv operators with 32b output,https://github.com/pytorch/pytorch/issues/55464,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2021-12-07T15:44:08Z
issue,29657,Improved detection of repeated observers,https://github.com/pytorch/pytorch/issues/29657,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2021-12-07T15:19:18Z
issue,58171,"Error while converting model for NNAPI ""Unsupported node kind ('quantized::batch_norm2d')""",https://github.com/pytorch/pytorch/issues/58171,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: android, oncall: mobile",scan when time allows,2021-12-06T20:54:51Z
issue,47546,Whats Pytorch's policy on adding support for a wider range of hardwares for training and inference?,https://github.com/pytorch/pytorch/issues/47546,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: backend,scan when time allows,2021-12-04T11:33:13Z
issue,69199,Protocol buffers library mismatch.,https://github.com/pytorch/pytorch/issues/69199,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: protobuf, module: tensorboard",scan when time allows,2021-12-03T13:54:18Z
issue,62933,Compilation error with gcc11 and libstdc++11 (thread_id != operator removed),https://github.com/pytorch/pytorch/issues/62933,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: mkldnn,scan when time allows,2021-12-03T09:35:06Z
issue,42502,[proposal] batch mode for randperm,https://github.com/pytorch/pytorch/issues/42502,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: random,scan when time allows,2021-12-03T04:01:20Z
issue,60932,Is Python version of Docker image on DockerHub downgraded?,https://github.com/pytorch/pytorch/issues/60932,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docker,scan when time allows,2021-12-03T01:11:25Z
issue,66410,CIFAR10 doesn't work on M1 MacBook,https://github.com/pytorch/pytorch/issues/66410,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: arm,scan when time allows,2021-12-02T14:33:57Z
issue,16899,"Give clearer guidance about multithreading in PyTorch, and how to disable it",https://github.com/pytorch/pytorch/issues/16899,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: multithreading",scan when time allows,2021-12-02T13:16:38Z
issue,4247,Feature request: sparse matrix max(axis),https://github.com/pytorch/pytorch/issues/4247,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2021-12-01T17:45:35Z
issue,67745,A `sieve` operation for separating values of a tensor into two tensors based on a condition.,https://github.com/pytorch/pytorch/issues/67745,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ux, module: viewing and reshaping",scan when time allows,2021-12-01T07:29:53Z
issue,68931,Dead lock happened ran pytorch 1.9.0 cuda11.2 on xeon gold 6326 cpu,https://github.com/pytorch/pytorch/issues/68931,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: data parallel, oncall: distributed",scan when time allows,2021-12-01T03:36:40Z
issue,59875,[ROCm] test_gather_stress_cuda is flaky,https://github.com/pytorch/pytorch/issues/59875,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: flaky-tests, module: rocm",scan when time allows,2021-11-30T22:11:04Z
issue,67937,Add support for LTC ops needed by backwards/autograd/autocast,https://github.com/pytorch/pytorch/issues/67937,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: lazy,scan when time allows,2021-11-30T17:36:55Z
issue,68925,How to implement `bucket_by_sequence_length` with IterableDataset and DataLoader,https://github.com/pytorch/pytorch/issues/68925,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: data, module: dataloader",scan when time allows,2021-11-30T15:32:48Z
issue,52743,torch.searchsorted issues,https://github.com/pytorch/pytorch/issues/52743,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: numpy, module: sorting and selection",scan when time allows,2021-11-30T12:09:08Z
issue,68935,repeat_interleave hangs in a forked subprocess,https://github.com/pytorch/pytorch/issues/68935,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: multiprocessing,scan when time allows,2021-11-30T01:08:06Z
issue,68967,[feature request] torch.argminmax,https://github.com/pytorch/pytorch/issues/68967,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-11-29T23:29:24Z
issue,62654,Checkpointing without re-entrant autograd,https://github.com/pytorch/pytorch/issues/62654,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2021-11-29T21:05:34Z
issue,68752,`softmin` and `softmax` operators support different input dtypes based on whether the `dtype` kwarg is passed,https://github.com/pytorch/pytorch/issues/68752,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: ux",scan when time allows,2021-11-29T18:20:48Z
issue,45412,Autocast but there are no kernels registered for this dispatch key. The operator is p�8ߎU Aborted (core dumped),https://github.com/pytorch/pytorch/issues/45412,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision),scan when time allows,2021-11-29T08:53:14Z
issue,27300,"Tensorboard add image with boxes, labels, and confidence scores.",https://github.com/pytorch/pytorch/issues/27300,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tensorboard,scan when time allows,2021-11-28T15:53:06Z
issue,29044,Add Wishart and inverse-Wishart distributions,https://github.com/pytorch/pytorch/issues/29044,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distributions,scan when time allows,2021-11-28T01:05:21Z
issue,66083,[feature request] Allow broadcasting for F.huber_loss and potentially other losses,https://github.com/pytorch/pytorch/issues/66083,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: loss, module: shape checking",scan when time allows,2021-11-26T00:20:12Z
issue,67751,Remove logic for constructing symbolic shapes in Profiler,https://github.com/pytorch/pytorch/issues/67751,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-11-25T00:23:05Z
issue,67687,"Trying to quantize and save a pre-trained deberta model, where i get a runtime error",https://github.com/pytorch/pytorch/issues/67687,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2021-11-23T18:32:36Z
issue,52747,[proposal] Make torchvision and other libraries' docs searchable via main docs search,https://github.com/pytorch/pytorch/issues/52747,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: doc infra, module: docs",scan when time allows,2021-11-23T17:19:27Z
issue,68622,"torch.{tensor, Tensor, LongTensor, ...} isn't captured under `enable_python_mode()`",https://github.com/pytorch/pytorch/issues/68622,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: __torch_dispatch__,scan when time allows,2021-11-22T14:27:45Z
issue,68422,There is a problem with GPU memory reclamation,https://github.com/pytorch/pytorch/issues/68422,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: memory usage,scan when time allows,2021-11-21T04:01:03Z
issue,18796,[Feature Request] Flattened indices option for max pooling,https://github.com/pytorch/pytorch/issues/18796,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: pooling",scan when time allows,2021-11-19T23:25:57Z
issue,17538,ONNX->TensorRT parser library duplicated,https://github.com/pytorch/pytorch/issues/17538,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-11-19T23:10:14Z
issue,55374,[RFC] Extend Autocast to CPU/CUDA with BF16 data type,https://github.com/pytorch/pytorch/issues/55374,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision), module: bfloat16",scan when time allows,2021-11-19T18:50:10Z
issue,47311,Quantized modules should properly implement __getstate__ and __setstate__ (copy.deepcopy doesn't work on quantized model),https://github.com/pytorch/pytorch/issues/47311,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2021-11-18T23:25:12Z
issue,68610,`torch.unique_consecutive`: passing positional optional arguments results in empty tensors,https://github.com/pytorch/pytorch/issues/68610,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sorting and selection,scan when time allows,2021-11-18T20:54:50Z
issue,68575,[ux] Print tensor's dtype by default,https://github.com/pytorch/pytorch/issues/68575,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ux,scan when time allows,2021-11-18T20:00:05Z
issue,68512,'replicate' padding in convolution is 77 times slower on cpu than 'zeros',https://github.com/pytorch/pytorch/issues/68512,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: padding, module: performance",scan when time allows,2021-11-18T18:08:13Z
issue,68420,`nn.functional.max_unpool(2|3)d`: failing shape check for correct inputs (with `dilation > 1`) with specified `output_size`,https://github.com/pytorch/pytorch/issues/68420,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: pooling",scan when time allows,2021-11-18T15:24:57Z
issue,68519,dataloader will miss batch data when num worker>0,https://github.com/pytorch/pytorch/issues/68519,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2021-11-18T14:59:42Z
issue,68507,Multiprocessing hang and queue short circuiting,https://github.com/pytorch/pytorch/issues/68507,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: deadlock, module: multiprocessing",scan when time allows,2021-11-18T09:05:49Z
issue,9624,Request: Pycuda interoperability,https://github.com/pytorch/pytorch/issues/9624,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-11-17T08:39:29Z
issue,68430,Strides issue on unsqueezed channels last tensor,https://github.com/pytorch/pytorch/issues/68430,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: memory format,scan when time allows,2021-11-17T08:14:47Z
issue,68498,Review disabling tests workflow,https://github.com/pytorch/pytorch/issues/68498,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-11-17T01:23:49Z
issue,62926,cudnn_convolution_add_relu fails under basic conditions,https://github.com/pytorch/pytorch/issues/62926,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cudnn, module: dependency bug",scan when time allows,2021-11-17T00:02:30Z
issue,68423,`torch.utils.checkpoint.checkpoint_sequential` is not optimal,https://github.com/pytorch/pytorch/issues/68423,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: activation checkpointing,scan when time allows,2021-11-17T00:01:32Z
issue,68474,[LTC][BE] Fix all compile warnings,https://github.com/pytorch/pytorch/issues/68474,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-11-16T23:54:43Z
issue,68475,[LTC][BE] Make the compile flags more strict,https://github.com/pytorch/pytorch/issues/68475,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-11-16T23:54:33Z
issue,68337,"`nn.functional.max_unpool{n}d`: shape checks fail with `output_size=(C, ...)`.",https://github.com/pytorch/pytorch/issues/68337,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: pooling",scan when time allows,2021-11-16T18:04:30Z
issue,47351,no member named 'beta1' in 'torch::optim::AdamOptions',https://github.com/pytorch/pytorch/issues/47351,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: optimizer",scan when time allows,2021-11-16T16:03:14Z
issue,67535,[LTC] `UncachedCompile` numbers increases for several iterations until fixed,https://github.com/pytorch/pytorch/issues/67535,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: lazy,scan when time allows,2021-11-16T04:54:54Z
issue,68315,[docs] Strange signatures for torch.autocast,https://github.com/pytorch/pytorch/issues/68315,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision), module: docs",scan when time allows,2021-11-15T22:14:35Z
issue,68288,List of issues to unblock dynamic shapes in Lazy Tensor,https://github.com/pytorch/pytorch/issues/68288,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-11-15T22:09:59Z
issue,68105,Some type combinations of cublas gemm are not supported when they should,https://github.com/pytorch/pytorch/issues/68105,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: linear algebra, module: type promotion",scan when time allows,2021-11-15T17:14:44Z
issue,53879,Linear algebra GPU library function bug tracking issue [magma/cusolver/cublas],https://github.com/pytorch/pytorch/issues/53879,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: linear algebra",scan when time allows,2021-11-15T16:16:25Z
issue,67968,"unresolved external symbol ""__declspec(dllimport) struct _object * __cdecl THPVariable_Wrap",https://github.com/pytorch/pytorch/issues/67968,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2021-11-15T14:21:06Z
issue,63479,Cannot include extension.h under Windows - linker error THPVariable_Wrap,https://github.com/pytorch/pytorch/issues/63479,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: cpp-extensions",scan when time allows,2021-11-15T00:12:58Z
issue,49394,Apparent Memory Leak with torch.as_tensor,https://github.com/pytorch/pytorch/issues/49394,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: memory usage, module: numpy, module: tensor creation",scan when time allows,2021-11-14T23:55:38Z
issue,32358,torch.bartlett_window not jitable,https://github.com/pytorch/pytorch/issues/32358,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2021-11-14T21:51:49Z
issue,32365,[jit] Calling `torch.jit.script` on `@staticmethod`s which in turn call other `@staticmethod`s results in a `TypeError`,https://github.com/pytorch/pytorch/issues/32365,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2021-11-14T21:51:45Z
issue,32039,Internal assert failure for user IValue unwrapping,https://github.com/pytorch/pytorch/issues/32039,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2021-11-14T21:51:43Z
issue,19628,[jit] nn.LSTM errors in nn.ScriptModule,https://github.com/pytorch/pytorch/issues/19628,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2021-11-14T21:51:40Z
issue,25591,IValue pickle does not work properly if an empty tensor table is not provided,https://github.com/pytorch/pytorch/issues/25591,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2021-11-14T21:51:38Z
issue,21275,weight_norm is not supported in TorchScript,https://github.com/pytorch/pytorch/issues/21275,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2021-11-14T21:51:36Z
issue,20591,@ignore annotation for user defined type,https://github.com/pytorch/pytorch/issues/20591,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2021-11-14T21:51:34Z
issue,35074,"[JIT] If a python function type comment is referring to a wrong type, JIT frontend gives a not helpful error message",https://github.com/pytorch/pytorch/issues/35074,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2021-11-14T21:50:55Z
issue,32300,[jit] Various problems calling `@staticmethod`s in 1.4.0,https://github.com/pytorch/pytorch/issues/32300,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2021-11-14T21:50:53Z
issue,35520,TorchScript can't use lists in conditionals,https://github.com/pytorch/pytorch/issues/35520,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2021-11-14T21:50:52Z
issue,56110,caffe2/utils/signal_handler.cc is failing to symbolize libtorch_cpu.so,https://github.com/pytorch/pytorch/issues/56110,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2021-11-14T21:50:47Z
issue,56397,Symbolicate crashes from releases,https://github.com/pytorch/pytorch/issues/56397,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-11-14T21:50:45Z
issue,55952,Compiler warnings tracking,https://github.com/pytorch/pytorch/issues/55952,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build warnings,scan when time allows,2021-11-14T21:50:43Z
issue,57404,[lint] Make sure that all CMake options have a corresponding output in the summary,https://github.com/pytorch/pytorch/issues/57404,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-11-14T21:50:38Z
issue,57987,`type: ignore` everything,https://github.com/pytorch/pytorch/issues/57987,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-11-14T21:50:36Z
issue,58930,[local lint] Shellcheck quicklint file filter is incorrect,https://github.com/pytorch/pytorch/issues/58930,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-11-14T21:50:35Z
issue,59682,Add a lint rule to recommend TORCH_CHECK(false) over throw std::runtime_error,https://github.com/pytorch/pytorch/issues/59682,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-11-14T21:50:32Z
issue,60924,[rfc][local lint] Create lint runner as a setup.py target,https://github.com/pytorch/pytorch/issues/60924,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-11-14T21:50:28Z
issue,60999,Merge seemethere/upload-artifact-s3 and driazati/upload-artifact-s3,https://github.com/pytorch/pytorch/issues/60999,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-11-14T21:50:26Z
issue,61079,"[clang-tidy] Add a custom rule for AT_ERROR -> TORCH_CHECK(false, ...)",https://github.com/pytorch/pytorch/issues/61079,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-11-14T21:50:24Z
issue,61384,Merge fork of upload-artifact-s3 to a single commit,https://github.com/pytorch/pytorch/issues/61384,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-11-14T21:50:20Z
issue,62673,Make printing to stdout / stderr in tests in CI an error,https://github.com/pytorch/pytorch/issues/62673,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-11-14T21:50:18Z
issue,36922,Multiprocess DataLoader with DLPack conversion sometimes corrupts memory,https://github.com/pytorch/pytorch/issues/36922,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: multiprocessing, module: serialization, module: xnnpack",scan when time allows,2021-11-13T17:24:18Z
issue,55752,"[question] Influence of divisibility of B, C, T by 16 on Conv1d (and Conv2d perf) with CuDNN, including presence of padding",https://github.com/pytorch/pytorch/issues/55752,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: performance",scan when time allows,2021-11-13T04:33:29Z
issue,68298,[Lazy] Add torch.autograd.Function wrappers for the following ops in order to support dynamic shapes,https://github.com/pytorch/pytorch/issues/68298,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-11-12T23:27:59Z
issue,68287,[Lazy] Add torch.autograd.Function wrappers for the following ops in order to support dynamic shapes,https://github.com/pytorch/pytorch/issues/68287,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-11-12T22:16:43Z
issue,68204,Feature Request: Scriptable Python Source-of-Truth for operator shape functions,https://github.com/pytorch/pytorch/issues/68204,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-11-12T22:03:28Z
issue,68193,clang-tidy shows false positive failure on PRs that rename files,https://github.com/pytorch/pytorch/issues/68193,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ci, module: lint",scan when time allows,2021-11-11T22:24:54Z
issue,68165,Upgrade NCCL2.11.4 in the new PyTorch,https://github.com/pytorch/pytorch/issues/68165,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2021-11-11T15:51:10Z
issue,68163,wsl2 install failed from source code,https://github.com/pytorch/pytorch/issues/68163,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: mkl",scan when time allows,2021-11-11T15:50:35Z
issue,65950,"Storage `is` should work if the storages ""are the same""",https://github.com/pytorch/pytorch/issues/65950,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-11-10T16:24:56Z
issue,67333,Feature request: Complex Number Support for Special Functions,https://github.com/pytorch/pytorch/issues/67333,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: complex, module: special",scan when time allows,2021-11-10T12:24:17Z
issue,67594,Feature request: set_grad_enabled(*mods_or_params) as a safe context manager,https://github.com/pytorch/pytorch/issues/67594,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: nn",scan when time allows,2021-11-10T11:44:49Z
issue,67482,Improve docs coverage testing,https://github.com/pytorch/pytorch/issues/67482,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2021-11-10T09:52:02Z
issue,63004,ignore_index does not work as described with nn.CrossEntropyLoss,https://github.com/pytorch/pytorch/issues/63004,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2021-11-10T05:47:34Z
issue,65657,Transpose channels when exporting to ONNX,https://github.com/pytorch/pytorch/issues/65657,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2021-11-09T23:00:25Z
issue,13300,Use a dill-based multiprocessing library and serialization,https://github.com/pytorch/pytorch/issues/13300,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: multiprocessing,scan when time allows,2021-11-09T17:05:25Z
issue,68046,"register_ Hook causes the CUDA out of memory, and remove() is useless",https://github.com/pytorch/pytorch/issues/68046,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2021-11-09T16:50:58Z
issue,64897,Either support torch.mean on BoolTensors or fix the error message,https://github.com/pytorch/pytorch/issues/64897,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: boolean tensor, module: error checking, module: reductions",scan when time allows,2021-11-09T10:47:48Z
issue,51535,Why isn't there a pip-installable PyTorch for Raspberry Pi?,https://github.com/pytorch/pytorch/issues/51535,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: arm, module: binaries",scan when time allows,2021-11-09T08:55:11Z
issue,47433,unbalanced gpu memory when using DistributedDataParallel,https://github.com/pytorch/pytorch/issues/47433,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2021-11-09T05:27:15Z
issue,62034,cudnn_batch_norm_backward is extremely imprecise for some input shapes,https://github.com/pytorch/pytorch/issues/62034,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cudnn,scan when time allows,2021-11-09T05:14:27Z
issue,67989,the abnormal display of bottleneck of resnet in tensorboard,https://github.com/pytorch/pytorch/issues/67989,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tensorboard,scan when time allows,2021-11-08T21:23:15Z
issue,45823,"Traceback for ""Warning: Mixed memory format inputs detected while calling the operator.""",https://github.com/pytorch/pytorch/issues/45823,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: memory format,scan when time allows,2021-11-08T02:59:09Z
issue,66775,Feature request: FFT operations on Metal,https://github.com/pytorch/pytorch/issues/66775,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ci, module: fft, module: macos, oncall: mobile",scan when time allows,2021-11-07T13:06:58Z
issue,67879,AMP does not improve performance,https://github.com/pytorch/pytorch/issues/67879,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision),scan when time allows,2021-11-05T20:14:49Z
issue,65702,Indicate support for more general use of `pos_weight` in binary cross entropy,https://github.com/pytorch/pytorch/issues/65702,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: nn",scan when time allows,2021-11-05T17:35:57Z
issue,56838,Comprehensively test NCCL's `get_future()` API,https://github.com/pytorch/pytorch/issues/56838,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2021-11-05T17:33:17Z
issue,66874,Error using _stateless version of Module,https://github.com/pytorch/pytorch/issues/66874,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2021-11-05T17:32:13Z
issue,31994,Trying to disable cuda to run torch on OSX run it on CPU,https://github.com/pytorch/pytorch/issues/31994,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cuda",scan when time allows,2021-11-05T12:56:36Z
issue,61636,Enhance Distributed `get_future()` docs,https://github.com/pytorch/pytorch/issues/61636,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2021-11-04T19:19:32Z
issue,67758,Building libtorch from source requires too much RAM,https://github.com/pytorch/pytorch/issues/67758,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cuda",scan when time allows,2021-11-04T13:44:30Z
issue,67772,Functions are rendered incorrectly,https://github.com/pytorch/pytorch/issues/67772,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: docs",scan when time allows,2021-11-03T16:21:57Z
issue,45042,"torch.distributed launch.py is hanged. (pid, sts) = os.waitpid(self.pid, wait_flags)",https://github.com/pytorch/pytorch/issues/45042,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2021-11-03T14:26:55Z
issue,67008,kwonly arguments without defaults don't work with test_overrides.py,https://github.com/pytorch/pytorch/issues/67008,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tests,scan when time allows,2021-11-02T12:26:22Z
issue,67589,Expose the `OptState` in `torch.cuda.amp.GradScaler`,https://github.com/pytorch/pytorch/issues/67589,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision),scan when time allows,2021-11-02T11:45:09Z
issue,67642,Is there a pre-built pytorch package for CUDA 10.1?,https://github.com/pytorch/pytorch/issues/67642,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2021-11-02T11:34:22Z
issue,32822,Error tracing custom autograd.Function,https://github.com/pytorch/pytorch/issues/32822,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2021-11-01T20:49:24Z
issue,67481,Batchnorm2D does behaves differently with different shapes when batch_size=1,https://github.com/pytorch/pytorch/issues/67481,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: norms and normalization",scan when time allows,2021-11-01T17:50:24Z
issue,67540,LTC asynchronous executor is not synced with python code and ignores context manager.,https://github.com/pytorch/pytorch/issues/67540,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: lazy,scan when time allows,2021-10-31T14:05:13Z
issue,67493,XLA not being tested in TestAutogradDeviceType,https://github.com/pytorch/pytorch/issues/67493,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: tests, module: xla",scan when time allows,2021-10-29T23:00:54Z
issue,66284,Memory leak issue with pytorch_java_only 1.9.0 and libtorch 1.9.0+cpu,https://github.com/pytorch/pytorch/issues/66284,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: android, module: memory usage, oncall: java, oncall: mobile",scan when time allows,2021-10-29T15:58:21Z
issue,54630,torch-tensor-repr gdb macro doesn't seem to work in opt mode,https://github.com/pytorch/pytorch/issues/54630,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-10-29T07:32:36Z
issue,67312,ProcessGroupNCCLTest.testSequenceNumInit is failing on 4xlarge,https://github.com/pytorch/pytorch/issues/67312,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: c10d, oncall: distributed",scan when time allows,2021-10-28T18:17:34Z
issue,67439,Exception in thread when using dataloader,https://github.com/pytorch/pytorch/issues/67439,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dataloader, module: multithreading",scan when time allows,2021-10-28T18:04:25Z
issue,61030,[Docker] Incompatible torchvision for 1.9.0-cuda10.2-cudnn7-runtime tag.,https://github.com/pytorch/pytorch/issues/61030,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docker, module: vision",scan when time allows,2021-10-28T05:34:36Z
issue,67377,[LTC] Replace mv with addmv in ts_native_functions.yaml,https://github.com/pytorch/pytorch/issues/67377,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-10-27T22:43:58Z
issue,67390,Missing doc for torch.distributions functions,https://github.com/pytorch/pytorch/issues/67390,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: distributions, module: docs",scan when time allows,2021-10-27T22:43:35Z
issue,67388,Missing doc for torch.cuda functions,https://github.com/pytorch/pytorch/issues/67388,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: docs",scan when time allows,2021-10-27T22:39:50Z
issue,67387,Missing doc for torch.autograd functions,https://github.com/pytorch/pytorch/issues/67387,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: docs",scan when time allows,2021-10-27T22:36:36Z
issue,63802,[torch.distributed.launch|run] Hangs on SIGINT when using a TCPStore backed rdzv_backend,https://github.com/pytorch/pytorch/issues/63802,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: c10d, oncall: distributed",scan when time allows,2021-10-27T21:17:06Z
issue,67279,register_dispatch_key should provide a way to tell if a native function has a no-op meta kernel,https://github.com/pytorch/pytorch/issues/67279,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dispatch, module: lazy",scan when time allows,2021-10-27T13:37:39Z
issue,67284,Failed to build with master.,https://github.com/pytorch/pytorch/issues/67284,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: mkl",scan when time allows,2021-10-27T13:30:57Z
issue,67208,TestForeachCUDA.test_binary_op_tensorlists_fastpath__foreach_add_cuda_bool and TestForeachCUDA.test_pointwise_op_fastpath__foreach_addcmul_cuda_uint8 fail intermittently,https://github.com/pytorch/pytorch/issues/67208,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: mta, module: tests",scan when time allows,2021-10-27T13:29:55Z
issue,63846,[torch.distributed] Make dynamic_rendezvous log handler configurable,https://github.com/pytorch/pytorch/issues/63846,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: elastic, oncall: r2p",scan when time allows,2021-10-27T04:46:32Z
issue,66755,Domain Transformation APIs for LibTorch and LibTorch-Lite,https://github.com/pytorch/pytorch/issues/66755,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: data, module: vision",scan when time allows,2021-10-26T19:44:07Z
issue,67071,Forward results vary depending on batch size on A100 machine,https://github.com/pytorch/pytorch/issues/67071,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: cuda, module: numerical-reproducibility",scan when time allows,2021-10-26T16:22:01Z
issue,67153,Build not working with cuda 11.5,https://github.com/pytorch/pytorch/issues/67153,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2021-10-26T15:44:26Z
issue,67168,Multiply two named tensor causes RuntimeError,https://github.com/pytorch/pytorch/issues/67168,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: named tensor,scan when time allows,2021-10-26T15:35:03Z
issue,58172,Mobile Android: Could not run 'aten::quantize_per_tensor' with arguments from the 'Vulkan' backend,https://github.com/pytorch/pytorch/issues/58172,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: android, oncall: mobile",scan when time allows,2021-10-25T22:29:42Z
issue,58548,Pixel shuffle support for PyTorch Mobile NN-API,https://github.com/pytorch/pytorch/issues/58548,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: android, oncall: mobile",scan when time allows,2021-10-25T22:28:51Z
issue,63223,[bug] torch.distributed.elastic logging: Failed to print the first info statement,https://github.com/pytorch/pytorch/issues/63223,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: elastic, oncall: r2p",scan when time allows,2021-10-25T17:57:27Z
issue,64636,[docs or feature request] torchelastic: OOM recovery / skipping batches (e.g. if inf loss or nan gradients),https://github.com/pytorch/pytorch/issues/64636,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ddp, module: docs, oncall: r2p",scan when time allows,2021-10-25T17:55:40Z
issue,66992,Bug? :Run torch.unique twice get different running time?,https://github.com/pytorch/pytorch/issues/66992,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: benchmark, module: performance, module: sorting and selection",scan when time allows,2021-10-25T16:12:27Z
issue,67187,Expose `isDifferentiableType` to python,https://github.com/pytorch/pytorch/issues/67187,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2021-10-25T15:24:46Z
issue,66963,Significantly difference in execution time when convolution is run as nn.Conv2d and as nn.Sequential,https://github.com/pytorch/pytorch/issues/66963,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: cuda, module: performance",scan when time allows,2021-10-21T14:04:22Z
issue,32695,torch.tensordot has inconsistent signature with torch script,https://github.com/pytorch/pytorch/issues/32695,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2021-10-21T12:11:42Z
issue,66902,Modify Dr. CI so it could detect runner disconnection failures,https://github.com/pytorch/pytorch/issues/66902,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2021-10-20T18:09:18Z
issue,66921,readme not update,https://github.com/pytorch/pytorch/issues/66921,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2021-10-20T15:21:15Z
issue,66907,torch.chunk return type may not be documented correctly,https://github.com/pytorch/pytorch/issues/66907,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2021-10-20T15:16:08Z
issue,66936,Lifetime issues when recording external CUDA streams with the caching allocator,https://github.com/pytorch/pytorch/issues/66936,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2021-10-20T14:53:19Z
issue,66868,torch.triu behaves differently when diagonal out of range,https://github.com/pytorch/pytorch/issues/66868,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: error checking,scan when time allows,2021-10-19T16:55:22Z
issue,66875,"Make the evaluated value of function f(x) accessible from `torch.autograd.functional.jacobian(f,x)`",https://github.com/pytorch/pytorch/issues/66875,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2021-10-19T16:53:40Z
issue,66824,Migrate C++ tests to Python runner,https://github.com/pytorch/pytorch/issues/66824,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tests,scan when time allows,2021-10-19T16:45:02Z
issue,66751,torch.nn.functional.embedding behave differently in two cases of cpu and cuda,https://github.com/pytorch/pytorch/issues/66751,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: embedding, module: error checking",scan when time allows,2021-10-19T14:40:05Z
issue,65526,[docs] Better explain signature of toch.randn,https://github.com/pytorch/pytorch/issues/65526,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2021-10-18T17:44:39Z
issue,66640,Static Build of Libtorch not linking correctly,https://github.com/pytorch/pytorch/issues/66640,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2021-10-18T15:43:44Z
issue,66750,torch.nn.EmbeddingBag behave differently in two cases of cpu and cuda,https://github.com/pytorch/pytorch/issues/66750,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: embedding, module: error checking",scan when time allows,2021-10-18T15:42:53Z
issue,20056,Creation of too big multidimensional array returns empty tensor.,https://github.com/pytorch/pytorch/issues/20056,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: error checking, module: tensor creation",scan when time allows,2021-10-17T22:18:41Z
issue,44511,"LR scheduler throws warning when using scaler.step instead of optimizer.step, and when saving optimizer state",https://github.com/pytorch/pytorch/issues/44511,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision),scan when time allows,2021-10-17T13:06:54Z
issue,53462,Second order derivatives for F.grid_sample(),https://github.com/pytorch/pytorch/issues/53462,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-10-16T08:30:13Z
issue,56707,[TorchScript JIT] `dtype` propagation pass (meta-issue),https://github.com/pytorch/pytorch/issues/56707,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-10-14T20:24:06Z
issue,43127,`onnxifi_load` symbol undefined in `libtorch_cpu.a` library built from source,https://github.com/pytorch/pytorch/issues/43127,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-10-14T19:18:00Z
issue,65837,NVFuser Examples to Help with Integration with Lazy Tensor Core,https://github.com/pytorch/pytorch/issues/65837,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: lazy,scan when time allows,2021-10-14T02:20:08Z
issue,66563,Give a better error message when REGISTER_DISPATCH is used in improper context,https://github.com/pytorch/pytorch/issues/66563,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dispatch, module: internals",scan when time allows,2021-10-13T23:10:10Z
issue,66489,The uniform operator param names in the C++ impl use python keywords,https://github.com/pytorch/pytorch/issues/66489,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: numpy,scan when time allows,2021-10-13T22:41:22Z
issue,66460,Inconsistencies in LSTM outputs when processing sequence stepwise on CPU,https://github.com/pytorch/pytorch/issues/66460,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: numerical-stability, module: rnn",scan when time allows,2021-10-13T22:05:44Z
issue,66223,Data Loader tests hang when run in ASAN test job,https://github.com/pytorch/pytorch/issues/66223,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dataloader, module: tests",scan when time allows,2021-10-13T18:31:51Z
issue,66366,Feature Request: Add constant padding_mode for grid_sample,https://github.com/pytorch/pytorch/issues/66366,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: interpolation, module: nn, module: padding",scan when time allows,2021-10-13T16:07:11Z
issue,66403,[opinfo] Confusing interface for `ops` decorator,https://github.com/pytorch/pytorch/issues/66403,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tests,scan when time allows,2021-10-13T07:48:38Z
issue,51965,Feature Request:Fast Cosine Transform,https://github.com/pytorch/pytorch/issues/51965,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fft,scan when time allows,2021-10-13T07:04:49Z
issue,66476,"0INTERNAL ASSERT FAILED, We don't have an op for aten::eq but it isn't a special case",https://github.com/pytorch/pytorch/issues/66476,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2021-10-12T16:35:17Z
issue,66473,nccl comm init needs a global barrier.,https://github.com/pytorch/pytorch/issues/66473,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: nccl",scan when time allows,2021-10-12T16:33:43Z
issue,65301,Anomaly detection: Error detected in CudnnRnnBackward0,https://github.com/pytorch/pytorch/issues/65301,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: cudnn, module: rnn",scan when time allows,2021-10-12T15:52:23Z
issue,66341,Invalid handling of out args with type Tensor[] in native_functions.yaml,https://github.com/pytorch/pytorch/issues/66341,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: codegen, module: internals",scan when time allows,2021-10-12T13:54:58Z
issue,55340,[RFC] Plan to deduplicate test_c10d and distributed_test,https://github.com/pytorch/pytorch/issues/55340,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2021-10-11T23:53:48Z
issue,66359,question on installing GPU-enabled Pytorch,https://github.com/pytorch/pytorch/issues/66359,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: cuda, module: lts",scan when time allows,2021-10-11T21:14:38Z
issue,66196,Specifying USE_VULKAN=0 in launching build_android.sh does not disable Vulkan,https://github.com/pytorch/pytorch/issues/66196,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: vulkan",scan when time allows,2021-10-11T20:13:18Z
issue,66305,Modify convolution kernels for ops,https://github.com/pytorch/pytorch/issues/66305,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fft,scan when time allows,2021-10-11T19:04:50Z
issue,66208,Add LTC-TS support for ops used in SGD optimizer,https://github.com/pytorch/pytorch/issues/66208,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: lazy,scan when time allows,2021-10-11T14:36:59Z
issue,66224,[feature request][quant] Support FakeQuant qconfigs in `test_module_init`,https://github.com/pytorch/pytorch/issues/66224,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tests,scan when time allows,2021-10-11T14:36:53Z
issue,66249,Latency issue with torch.sin,https://github.com/pytorch/pytorch/issues/66249,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: performance,scan when time allows,2021-10-11T14:23:19Z
issue,66364,Hope to obtain authorization for Chinese translation of official tutorials.,https://github.com/pytorch/pytorch/issues/66364,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2021-10-11T14:18:18Z
issue,66251,Should the drop_last parameter of Dataloader be mutually exclusive with the batch samplers?,https://github.com/pytorch/pytorch/issues/66251,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2021-10-11T14:05:18Z
issue,65949,torch.unique signature is not descriptive,https://github.com/pytorch/pytorch/issues/65949,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2021-10-11T01:49:41Z
issue,65038,"[pytorch v1.9.0] for torch sync.pipeline in a single GPU, passing tensor layer by layer is better or using the skip-connection API is better?",https://github.com/pytorch/pytorch/issues/65038,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2021-10-09T06:46:57Z
issue,38471,Immediate mode API (with functional flavor) for optimizers,https://github.com/pytorch/pytorch/issues/38471,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2021-10-09T03:25:18Z
issue,48287,"[doc] Error ""You've reached a dead end"" when opening torch.__config__ in the docs",https://github.com/pytorch/pytorch/issues/48287,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: doc infra, module: docs",scan when time allows,2021-10-08T22:57:58Z
issue,64766,Ability to explicitly close/dispose a DataLoader,https://github.com/pytorch/pytorch/issues/64766,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: data, module: dataloader, module: memory usage",scan when time allows,2021-10-08T02:50:06Z
issue,58139,[docs] Strange torch.unique function signature,https://github.com/pytorch/pytorch/issues/58139,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, module: docs",scan when time allows,2021-10-07T17:07:48Z
issue,65984,libtorch: collate_fn equivalent,https://github.com/pytorch/pytorch/issues/65984,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: dataloader",scan when time allows,2021-10-07T13:57:30Z
issue,64874,I am unable to detect torch.Tensor._version change during __torch_dispatch__ call,https://github.com/pytorch/pytorch/issues/64874,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: __torch_dispatch__,scan when time allows,2021-10-06T15:53:18Z
issue,66199,Move the `queue_callback()` API out of `Variable._execution_engine` and into a public API,https://github.com/pytorch/pytorch/issues/66199,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2021-10-06T14:18:56Z
issue,45769,F.conv2d() causes RuntimeError: cuDNN error: CUDNN_STATUS_INTERNAL_ERROR,https://github.com/pytorch/pytorch/issues/45769,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cudnn,scan when time allows,2021-10-06T14:14:56Z
issue,65811,"Beta distribution yields wrong results for a,b >1 and x=0 or x=1",https://github.com/pytorch/pytorch/issues/65811,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distributions,scan when time allows,2021-10-06T12:15:04Z
issue,65910,Deprecation: Remove nn.functional.sigmoid,https://github.com/pytorch/pytorch/issues/65910,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: deprecation, module: nn",scan when time allows,2021-10-06T04:38:15Z
issue,66026,`__torch_dispatch__` can result in returning `None` for an op that should return Tensors.,https://github.com/pytorch/pytorch/issues/66026,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: __torch_dispatch__, module: pybind",scan when time allows,2021-10-06T03:52:40Z
issue,64186,Complex recurrent layers produce NaN as grad,https://github.com/pytorch/pytorch/issues/64186,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: complex, module: nans and infs, module: nn",scan when time allows,2021-10-05T20:39:26Z
issue,65306,[numpy] Add `iscomplexobj` and `isrealobj`,https://github.com/pytorch/pytorch/issues/65306,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: complex, module: numpy",scan when time allows,2021-10-05T19:05:23Z
issue,66116,Removal of BufferedShuffleDataset,https://github.com/pytorch/pytorch/issues/66116,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: data, module: docs",scan when time allows,2021-10-05T13:05:50Z
issue,65739,Sobol state API,https://github.com/pytorch/pytorch/issues/65739,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: random",scan when time allows,2021-10-05T12:40:28Z
issue,65753,[POC] Convolution-relu fusion via torch function and one-step lazy evaluation,https://github.com/pytorch/pytorch/issues/65753,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: __torch_function__, module: lazy",scan when time allows,2021-10-04T23:10:13Z
issue,65229,at::parallel_for created max_threads for inputs larger than GRAIN_SIZE,https://github.com/pytorch/pytorch/issues/65229,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: multithreading, module: openmp, module: performance",scan when time allows,2021-10-04T17:35:35Z
issue,65426,Add a `__torch_pre_dispatch__` that will live at the top of the dispatcher,https://github.com/pytorch/pytorch/issues/65426,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dispatch,scan when time allows,2021-10-04T17:33:36Z
issue,65694,Plain Tensor serialization does not save the content of `__dict__`.,https://github.com/pytorch/pytorch/issues/65694,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: serialization,scan when time allows,2021-10-04T17:24:59Z
issue,65494,Feature request: add bool dtype support to CPU abs(),https://github.com/pytorch/pytorch/issues/65494,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-10-04T17:23:37Z
issue,65665,Pipe method for torch tensors,https://github.com/pytorch/pytorch/issues/65665,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2021-10-04T17:21:42Z
issue,65688,Parametrization silently disables RNN weight updates,https://github.com/pytorch/pytorch/issues/65688,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: rnn",scan when time allows,2021-10-04T17:19:41Z
issue,66076,Bad Input to torch.nn.NLLLoss causes CUDA Error,https://github.com/pytorch/pytorch/issues/66076,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: loss",scan when time allows,2021-10-04T16:32:45Z
issue,28320,Unable to install pytorch with cuda 10.0 using conda,https://github.com/pytorch/pytorch/issues/28320,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2021-10-04T16:04:13Z
issue,65067,[DataPipe] Add Test Cases to Ensure Correct Behaviors When IterDataPipes Reset,https://github.com/pytorch/pytorch/issues/65067,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: data,scan when time allows,2021-10-04T14:54:36Z
issue,65339,Make it easier to accurately reflect storage/view relationships in meta/wrapper tensors,https://github.com/pytorch/pytorch/issues/65339,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: __torch_dispatch__, module: meta tensors",scan when time allows,2021-10-04T14:36:57Z
issue,65661,[BUG] ConvTranspose with out_channels=1,https://github.com/pytorch/pytorch/issues/65661,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: convolution,scan when time allows,2021-10-03T01:57:58Z
issue,20005,install error from source,https://github.com/pytorch/pytorch/issues/20005,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: internals,scan when time allows,2021-10-02T01:44:04Z
issue,65788,Autocast sometimes discards changes to mutable arguments,https://github.com/pytorch/pytorch/issues/65788,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision), module: intel",scan when time allows,2021-10-01T21:02:11Z
issue,16499,TorchConfig.cmake always sets _GLIBCXX_USE_CXX11_ABI,https://github.com/pytorch/pytorch/issues/16499,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2021-10-01T19:25:20Z
issue,16187,Sparse matrix multiplication is too slow,https://github.com/pytorch/pytorch/issues/16187,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2021-10-01T19:25:20Z
issue,15161,Maybe a bug when using DataParallel,https://github.com/pytorch/pytorch/issues/15161,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: data parallel,scan when time allows,2021-10-01T19:25:20Z
issue,13222,Memory inefficient in batched matmul when requiring gradients,https://github.com/pytorch/pytorch/issues/13222,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: linear algebra,scan when time allows,2021-10-01T19:25:20Z
issue,13018,Improved performance for torch.multinomial with small batches,https://github.com/pytorch/pytorch/issues/13018,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: performance",scan when time allows,2021-10-01T19:25:20Z
issue,10956,Better dev docs for writing native CPU kernels with Vec256,https://github.com/pytorch/pytorch/issues/10956,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: vectorization,scan when time allows,2021-10-01T19:25:20Z
issue,9898,[Feature Request] Checkpoint manager,https://github.com/pytorch/pytorch/issues/9898,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-10-01T19:25:20Z
issue,9433,Build error: cannot find -lonnxifi_loader,https://github.com/pytorch/pytorch/issues/9433,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2021-10-01T19:25:20Z
issue,7066,LBFGS contribution,https://github.com/pytorch/pytorch/issues/7066,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2021-10-01T19:25:20Z
issue,5912,Conv-RNN combination slow in backward pass,https://github.com/pytorch/pytorch/issues/5912,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: performance",scan when time allows,2021-10-01T19:25:20Z
issue,1505,Cannot find Intel MKL,https://github.com/pytorch/pytorch/issues/1505,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: dependency bug, module: mkl",scan when time allows,2021-10-01T19:25:20Z
issue,1462,"Change sparse_mask to take indexing mask, rather than entire sparse tensor",https://github.com/pytorch/pytorch/issues/1462,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2021-10-01T19:25:20Z
issue,1442,Optimizers can't be moved to a different GPU,https://github.com/pytorch/pytorch/issues/1442,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-10-01T19:25:20Z
issue,65760,Codegen issues with Tensor(a!)? arguments in native_functions.yaml schemas,https://github.com/pytorch/pytorch/issues/65760,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: codegen,scan when time allows,2021-10-01T18:12:22Z
issue,65622,Transformer modules should not hard-code the activation function,https://github.com/pytorch/pytorch/issues/65622,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2021-10-01T18:04:47Z
issue,65881,torch.hub.load can confuse external python package with local python package.,https://github.com/pytorch/pytorch/issues/65881,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: hub,scan when time allows,2021-10-01T07:56:39Z
issue,65890,`pad_packed_sequence` is not exactly the inverse of `pack_padded_sequence`,https://github.com/pytorch/pytorch/issues/65890,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rnn,scan when time allows,2021-10-01T01:51:36Z
issue,65868,torch.mul is not consistent with torch.multiply,https://github.com/pytorch/pytorch/issues/65868,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: numerical-stability, module: numpy",scan when time allows,2021-10-01T01:46:24Z
issue,65825,Incorrect reference to previous kernel in the warning displayed when overriding a previously registered kernel for the same operator and the same dispatch,https://github.com/pytorch/pytorch/issues/65825,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dispatch,scan when time allows,2021-09-30T19:05:10Z
issue,65619,Integration of __torch_dispatch__ with masked tensor semantics,https://github.com/pytorch/pytorch/issues/65619,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: __torch_dispatch__,scan when time allows,2021-09-30T17:54:57Z
issue,42053,Function request: np.copy (alias of clone?),https://github.com/pytorch/pytorch/issues/42053,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: numpy,scan when time allows,2021-09-30T08:27:57Z
issue,65522,"pytorch framework tests using make_tensor hangs with pytest's boxed exec option ""--forked""",https://github.com/pytorch/pytorch/issues/65522,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tests,scan when time allows,2021-09-30T05:45:34Z
issue,62648,Feature request: Generate Sobol points directly on the GPU,https://github.com/pytorch/pytorch/issues/62648,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: distributions, module: tensor creation",scan when time allows,2021-09-30T05:39:04Z
issue,57593,[RFC] Provide an API for Structural Performance Tips in DDP,https://github.com/pytorch/pytorch/issues/57593,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ddp, oncall: distributed",scan when time allows,2021-09-28T20:02:12Z
issue,65428,module 'torch.distributed' has no attribute 'optim',https://github.com/pytorch/pytorch/issues/65428,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2021-09-28T19:31:03Z
issue,65752,Deprecate torch.cross with optional dim,https://github.com/pytorch/pytorch/issues/65752,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: deprecation,scan when time allows,2021-09-28T19:04:34Z
issue,55191,"`torch.tensor(..., device='meta')` doesn't work",https://github.com/pytorch/pytorch/issues/55191,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: bootcamp,scan when time allows,2021-09-28T17:03:48Z
issue,61122,Make copy_ use dispatcher,https://github.com/pytorch/pytorch/issues/61122,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: bootcamp,scan when time allows,2021-09-28T10:13:47Z
issue,65683,`torch.nn.functional.l1_loss` fails gradgradcheck for complex inputs,https://github.com/pytorch/pytorch/issues/65683,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: complex, module: nn",scan when time allows,2021-09-27T23:31:39Z
issue,65650,Does ZeroRedundancyOptimizer support re-split partition on load_state_dict?,https://github.com/pytorch/pytorch/issues/65650,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2021-09-27T22:25:39Z
issue,65637,Test ZeRO on gloo backend for GPU,https://github.com/pytorch/pytorch/issues/65637,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ddp, oncall: distributed",scan when time allows,2021-09-27T21:55:33Z
issue,65670,Memory Leak in Distributed RPC nightly,https://github.com/pytorch/pytorch/issues/65670,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2021-09-27T20:25:59Z
issue,65646,[LTC] Building unit tests on Debian results in double defined _GLIBCXX_USE_CXX11_ABI flag,https://github.com/pytorch/pytorch/issues/65646,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: lazy,scan when time allows,2021-09-27T20:21:19Z
issue,65687,"repeat_interleave changes memory format from ""channels last"" to ""contiguous""",https://github.com/pytorch/pytorch/issues/65687,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: memory format,scan when time allows,2021-09-27T18:43:22Z
issue,65673,[libtorch]incorrect sigmoid result on arm chip(rk3326),https://github.com/pytorch/pytorch/issues/65673,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: arm,scan when time allows,2021-09-27T18:32:31Z
issue,61835,Broadcasting documentation diverges from implementation,https://github.com/pytorch/pytorch/issues/61835,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, module: docs",scan when time allows,2021-09-27T17:46:51Z
issue,65664,C10dProcessGroupSerialization.test_process_group_as_module_member is flaky,https://github.com/pytorch/pytorch/issues/65664,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2021-09-27T04:54:19Z
issue,61727,```torch.distributions.Categorical``` unintended ```log_prob``` gradient w.r.t ```probs```,https://github.com/pytorch/pytorch/issues/61727,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distributions,scan when time allows,2021-09-26T09:12:50Z
issue,25066,[FR][jit] torch.jit.script as a class decorator,https://github.com/pytorch/pytorch/issues/25066,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2021-09-25T20:10:44Z
issue,23509,Enable PyTorch Bfloat16 for CPU and add MKL-DNN bfloat16 optimization for Cooper Lake,https://github.com/pytorch/pytorch/issues/23509,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: performance",scan when time allows,2021-09-25T01:10:57Z
issue,65628,Add torch.distributed.shard package,https://github.com/pytorch/pytorch/issues/65628,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2021-09-24T19:57:00Z
issue,65604,pytorch serializes entire tensor when you try to pickle a slice,https://github.com/pytorch/pytorch/issues/65604,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: serialization,scan when time allows,2021-09-24T16:46:16Z
issue,65572,PyTorch setup.py should confirm with user if the user changes build flags,https://github.com/pytorch/pytorch/issues/65572,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2021-09-24T15:59:01Z
issue,65577,Support cat() for meta tensors,https://github.com/pytorch/pytorch/issues/65577,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: meta tensors,scan when time allows,2021-09-24T15:40:21Z
issue,65465,[LTC] Fail to run testcase of latest lazy_tensor_core branch,https://github.com/pytorch/pytorch/issues/65465,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: lazy,scan when time allows,2021-09-24T04:45:55Z
issue,65462,"Why here save `store`, rather than `prefix_store`?",https://github.com/pytorch/pytorch/issues/65462,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2021-09-24T04:19:11Z
issue,65261,Add an `all_paths` method to PackageExporter,https://github.com/pytorch/pytorch/issues/65261,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: package/deploy,scan when time allows,2021-09-24T04:07:50Z
issue,65464,linking error when building on Linux,https://github.com/pytorch/pytorch/issues/65464,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cuda",scan when time allows,2021-09-24T03:42:16Z
issue,65536,Pass backward flags such as retain_graph to context of custom torch.autograd.Function,https://github.com/pytorch/pytorch/issues/65536,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2021-09-23T21:20:21Z
issue,65520,Improve input checking for running_var of nn.functional.batch_norm on CPU/GPU,https://github.com/pytorch/pytorch/issues/65520,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2021-09-23T20:48:42Z
issue,65512,Performance problems of eigh operator on CPU,https://github.com/pytorch/pytorch/issues/65512,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: performance,scan when time allows,2021-09-23T20:32:49Z
issue,65386,[LTC] Setup a subset of TorchBench for the current implementation,https://github.com/pytorch/pytorch/issues/65386,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: lazy,scan when time allows,2021-09-23T20:27:16Z
issue,65569,Support 'meta' device for 'shard_parameter' API,https://github.com/pytorch/pytorch/issues/65569,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2021-09-23T19:56:55Z
issue,18317,Add a RandomBatchSampler ?,https://github.com/pytorch/pytorch/issues/18317,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2021-09-23T18:43:52Z
issue,65475,building pytorch from source without conda,https://github.com/pytorch/pytorch/issues/65475,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2021-09-23T17:32:39Z
issue,36175,New dtype ComplexPolarFloat (phasor),https://github.com/pytorch/pytorch/issues/36175,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: complex,scan when time allows,2021-09-22T16:47:31Z
issue,65329,How can I re-weight a sample based on both class weights and instance weights?,https://github.com/pytorch/pytorch/issues/65329,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: loss,scan when time allows,2021-09-22T16:31:49Z
issue,65446,libtorch compile problem. How to get the correct protobuf version? what PROTOBUF_VERSION <3011000 and 3011004 <PROTOBUF_MIN_PROTOC_VERSION?,https://github.com/pytorch/pytorch/issues/65446,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: protobuf",scan when time allows,2021-09-22T14:27:18Z
issue,65447,No support for torch.trace on CPU for float16 tensors,https://github.com/pytorch/pytorch/issues/65447,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: half,scan when time allows,2021-09-22T14:16:52Z
issue,35736,"Simple C++ custom autograd function code throws error ""CUDA error: driver shutting down""",https://github.com/pytorch/pytorch/issues/35736,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: cpp",scan when time allows,2021-09-22T13:58:51Z
issue,65448,Magma : Intel MKL Errors,https://github.com/pytorch/pytorch/issues/65448,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: intel, module: magma, module: mkldnn",scan when time allows,2021-09-22T13:27:18Z
issue,59437,pytorch/manylinux-cuda102 support for aarch64,https://github.com/pytorch/pytorch/issues/59437,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: arm, module: build, module: docker",scan when time allows,2021-09-22T11:06:04Z
issue,65436,Make DDP uneven inputs work with custom buffer reduction,https://github.com/pytorch/pytorch/issues/65436,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ddp, oncall: distributed",scan when time allows,2021-09-22T07:45:26Z
issue,30807,Sending sparse tensors over RPC not yet supported,https://github.com/pytorch/pytorch/issues/30807,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: rpc, module: serialization",scan when time allows,2021-09-22T02:56:49Z
issue,10377,[Docs] Update broadcasting documentation for scalars / n-dimensional empty tensors,https://github.com/pytorch/pytorch/issues/10377,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2021-09-22T02:07:47Z
issue,65375,Make python_error use c10/util/Exception.h C++ backtrace machinery,https://github.com/pytorch/pytorch/issues/65375,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: error checking,scan when time allows,2021-09-21T22:59:19Z
issue,64931,native functions should not be allowed to take in a `grad` argument,https://github.com/pytorch/pytorch/issues/64931,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: codegen",scan when time allows,2021-09-21T15:49:00Z
issue,65400,torch.sparse.sum on scalar sparse tensor fails when dim is specified,https://github.com/pytorch/pytorch/issues/65400,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2021-09-21T15:01:59Z
issue,53365,Pytorch mobile Vulkan API usage,https://github.com/pytorch/pytorch/issues/53365,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: mobile,scan when time allows,2021-09-21T10:30:52Z
issue,63812,Calling backward with create_graph on the output of a DistributedDataParallel throws error,https://github.com/pytorch/pytorch/issues/63812,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ddp, oncall: distributed",scan when time allows,2021-09-20T17:54:25Z
issue,23938,torch.nn.DataParallel causes incorrect gradients,https://github.com/pytorch/pytorch/issues/23938,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, oncall: distributed",scan when time allows,2021-09-20T15:22:00Z
issue,65215,_disabled_torch_function_impl should not be necessary for __torch_dispatch__,https://github.com/pytorch/pytorch/issues/65215,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: __torch_function__,scan when time allows,2021-09-20T13:46:28Z
issue,65062,Gloo and TensorPipe depend on different version of libuv,https://github.com/pytorch/pytorch/issues/65062,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: tensorpipe, module: third_party, oncall: distributed",scan when time allows,2021-09-20T11:28:59Z
issue,62876,Key already registered with the same priority: uv,https://github.com/pytorch/pytorch/issues/62876,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: internals, module: tensorpipe",scan when time allows,2021-09-20T10:51:11Z
issue,39437,error when specifying sparse=True in embedding,https://github.com/pytorch/pytorch/issues/39437,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2021-09-20T07:41:21Z
issue,44279,How to use CUDA Dynamic Parallelism in PyTorch CPP extension?,https://github.com/pytorch/pytorch/issues/44279,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2021-09-18T11:30:15Z
issue,65163,Parametrization cannot be parametrized,https://github.com/pytorch/pytorch/issues/65163,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2021-09-17T18:11:37Z
issue,50802,Link to `torch.einsum` in `torch.tensordot`,https://github.com/pytorch/pytorch/issues/50802,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2021-09-17T17:06:25Z
issue,53704,`test_reductions` ignoring some tests,https://github.com/pytorch/pytorch/issues/53704,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: reductions, module: tests",scan when time allows,2021-09-17T17:06:24Z
issue,62354,Some way to specify expected failures for OpInfo-based tests,https://github.com/pytorch/pytorch/issues/62354,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tests,scan when time allows,2021-09-17T17:06:20Z
issue,63645,Create a new OptionalArrayRef template class to replace Optional<ArrayRef>,https://github.com/pytorch/pytorch/issues/63645,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: internals,scan when time allows,2021-09-17T17:06:16Z
issue,64690,_disabled_torch_function_impl is unimplementable (must be special cased),https://github.com/pytorch/pytorch/issues/64690,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: __torch_function__,scan when time allows,2021-09-17T16:33:03Z
issue,65049,Setting a random seed or passing a generator to Modules for deterministic sampling.,https://github.com/pytorch/pytorch/issues/65049,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: determinism, module: nn, module: random",scan when time allows,2021-09-17T15:43:05Z
issue,65100,[Bug] [CUDA IPC] CUDA IPC memory cost,https://github.com/pytorch/pytorch/issues/65100,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: multiprocessing, module: performance",scan when time allows,2021-09-17T12:17:15Z
issue,65132,How to reference a tensor variable from a superclass of `torch.Tensor`?,https://github.com/pytorch/pytorch/issues/65132,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: xla,scan when time allows,2021-09-16T21:35:37Z
issue,65131,Adding a device variable to a `torch.Tensor` subclass fails,https://github.com/pytorch/pytorch/issues/65131,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: xla,scan when time allows,2021-09-16T21:31:39Z
issue,65166,Review the Tensor and Operator Basics Onboarding page,https://github.com/pytorch/pytorch/issues/65166,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2021-09-16T20:52:59Z
issue,39671,TestTorchDeviceTypeCPU.test_float_to_int_conversion_finite_cpu_uint8 is broken on PowerPC,https://github.com/pytorch/pytorch/issues/39671,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: power, module: tests",scan when time allows,2021-09-16T09:34:25Z
issue,18434,improve jit error message for legacy constructor,https://github.com/pytorch/pytorch/issues/18434,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2021-09-16T01:04:32Z
issue,50962,Add option for Convolutions to operate over NTC and NHWC tensors,https://github.com/pytorch/pytorch/issues/50962,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: memory format,scan when time allows,2021-09-15T22:21:28Z
issue,65069,torch.linalg.cholesky does not raise a RuntimeError when the matrix is not symmetric,https://github.com/pytorch/pytorch/issues/65069,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: linear algebra",scan when time allows,2021-09-15T21:43:08Z
issue,65050,torch.cross precision problem,https://github.com/pytorch/pytorch/issues/65050,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: numerical-stability,scan when time allows,2021-09-15T18:14:48Z
issue,64743,Named Tensors to support custom unify functions,https://github.com/pytorch/pytorch/issues/64743,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: named tensor,scan when time allows,2021-09-15T12:24:43Z
issue,65002,Traceback tensor mode,https://github.com/pytorch/pytorch/issues/65002,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-09-15T00:01:19Z
issue,16589,Make sure `data_ptr` for non-zero-size input tensors stays the same after the VariableType dispatch,https://github.com/pytorch/pytorch/issues/16589,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: assert failure, module: autograd, module: molly-guard",scan when time allows,2021-09-14T23:05:01Z
issue,55192,"Consider adding meta:1, meta:2 devices",https://github.com/pytorch/pytorch/issues/55192,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: meta tensors, module: structured kernels",scan when time allows,2021-09-14T20:46:06Z
issue,19978,[JIT] traced model with optimization shows no performance improvement,https://github.com/pytorch/pytorch/issues/19978,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2021-09-14T17:18:26Z
issue,18157,JIT does not batch linear layers in an ensemble,https://github.com/pytorch/pytorch/issues/18157,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2021-09-14T17:17:19Z
issue,17655,TracedModule 'to' attribute doesn't work for tensors created on forward.,https://github.com/pytorch/pytorch/issues/17655,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2021-09-14T17:17:08Z
issue,40507,SyncBatchNorm for JIT and a list of not supported operations,https://github.com/pytorch/pytorch/issues/40507,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2021-09-14T17:16:12Z
issue,30754,"JIT, nn.utils.weight_norm and {save,load}_state_dict produce wrong results",https://github.com/pytorch/pytorch/issues/30754,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2021-09-14T17:15:10Z
issue,22155,[jit] Optional type refinement on non-named expressions,https://github.com/pytorch/pytorch/issues/22155,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2021-09-14T17:13:36Z
issue,64939,BC CI error message should link to some information about how to squash the warning,https://github.com/pytorch/pytorch/issues/64939,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2021-09-13T17:38:17Z
issue,58380,Clarity of error message in einsum regressed in performance improvements,https://github.com/pytorch/pytorch/issues/58380,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: error checking, module: linear algebra",scan when time allows,2021-09-13T16:08:36Z
issue,41099,max_pool2d always compute indices even when it's not required,https://github.com/pytorch/pytorch/issues/41099,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: pooling",scan when time allows,2021-09-13T16:07:14Z
issue,40570,Generalized CPU vector reductions,https://github.com/pytorch/pytorch/issues/40570,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: performance, module: reductions",scan when time allows,2021-09-13T16:07:05Z
issue,64881,Add support for GPU based data loading and transformation,https://github.com/pytorch/pytorch/issues/64881,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2021-09-13T14:55:41Z
issue,17966,torch.cuda.is_available() returns misleading value,https://github.com/pytorch/pytorch/issues/17966,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: ux",scan when time allows,2021-09-13T14:02:16Z
issue,64812,PyTorch inference on tensors of a particular size cause Illegal Instruction (core dumped) on Jetson Nano,https://github.com/pytorch/pytorch/issues/64812,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: arm, module: jetson",scan when time allows,2021-09-13T08:16:28Z
issue,23067,creation of a tensor from a numba.cuda array,https://github.com/pytorch/pytorch/issues/23067,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: numba,scan when time allows,2021-09-13T05:07:58Z
issue,64818,Inconsistent NaN handling by cholesky between CPU and CUDA,https://github.com/pytorch/pytorch/issues/64818,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: linear algebra, module: nans and infs",scan when time allows,2021-09-11T20:41:37Z
issue,30873,ReduceLROnPlateau detects a plateau during a steady decrease after a spike,https://github.com/pytorch/pytorch/issues/30873,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2021-09-11T07:45:12Z
issue,64845,Track Reliability of Test Reporting and Uploading,https://github.com/pytorch/pytorch/issues/64845,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-09-11T01:30:57Z
issue,17849,Tensor::options() returns false for requires_grad when it is true,https://github.com/pytorch/pytorch/issues/17849,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp-extensions,scan when time allows,2021-09-11T01:25:33Z
issue,17879,C++ nn::Sequential push_back() copies module if the module is concrete type,https://github.com/pytorch/pytorch/issues/17879,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: nn",scan when time allows,2021-09-11T01:23:35Z
issue,64868,CMake outputs lots of warnings in CI,https://github.com/pytorch/pytorch/issues/64868,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2021-09-11T00:14:47Z
issue,64311,issues flutter in android studio,https://github.com/pytorch/pytorch/issues/64311,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: android, oncall: mobile",scan when time allows,2021-09-10T20:09:10Z
issue,64407,Create a boxed/templated ADInplaceOrView kernel,https://github.com/pytorch/pytorch/issues/64407,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2021-09-09T19:39:01Z
issue,63041,[RFC] Should DDP support custom reduction logic for registered module buffers?,https://github.com/pytorch/pytorch/issues/63041,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ddp, oncall: distributed",scan when time allows,2021-09-09T19:02:05Z
issue,64745,Consolidate ProcessGroup allgather_coalesced and allgather,https://github.com/pytorch/pytorch/issues/64745,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, module: c10d, oncall: distributed",scan when time allows,2021-09-09T14:57:50Z
issue,64637,Evaluating on single GPU (DDP),https://github.com/pytorch/pytorch/issues/64637,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ddp,scan when time allows,2021-09-09T00:42:52Z
issue,64644,Improve meta tensor testing,https://github.com/pytorch/pytorch/issues/64644,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: structured kernels, module: tests",scan when time allows,2021-09-09T00:41:57Z
issue,64709,Add out= resize warning to cat and maybe_native_stack,https://github.com/pytorch/pytorch/issues/64709,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: viewing and reshaping,scan when time allows,2021-09-08T22:59:05Z
issue,17150,Conv2d layers should accept 4-tuple for padding argument,https://github.com/pytorch/pytorch/issues/17150,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: nn",scan when time allows,2021-09-08T18:50:53Z
issue,64544,[Feature Request] Any plan to add 'Sparse Convolution' as default nn module?,https://github.com/pytorch/pytorch/issues/64544,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: sparse",scan when time allows,2021-09-08T14:41:52Z
issue,61468,Version is set to 1.8.0 in the torch tag v1.8.1,https://github.com/pytorch/pytorch/issues/61468,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, oncall: releng",scan when time allows,2021-09-08T02:20:09Z
issue,64539,[LTC][Documentation] Python examples in API_GUIDE.md don't work,https://github.com/pytorch/pytorch/issues/64539,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: lazy",scan when time allows,2021-09-07T22:42:54Z
issue,64527,"Asking for ""git submodule update"" when building a specific version",https://github.com/pytorch/pytorch/issues/64527,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2021-09-07T20:36:26Z
issue,64548,Interpreter initialized state changed by torch dispatch,https://github.com/pytorch/pytorch/issues/64548,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: __torch_dispatch__, module: deploy",scan when time allows,2021-09-07T20:32:43Z
issue,64541,[docs] Version switcher to not reset home page unconditionally,https://github.com/pytorch/pytorch/issues/64541,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: doc infra, module: docs",scan when time allows,2021-09-07T20:28:39Z
issue,64458,Rename C++ Serialization APIs,https://github.com/pytorch/pytorch/issues/64458,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: serialization,scan when time allows,2021-09-07T03:07:18Z
issue,64292,torch.empty_like not taking None as layout?,https://github.com/pytorch/pytorch/issues/64292,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: tensor creation",scan when time allows,2021-09-06T05:57:59Z
issue,64107,as_tensor and negative strided np arrays,https://github.com/pytorch/pytorch/issues/64107,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: numpy, module: tensor creation",scan when time allows,2021-09-06T05:30:22Z
issue,64106,modulefinder_determinator is incompatible with imports into the test/ directory,https://github.com/pytorch/pytorch/issues/64106,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tests,scan when time allows,2021-09-05T03:30:55Z
issue,17612,The documentation to Module._version isn't visible.,https://github.com/pytorch/pytorch/issues/17612,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: nn, module: serialization",scan when time allows,2021-09-05T03:13:31Z
issue,26456,[libtorch]Same model in CUDA and CPU got different result?,https://github.com/pytorch/pytorch/issues/26456,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: nn",scan when time allows,2021-09-03T01:58:27Z
issue,64426,Add support for ROCm MIOpen miopenConvolutionForwardBias and miopenConvolutionBackwardBias,https://github.com/pytorch/pytorch/issues/64426,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rocm,scan when time allows,2021-09-02T16:57:49Z
issue,17644,"Not depend on third_party submodules, but self-built libraries?",https://github.com/pytorch/pytorch/issues/17644,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2021-09-02T15:04:49Z
issue,62032,"Create a boxed fallback / template recipe for Autograd that forwards, but errors on backwards",https://github.com/pytorch/pytorch/issues/62032,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: dispatch, module: internals",scan when time allows,2021-09-02T04:30:02Z
issue,14112,[sparse] add descriptions and examples for methods at torch.sparse doc page,https://github.com/pytorch/pytorch/issues/14112,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2021-09-01T22:33:10Z
issue,64308,Building from source results in no member named 'cerr' in namespace 'std',https://github.com/pytorch/pytorch/issues/64308,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2021-09-01T20:22:56Z
issue,64384,[doc] incomplete `get_num_interop_threads` doc,https://github.com/pytorch/pytorch/issues/64384,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2021-09-01T19:01:31Z
issue,64306,GradGrad of max_pool2d fails with empty batch dimension,https://github.com/pytorch/pytorch/issues/64306,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: nn",scan when time allows,2021-09-01T17:54:28Z
issue,63297,Named tensor in tracer,https://github.com/pytorch/pytorch/issues/63297,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: named tensor,scan when time allows,2021-09-01T14:16:46Z
issue,51558,torch.reshape fails to keep the memory format,https://github.com/pytorch/pytorch/issues/51558,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: memory format, module: viewing and reshaping",scan when time allows,2021-09-01T12:20:41Z
issue,48483,Vulkan Api Backend Build Error with GCC,https://github.com/pytorch/pytorch/issues/48483,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2021-09-01T07:22:31Z
issue,63870,torch median / nanmedian w/ nans speed,https://github.com/pytorch/pytorch/issues/63870,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nans and infs, module: performance, module: reductions, module: sorting and selection",scan when time allows,2021-09-01T00:06:39Z
issue,62788,PyTorch(with mkldnn) in inference mode using non-optimal prop_kind for convolution,https://github.com/pytorch/pytorch/issues/62788,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: mkldnn,scan when time allows,2021-08-31T19:44:23Z
issue,62473,Sparse updates to logits in distributions.Categorical,https://github.com/pytorch/pytorch/issues/62473,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: distributions, module: memory usage, module: sparse",scan when time allows,2021-08-31T17:40:30Z
issue,64284,FR: Record results of OpInfo reference tests and detect when numerics of an operator change,https://github.com/pytorch/pytorch/issues/64284,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tests,scan when time allows,2021-08-31T17:00:36Z
issue,64206,Document how to generate Pybind bindings for C++ Autograd,https://github.com/pytorch/pytorch/issues/64206,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp,scan when time allows,2021-08-31T14:20:23Z
issue,38185,Feature Request: Sampled softmax loss,https://github.com/pytorch/pytorch/issues/38185,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: loss, module: nn",scan when time allows,2021-08-31T07:19:46Z
issue,38205,RESOLVED: Disable zero-dim CUDA tensors interacting with CUDA tensors on other devices,https://github.com/pytorch/pytorch/issues/38205,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bc-breaking, module: cuda",scan when time allows,2021-08-31T07:15:20Z
issue,64138,Longer int and normal Floats conversion to FloatTensor is strange,https://github.com/pytorch/pytorch/issues/64138,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2021-08-31T06:45:25Z
issue,57447,`torch._dirichlet_grad` returns `nan` value on cuda device,https://github.com/pytorch/pytorch/issues/57447,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: cuda, module: nans and infs",scan when time allows,2021-08-31T00:47:52Z
issue,51906,torch.nn.functional.binary_cross_entropy(_with_logits) outputs NaN,https://github.com/pytorch/pytorch/issues/51906,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nans and infs, module: nn",scan when time allows,2021-08-30T21:14:39Z
issue,51714,Inconsistent result between NumPy and PyTorch for 0^(z) when Re(z) < 0,https://github.com/pytorch/pytorch/issues/51714,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: complex, module: nans and infs",scan when time allows,2021-08-30T21:13:57Z
issue,48000,Inconsistent complex results with NumPy when computing non-positive power of 0,https://github.com/pytorch/pytorch/issues/48000,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: complex, module: nans and infs, module: numpy",scan when time allows,2021-08-30T21:06:40Z
issue,60825,Where OpInfo doesn't handle cases where one of the inputs is a scalar,https://github.com/pytorch/pytorch/issues/60825,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: sorting and selection, module: tests",scan when time allows,2021-08-30T20:34:56Z
issue,64133,nn.MultiheadAttention does not belong to activation function,https://github.com/pytorch/pytorch/issues/64133,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: nn",scan when time allows,2021-08-30T17:58:34Z
issue,64038,torch.nn.functional.softmax _stacklevel undocumented,https://github.com/pytorch/pytorch/issues/64038,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: nn",scan when time allows,2021-08-30T17:32:31Z
issue,64187,Uniformly test that defaulted Tensor arguments appropriately handle __torch_function__,https://github.com/pytorch/pytorch/issues/64187,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: __torch_function__, module: tests",scan when time allows,2021-08-30T17:29:18Z
issue,63970,[LTC] Investigate a testing utility to make sure correct lowering has happened,https://github.com/pytorch/pytorch/issues/63970,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: lazy,scan when time allows,2021-08-30T17:00:16Z
issue,63113,Implement half_to_float flag for both _log_softmax and _softmax operators,https://github.com/pytorch/pytorch/issues/63113,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: lazy,scan when time allows,2021-08-30T17:00:15Z
issue,62614,LazyTensor recomputes intermediate tensor in fragmented computation,https://github.com/pytorch/pytorch/issues/62614,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: lazy,scan when time allows,2021-08-30T17:00:13Z
issue,64162,[feature request] torch.Generator constructor to accept seed directly,https://github.com/pytorch/pytorch/issues/64162,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: random,scan when time allows,2021-08-30T15:58:23Z
issue,63920,quicklint clang-tidy step fails with many unrelated errors,https://github.com/pytorch/pytorch/issues/63920,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: lint,scan when time allows,2021-08-30T15:48:24Z
issue,63919,quicklint mypy step fails with assertion,https://github.com/pytorch/pytorch/issues/63919,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: lint, module: typing",scan when time allows,2021-08-30T15:48:01Z
issue,17554,Add ASSERT for calling accessor on a GPU tensor,https://github.com/pytorch/pytorch/issues/17554,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: molly-guard",scan when time allows,2021-08-30T15:39:31Z
issue,57272,Consistent treatment of non-differentiability in linear algebra operations,https://github.com/pytorch/pytorch/issues/57272,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: linear algebra",scan when time allows,2021-08-30T15:25:42Z
issue,17425,"improved assert message in the case of ""CUDA error: device-side assert triggered""",https://github.com/pytorch/pytorch/issues/17425,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, module: cuda, module: molly-guard",scan when time allows,2021-08-30T15:23:18Z
issue,63293,Support `torch.linalg.outer`,https://github.com/pytorch/pytorch/issues/63293,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: linear algebra, module: python array api",scan when time allows,2021-08-30T12:53:29Z
issue,56235,`test_cholesky_solve` gradgradcheck fails sometimes,https://github.com/pytorch/pytorch/issues/56235,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: linear algebra, module: magma, module: tests",scan when time allows,2021-08-30T08:29:01Z
issue,58015,Support a dist.group_like API,https://github.com/pytorch/pytorch/issues/58015,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, module: c10d, oncall: distributed",scan when time allows,2021-08-30T07:13:01Z
issue,8577,Update tests to no longer spew debug info,https://github.com/pytorch/pytorch/issues/8577,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tests,scan when time allows,2021-08-30T03:11:59Z
issue,11496,"Tests with ""."" in the name cannot be run standalone",https://github.com/pytorch/pytorch/issues/11496,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tests,scan when time allows,2021-08-30T03:09:43Z
issue,52738,[OpInfo] Improvements for sparse ops tests,https://github.com/pytorch/pytorch/issues/52738,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: sparse, module: tests",scan when time allows,2021-08-30T03:08:01Z
issue,52984,Tracker: pytest-related test improvements,https://github.com/pytorch/pytorch/issues/52984,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tests,scan when time allows,2021-08-30T03:07:16Z
issue,53352,OpInfo mechanism to test for nondeterminism,https://github.com/pytorch/pytorch/issues/53352,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: determinism, module: tests",scan when time allows,2021-08-30T03:06:57Z
issue,53510,Improve memory format testing,https://github.com/pytorch/pytorch/issues/53510,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: memory format, module: tests",scan when time allows,2021-08-30T03:05:03Z
issue,54006,enable parallel test execution for GPU tests,https://github.com/pytorch/pytorch/issues/54006,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ci, module: tests",scan when time allows,2021-08-30T03:03:28Z
issue,55159,Better syntax for OpInfo,https://github.com/pytorch/pytorch/issues/55159,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tests,scan when time allows,2021-08-30T03:03:08Z
issue,54346,Missing tests for gradcheck,https://github.com/pytorch/pytorch/issues/54346,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: tests",scan when time allows,2021-08-30T03:03:00Z
issue,55907,test_variant_consistency_eager_addbmm fails on both cpu and cuda,https://github.com/pytorch/pytorch/issues/55907,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: linear algebra, module: tests",scan when time allows,2021-08-30T03:00:23Z
issue,56202,Tolerance for non-determinism operators in gradcheck,https://github.com/pytorch/pytorch/issues/56202,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: determinism, module: tests",scan when time allows,2021-08-30T02:59:46Z
issue,57301,Avoid code repeat in create sample inputs for sort/msort,https://github.com/pytorch/pytorch/issues/57301,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: sorting and selection, module: tests",scan when time allows,2021-08-30T02:57:16Z
issue,56275,cumprod gradgradcheck fails in fast_mode=True,https://github.com/pytorch/pytorch/issues/56275,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: tests",scan when time allows,2021-08-30T02:56:35Z
issue,60748,"Record file/line number when creating test data, and then report it in backtraces associated with this data",https://github.com/pytorch/pytorch/issues/60748,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tests,scan when time allows,2021-08-30T02:52:43Z
issue,60916,Scipy 1.7.0 may cause some test failures,https://github.com/pytorch/pytorch/issues/60916,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: distributions, module: tests",scan when time allows,2021-08-30T02:50:40Z
issue,62925,[Testing] memory_format decorator,https://github.com/pytorch/pytorch/issues/62925,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: memory format, module: tests",scan when time allows,2021-08-30T02:45:34Z
issue,63837,`torch.scatter` doesn't fail correctly on CUDA (memory overlap),https://github.com/pytorch/pytorch/issues/63837,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: scatter & gather ops, module: structured kernels, module: tests",scan when time allows,2021-08-30T02:39:38Z
issue,49909,Test functionality to detect extra cross device synchronizations,https://github.com/pytorch/pytorch/issues/49909,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: tests",scan when time allows,2021-08-29T23:01:55Z
issue,64079,torch.equal does not support sparse tensors,https://github.com/pytorch/pytorch/issues/64079,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2021-08-27T05:16:46Z
issue,63555,pytorch 1.8.1+cu111 used much more CPU RAM than pytorch 1.8.1 after run `import torch`,https://github.com/pytorch/pytorch/issues/63555,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: cpu, module: cuda, module: memory usage",scan when time allows,2021-08-27T03:28:17Z
issue,61890,The matrix multiplication operator can't get correct results on 3090 !!,https://github.com/pytorch/pytorch/issues/61890,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: tf32",scan when time allows,2021-08-26T20:40:53Z
issue,52052,common_jit.check_against_reference() should check gradients w.r.t. weights,https://github.com/pytorch/pytorch/issues/52052,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, oncall: jit",scan when time allows,2021-08-26T16:09:57Z
issue,58720,`pickle_save`/`pickle_load` are undocumented,https://github.com/pytorch/pytorch/issues/58720,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2021-08-26T16:09:07Z
issue,54394,Half Normal Log_Prob not defined for 0,https://github.com/pytorch/pytorch/issues/54394,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: distributions, module: half, module: numerical-stability, module: random",scan when time allows,2021-08-26T14:55:50Z
issue,63781,[rfc] Hardcoded Target Determination,https://github.com/pytorch/pytorch/issues/63781,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tests,scan when time allows,2021-08-26T06:01:09Z
issue,63998,cat out= variant memory_format inconsistent for cpu and cuda,https://github.com/pytorch/pytorch/issues/63998,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: memory format",scan when time allows,2021-08-26T03:47:46Z
issue,16458,TensorRTOpTest.test_vgg19 is flaky,https://github.com/pytorch/pytorch/issues/16458,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: flaky-tests,scan when time allows,2021-08-25T23:50:41Z
issue,62849,"slice_embed, select_embed like diag_embed",https://github.com/pytorch/pytorch/issues/62849,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: viewing and reshaping",scan when time allows,2021-08-25T17:53:13Z
issue,63720,Torch.save with zip serialization causes memory bloat,https://github.com/pytorch/pytorch/issues/63720,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: serialization,scan when time allows,2021-08-25T16:43:54Z
issue,63475,C++ Inference error causes Identity Module,https://github.com/pytorch/pytorch/issues/63475,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2021-08-25T08:50:20Z
issue,54743,[Feature request] Add batched matrix support for torch.diag,https://github.com/pytorch/pytorch/issues/54743,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: batching, module: linear algebra, module: numpy",scan when time allows,2021-08-25T05:55:19Z
issue,31571,Bug in ForkingPickler for multiprocessing spawn context for shared storages on Linux,https://github.com/pytorch/pytorch/issues/31571,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: multiprocessing,scan when time allows,2021-08-25T04:50:42Z
issue,63897,Scalar construction in C++: PY`torch.tensor(7)` !~= C++`at::tensor(7)`,https://github.com/pytorch/pytorch/issues/63897,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: tensor creation",scan when time allows,2021-08-24T23:27:37Z
issue,17234,[feature request] build and move distributions w/ device and/or dtype,https://github.com/pytorch/pytorch/issues/17234,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: distributions",scan when time allows,2021-08-24T19:54:31Z
issue,63849,"Installed successfully the torch-1.9 and torch_xla-1.9. RuntimeError requested XLA_GPU:0，but available devices are [CPU:0, XLA_CPU:0 ]",https://github.com/pytorch/pytorch/issues/63849,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: xla",scan when time allows,2021-08-24T17:37:15Z
issue,17257,Generated `__init__.pyi` contains invalid default values,https://github.com/pytorch/pytorch/issues/17257,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: codegen,scan when time allows,2021-08-24T15:36:00Z
issue,17268,Generic object to tensor dispatching,https://github.com/pytorch/pytorch/issues/17268,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-08-24T15:34:59Z
issue,17286,Use standard docker image for XLA build,https://github.com/pytorch/pytorch/issues/17286,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docker, module: xla, oncall: releng",scan when time allows,2021-08-24T15:33:48Z
issue,20117,[FR] [RFC] add Sequential.append & .extend,https://github.com/pytorch/pytorch/issues/20117,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2021-08-24T13:43:30Z
issue,62588,static library kineto_LIBRARY-NOTFOUND not found.,https://github.com/pytorch/pytorch/issues/62588,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build warnings,scan when time allows,2021-08-24T02:11:19Z
issue,7343,[memory leak] [PyTorch] .backward(create_graph=True),https://github.com/pytorch/pytorch/issues/7343,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: memory usage",scan when time allows,2021-08-23T23:15:42Z
issue,62162,Clarify sparse COO tensor coalesce behavior wrt overflow + how to binarize a sparse tensor,https://github.com/pytorch/pytorch/issues/62162,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: sparse",scan when time allows,2021-08-23T14:53:45Z
issue,63644,LeakyReLU and Elu use more VRAM than needed,https://github.com/pytorch/pytorch/issues/63644,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: nn",scan when time allows,2021-08-23T14:14:14Z
issue,59263,Ubuntu 20.04 GeForce RTX 3070 / NVIDIA-SMI 460.73.01 Driver Version: 460.73.01 / GPU Visibility Issue,https://github.com/pytorch/pytorch/issues/59263,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: cuda",scan when time allows,2021-08-22T05:38:28Z
issue,61182,"Feature Request: torch.functional.interpolate to quietly ignore ""align_corners"" when mode is set to ""nearest""",https://github.com/pytorch/pytorch/issues/61182,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: interpolation, module: nn",scan when time allows,2021-08-21T23:52:14Z
issue,62352,[feature request] Expand a given dim,https://github.com/pytorch/pytorch/issues/62352,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: viewing and reshaping,scan when time allows,2021-08-21T16:21:18Z
issue,63618,C++ torch::cuda::synchronize speeds up training,https://github.com/pytorch/pytorch/issues/63618,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2021-08-21T13:29:31Z
issue,40427,"Can we make torch.inverse FP16? - RuntimeError: ""inverse_cuda"" not implemented for 'Half'",https://github.com/pytorch/pytorch/issues/40427,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision), module: cuda, module: linear algebra",scan when time allows,2021-08-20T16:18:47Z
issue,63624,Raw saved tensors can survive the deletion of the underlying SavedVariable object,https://github.com/pytorch/pytorch/issues/63624,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2021-08-20T14:12:30Z
issue,63559,"Functional grid_sample: example of padding_mode=""reflection"" description error",https://github.com/pytorch/pytorch/issues/63559,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: interpolation, module: nn",scan when time allows,2021-08-19T21:52:19Z
issue,63574,compile_commands.json doesn't contain nvcc invocations,https://github.com/pytorch/pytorch/issues/63574,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2021-08-19T21:42:37Z
issue,63485,Preserve tensor subclasses when unpacking a SavedTensor,https://github.com/pytorch/pytorch/issues/63485,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2021-08-19T14:54:29Z
issue,18106,copy.copy not working for ScriptModule,https://github.com/pytorch/pytorch/issues/18106,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2021-08-18T23:39:42Z
issue,59526,Documentation for torch.finfo doesn't match implementation,https://github.com/pytorch/pytorch/issues/59526,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: numpy",scan when time allows,2021-08-18T21:06:57Z
issue,38642,Synchronization problem in torch.distributed with MPI on CUDA,https://github.com/pytorch/pytorch/issues/38642,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2021-08-18T16:00:07Z
issue,63477,Supporting Pytorch for Custom Compiler Backend,https://github.com/pytorch/pytorch/issues/63477,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: backend, module: dispatch",scan when time allows,2021-08-18T13:30:33Z
issue,63441,OpenCV causes backpropagation to get stuck,https://github.com/pytorch/pytorch/issues/63441,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: deadlock, module: multithreading",scan when time allows,2021-08-18T04:00:31Z
issue,57610,[perf] 10x improvement when doing `x.sum(-1)` manually,https://github.com/pytorch/pytorch/issues/57610,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: performance, module: reductions",scan when time allows,2021-08-17T22:00:08Z
issue,63290,DistributedDataParallel documentation,https://github.com/pytorch/pytorch/issues/63290,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ddp, oncall: distributed",scan when time allows,2021-08-17T19:24:45Z
issue,16896,`nn.Linear` allows 1d input tensors,https://github.com/pytorch/pytorch/issues/16896,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2021-08-17T17:59:49Z
issue,62955,Make axes selection keyword arguments in torch.diagonal and torch.transpose consistent,https://github.com/pytorch/pytorch/issues/62955,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: linear algebra, module: numpy, module: python array api",scan when time allows,2021-08-17T08:32:19Z
issue,62265,`periodic-pytorch-linux-xenial-cuda11.3-cudnn8-py3.6-gcc7-build` started to fail,https://github.com/pytorch/pytorch/issues/62265,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2021-08-16T22:08:31Z
issue,63226,"cuda.is_available returns True in pycharms python console, False in code",https://github.com/pytorch/pytorch/issues/63226,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2021-08-16T21:56:19Z
issue,63217,Torch.nn.functional.interpolate() function memory release problem,https://github.com/pytorch/pytorch/issues/63217,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: interpolation, module: memory usage",scan when time allows,2021-08-16T21:54:46Z
issue,63313,Automatic mixed precision works worse for 3D neural networks.,https://github.com/pytorch/pytorch/issues/63313,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision), module: performance",scan when time allows,2021-08-16T21:54:25Z
issue,63295,pytorch 1.8.2 cuda test errors,https://github.com/pytorch/pytorch/issues/63295,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: build, module: cuda, module: linear algebra",scan when time allows,2021-08-16T21:53:39Z
issue,63356,[python codegen] correctly plumb TensorOptions defaults through the python binding layer.,https://github.com/pytorch/pytorch/issues/63356,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: codegen,scan when time allows,2021-08-16T20:52:02Z
issue,63299,Channels last mode doesn't work with inputs with 0 batch dimension,https://github.com/pytorch/pytorch/issues/63299,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: memory format, module: nn",scan when time allows,2021-08-16T17:39:32Z
issue,63023,[feature request] Type promotions for Boolean tensors with sub operation + Numpy compatability,https://github.com/pytorch/pytorch/issues/63023,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: boolean tensor, module: numpy, module: type promotion",scan when time allows,2021-08-16T17:17:56Z
issue,63282,Serialising `torch.bool` generates a warning about `np.bool` being deprecated.,https://github.com/pytorch/pytorch/issues/63282,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: error checking, module: numpy, module: serialization",scan when time allows,2021-08-16T01:13:05Z
issue,63257,`help(torch.finfo)` doesn't give useful information even though class has documentation,https://github.com/pytorch/pytorch/issues/63257,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2021-08-15T16:19:45Z
issue,63211,Memory leak in `ddp_zero_hook.py` hooks,https://github.com/pytorch/pytorch/issues/63211,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ddp, module: memory usage",scan when time allows,2021-08-15T16:14:59Z
issue,63207,"document our compatibility with upstream packages (e.g. languages, compilers, tools)",https://github.com/pytorch/pytorch/issues/63207,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2021-08-15T16:14:26Z
issue,63205,split up test/test_autograd.py,https://github.com/pytorch/pytorch/issues/63205,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2021-08-13T13:16:20Z
issue,63016,grid_sample should not require grid to be the same dimension as input.,https://github.com/pytorch/pytorch/issues/63016,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: interpolation, module: nn",scan when time allows,2021-08-13T02:54:32Z
issue,7020,[Caffe2] LT and GT cannot be exported to ONNX,https://github.com/pytorch/pytorch/issues/7020,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-08-12T23:25:07Z
issue,6505,[Caffe2] Onnx to Caffe2 Blob error,https://github.com/pytorch/pytorch/issues/6505,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-08-12T23:13:25Z
issue,63069,CUDA error: invalid configuration argument for torch.sparse tensor backward,https://github.com/pytorch/pytorch/issues/63069,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: sparse",scan when time allows,2021-08-12T22:47:30Z
issue,63057,Find out why TestSparseCPU.test_softmax_cpu_float64 grad and input dtypes are different,https://github.com/pytorch/pytorch/issues/63057,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: structured kernels,scan when time allows,2021-08-12T22:46:30Z
issue,60933,jit.load error when changed Folder Name,https://github.com/pytorch/pytorch/issues/60933,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: serialization,scan when time allows,2021-08-12T22:39:54Z
issue,62606,Guidance on where to add documentation,https://github.com/pytorch/pytorch/issues/62606,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2021-08-12T21:50:31Z
issue,63082,script/build_android nn::Module support,https://github.com/pytorch/pytorch/issues/63082,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: android, oncall: mobile",scan when time allows,2021-08-12T20:51:33Z
issue,62989,Doc page for functions with overload is not properly indented,https://github.com/pytorch/pytorch/issues/62989,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2021-08-12T19:31:30Z
issue,55609,requires_grad does not get propagated properly when using the JIT compiler,https://github.com/pytorch/pytorch/issues/55609,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2021-08-12T18:59:25Z
issue,63078,torch.cuda.amp fails with torch.sparse.softmax,https://github.com/pytorch/pytorch/issues/63078,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision), module: sparse",scan when time allows,2021-08-12T18:55:16Z
issue,38964,[JIT] jit can not recognize the imported function,https://github.com/pytorch/pytorch/issues/38964,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2021-08-12T06:05:49Z
issue,60724,bilinear interpolate is very slow under mixed precision training mode.,https://github.com/pytorch/pytorch/issues/60724,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision), module: cuda, module: interpolation, module: nn",scan when time allows,2021-08-12T04:16:20Z
issue,57815,copy.deepcopy fails on spectral-norm layers after the first forward,https://github.com/pytorch/pytorch/issues/57815,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2021-08-12T03:38:13Z
issue,40208,C++ API for torch.autograd.functional.jacobian,https://github.com/pytorch/pytorch/issues/40208,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: cpp",scan when time allows,2021-08-12T00:21:07Z
issue,60164,Training goes wrong with amp and no_grad since pytorch 1.8,https://github.com/pytorch/pytorch/issues/60164,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision), module: cuda",scan when time allows,2021-08-11T15:12:10Z
issue,20359,[feature request] Run examples from docs as tests,https://github.com/pytorch/pytorch/issues/20359,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2021-08-11T15:03:20Z
issue,42427,./scripts/build_android.sh need support nn::module nn::Functional nn::Linear,https://github.com/pytorch/pytorch/issues/42427,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: android, oncall: mobile",scan when time allows,2021-08-11T14:16:55Z
issue,35723,LibTorch API on Mobile,https://github.com/pytorch/pytorch/issues/35723,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, oncall: mobile",scan when time allows,2021-08-11T13:46:28Z
issue,45073,Compilation errors on power-pc,https://github.com/pytorch/pytorch/issues/45073,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: power",scan when time allows,2021-08-11T02:04:22Z
issue,14095,[feature request] bincount along specified dimension(s),https://github.com/pytorch/pytorch/issues/14095,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: numpy, module: sorting and selection",scan when time allows,2021-08-10T22:25:25Z
issue,62888,Add a mechanism to not universally dispatch to python for all for all operations when using __torch_dispatch__,https://github.com/pytorch/pytorch/issues/62888,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: __torch_dispatch__,scan when time allows,2021-08-10T18:19:13Z
issue,62171,[feature request] torch.nn.Conv3d on tensors having more than 5 dimensions,https://github.com/pytorch/pytorch/issues/62171,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: nn",scan when time allows,2021-08-09T21:05:20Z
issue,62931,GPU compatibility check without initializing CUDA,https://github.com/pytorch/pytorch/issues/62931,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2021-08-09T16:20:49Z
issue,16873,Allow building C++ custom ops that imports another custom ops,https://github.com/pytorch/pytorch/issues/16873,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp-extensions,scan when time allows,2021-08-09T15:47:43Z
issue,62934,`tools/autograd/derivatives.yaml` doesn't support methods on optional tensor,https://github.com/pytorch/pytorch/issues/62934,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: codegen",scan when time allows,2021-08-09T15:25:38Z
issue,45203,Staged backend boxed fallback (per-operator precomputation / precompute),https://github.com/pytorch/pytorch/issues/45203,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dispatch,scan when time allows,2021-08-09T14:49:11Z
issue,62845,allow specification of variadic arguments in native_functions.yaml,https://github.com/pytorch/pytorch/issues/62845,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: internals,scan when time allows,2021-08-09T13:49:25Z
issue,23151,Deterministic mode for scatter_add operation,https://github.com/pytorch/pytorch/issues/23151,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: determinism, module: scatter & gather ops",scan when time allows,2021-08-08T10:08:40Z
issue,58979,gradient layout contract,https://github.com/pytorch/pytorch/issues/58979,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: memory format,scan when time allows,2021-08-07T07:01:29Z
issue,62865,Multiplication of non-int torch.tensor with string gives unexpected TypeError,https://github.com/pytorch/pytorch/issues/62865,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: error checking,scan when time allows,2021-08-06T15:43:13Z
issue,62862,Convert unset variables in `Reducer::Timer` to use c10::optional,https://github.com/pytorch/pytorch/issues/62862,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ddp,scan when time allows,2021-08-06T14:39:30Z
issue,62884,Conjugate fallback dispatch key should be per-backend,https://github.com/pytorch/pytorch/issues/62884,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: complex, module: dispatch",scan when time allows,2021-08-06T14:36:45Z
issue,16708,Better include path when compiling mkldnn,https://github.com/pytorch/pytorch/issues/16708,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2021-08-06T14:35:21Z
issue,16700,[feature request] Store accumulated gradients in separate GPU or on CPU memory,https://github.com/pytorch/pytorch/issues/16700,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: data parallel,scan when time allows,2021-08-06T14:33:59Z
issue,62784,Runtime error when passing dim as None in torch.squeeze (and Tensor.squeeze),https://github.com/pytorch/pytorch/issues/62784,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: numpy,scan when time allows,2021-08-06T06:51:10Z
issue,48714,RTX 3090 setup vs 2x RTX 2080TI setup slower? Help..,https://github.com/pytorch/pytorch/issues/48714,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: data parallel, module: performance",scan when time allows,2021-08-06T06:04:08Z
issue,62790,Remove c10d::kDefaultFirstBucketBytes,https://github.com/pytorch/pytorch/issues/62790,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, module: ddp, oncall: distributed",scan when time allows,2021-08-05T01:11:23Z
issue,62742,Add a way to edit CI env variables from PRs,https://github.com/pytorch/pytorch/issues/62742,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-08-04T18:52:48Z
issue,56038,[Meta] PyTorch features build/test matrix,https://github.com/pytorch/pytorch/issues/56038,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: ci",scan when time allows,2021-08-04T18:35:44Z
issue,16695,Check whether _cudnn_rnn_flatten_weight can avoid changing the TensorImpl or Storage pointer of tensors in `weight_arr`,https://github.com/pytorch/pytorch/issues/16695,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: assert failure, module: cudnn",scan when time allows,2021-08-04T17:28:10Z
issue,16668,Reduce fragmentation with CUDA caching allocator when using many streams,https://github.com/pytorch/pytorch/issues/16668,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: memory usage",scan when time allows,2021-08-04T17:27:49Z
issue,16556,"KeyError: 'No translator registered for layer: name: ""res3b_relu_norm""\ntype: ""Normalize""\nbottom: ""res3b""\ntop: ""res3b_relu_norm""\nnorm_param",https://github.com/pytorch/pytorch/issues/16556,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-08-04T17:15:58Z
issue,16523,"The word ""ints"" is not rendered correclty in docs.",https://github.com/pytorch/pytorch/issues/16523,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2021-08-04T17:09:25Z
issue,16432,A bug in parallel.data_parallel when module_kwargs is not None,https://github.com/pytorch/pytorch/issues/16432,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: batching, module: cuda, module: data parallel, module: error checking",scan when time allows,2021-08-04T17:06:14Z
issue,62500,meta tensor `set_` seems fishy,https://github.com/pytorch/pytorch/issues/62500,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: internals, module: meta tensors",scan when time allows,2021-08-04T16:11:04Z
issue,62448,Dynamic tensor rematerialization,https://github.com/pytorch/pytorch/issues/62448,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: activation checkpointing, module: internals",scan when time allows,2021-08-04T16:06:57Z
issue,62699,Can not parse the caffe2 pretrain model. google.protobuf.message.DecodeError: Error parsing message,https://github.com/pytorch/pytorch/issues/62699,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-08-04T14:32:00Z
issue,61730,"Can't specify compiler, leading to SUPPORT_GLIBCXX_USE_C99 failed",https://github.com/pytorch/pytorch/issues/61730,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2021-08-04T09:40:36Z
issue,62591,Unnecessary python bindings and documentation for internal functions/ops,https://github.com/pytorch/pytorch/issues/62591,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: codegen,scan when time allows,2021-08-04T05:31:24Z
issue,40147,SyncBatchNorm doesn't work when I set track_running_stats False,https://github.com/pytorch/pytorch/issues/40147,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2021-08-04T05:17:14Z
issue,62655,Make KernelFunction::makeFromUnboxedFunctor infer KernelFunctor from input argument,https://github.com/pytorch/pytorch/issues/62655,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, module: dispatch",scan when time allows,2021-08-03T23:12:07Z
issue,62656,"Building pytorch from source results in ""an object with that name is already defined"" error on import",https://github.com/pytorch/pytorch/issues/62656,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: pybind",scan when time allows,2021-08-03T22:50:59Z
issue,62650,Error while importing torch: libtorch_python.so: undefined symbol: PyThread_tss_alloc,https://github.com/pytorch/pytorch/issues/62650,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: abi,scan when time allows,2021-08-03T15:49:36Z
issue,57136,Intermittent segmentation fault when using rpc_async/rpc_sync with CUDA tensors,https://github.com/pytorch/pytorch/issues/57136,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2021-08-03T14:27:14Z
issue,62602,"""test_events_wait"" is flaky on ROCm",https://github.com/pytorch/pytorch/issues/62602,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: flaky-tests, module: rocm",scan when time allows,2021-08-03T14:18:21Z
issue,62446,[docs] Can't get header links to h3 on docs pages,https://github.com/pytorch/pytorch/issues/62446,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: doc infra,scan when time allows,2021-08-03T08:06:39Z
issue,52236,CosineAnnealingWarmRestarts LR scheduler fails when lash_epoch != -1,https://github.com/pytorch/pytorch/issues/52236,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2021-08-03T07:39:34Z
issue,62566,`c10::optional<T>` operators should delegate to corresponding operators on `T`,https://github.com/pytorch/pytorch/issues/62566,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp,scan when time allows,2021-08-02T19:48:52Z
issue,62451,torch.Generator ignores the device argument?,https://github.com/pytorch/pytorch/issues/62451,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: random,scan when time allows,2021-08-02T15:28:59Z
issue,62449,Review replacing test/test_namedtuple_return_api.py with an OpInfo-based test,https://github.com/pytorch/pytorch/issues/62449,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tests,scan when time allows,2021-07-30T03:53:44Z
issue,61591,Add `map_location` arg for c10d collective comms,https://github.com/pytorch/pytorch/issues/61591,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2021-07-29T21:06:39Z
issue,62399,Understatement about how running_var is computed in BatchNorm,https://github.com/pytorch/pytorch/issues/62399,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: norms and normalization",scan when time allows,2021-07-29T19:15:09Z
issue,61659,Python program locks up when using some innocuous code inside mp.spawn(),https://github.com/pytorch/pytorch/issues/61659,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2021-07-28T23:47:44Z
issue,62343,maybe_wrap_dim should have an example attached to it,https://github.com/pytorch/pytorch/issues/62343,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2021-07-28T18:38:06Z
issue,62147,Tensor subclassing does not support bool input,https://github.com/pytorch/pytorch/issues/62147,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: boolean tensor, module: tensor creation",scan when time allows,2021-07-28T03:30:35Z
issue,16328,Should torch.arange take a layout parameter?,https://github.com/pytorch/pytorch/issues/16328,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: sparse, module: tensor creation, module: ux",scan when time allows,2021-07-27T19:26:00Z
issue,62273,Serialization map_location silently ignores xla/mlc/meta (any serialization mechanism that skips storage),https://github.com/pytorch/pytorch/issues/62273,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: serialization,scan when time allows,2021-07-27T18:47:57Z
issue,62271,Refactor serialization tests to use device parametrization,https://github.com/pytorch/pytorch/issues/62271,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: serialization, module: tests",scan when time allows,2021-07-27T18:47:33Z
issue,38995,[FR] [distributed] coalesced primitives,https://github.com/pytorch/pytorch/issues/38995,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2021-07-27T17:41:53Z
issue,37063,Misleading documentation in torch.nn.functional.fold,https://github.com/pytorch/pytorch/issues/37063,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: nn",scan when time allows,2021-07-27T05:37:52Z
issue,62026,Should we be upcasting integral types to int64 in torch.sum and torch.prod?,https://github.com/pytorch/pytorch/issues/62026,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: numpy, module: python array api, module: reductions",scan when time allows,2021-07-26T22:10:24Z
issue,61662,`super().__init__()` usage for `_Joinable`,https://github.com/pytorch/pytorch/issues/61662,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2021-07-26T19:01:40Z
issue,61661,Return `Future` instead of `Work` in `notify_join_context()`,https://github.com/pytorch/pytorch/issues/61661,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2021-07-26T19:01:31Z
issue,28794,torch.from_file not documented,https://github.com/pytorch/pytorch/issues/28794,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2021-07-26T02:45:13Z
issue,62130,"nn.DataParallel not working on A100 with nvidia-driver 470.42.01, Cuda 11.1, Debian 10",https://github.com/pytorch/pytorch/issues/62130,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: data parallel, module: multi-gpu",scan when time allows,2021-07-26T02:28:35Z
issue,32083,build pytorch from source fauled: undefined reference to `cusparseGetErrorString(cusparseStatus_t)',https://github.com/pytorch/pytorch/issues/32083,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2021-07-24T12:46:18Z
issue,62009,multi-input and multi-output support for nn.Module,https://github.com/pytorch/pytorch/issues/62009,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2021-07-23T02:36:11Z
issue,35041,Make operator registrations truly commutative using priority,https://github.com/pytorch/pytorch/issues/35041,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dispatch,scan when time allows,2021-07-22T15:57:03Z
issue,62031,Caffe2 tensor to ATen conversion doesn't initialize PyTorch CUDA state,https://github.com/pytorch/pytorch/issues/62031,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: internals",scan when time allows,2021-07-22T15:36:19Z
issue,62016,ppc64le build fail: invalid conversion from Bfloat16 in functional_base.h:180,https://github.com/pytorch/pytorch/issues/62016,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bfloat16, module: power",scan when time allows,2021-07-22T15:32:26Z
issue,61772,Can I set the algorithm of torch.nn.Conv to CUDNN_CONVOLUTION FWD_ALGO_GEMM by myself?,https://github.com/pytorch/pytorch/issues/61772,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: cuda, module: cudnn",scan when time allows,2021-07-21T23:44:01Z
issue,57406,nn.Module custom setattr leads very confusing behaviors,https://github.com/pytorch/pytorch/issues/57406,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2021-07-21T06:47:35Z
issue,61930,NNAPI delegate library cannot be built on MacOS,https://github.com/pytorch/pytorch/issues/61930,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: macos",scan when time allows,2021-07-20T21:30:38Z
issue,61923,Calling `torch.save()` twice writes a corrupted file,https://github.com/pytorch/pytorch/issues/61923,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: serialization,scan when time allows,2021-07-20T19:29:52Z
issue,61919,RuntimeError: !ref.requires_grad()INTERNAL ASSERT FAILED,https://github.com/pytorch/pytorch/issues/61919,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2021-07-20T19:14:53Z
issue,60426,"Output of non-inplace, non-CompositeImplicitAutograd op has TensorImpl > 1 or StorageImpl use_count != 1",https://github.com/pytorch/pytorch/issues/60426,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2021-07-20T15:46:27Z
issue,58093,"TrilinearBackward takes 98.4% of total computational time, is this to be expected?",https://github.com/pytorch/pytorch/issues/58093,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: performance",scan when time allows,2021-07-20T15:36:07Z
issue,16148,CategoricalCrossEntropy Loss runs with wrong tag,https://github.com/pytorch/pytorch/issues/16148,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: loss, module: nn",scan when time allows,2021-07-20T13:17:07Z
issue,61839,died with <Signals.SIGSEGV: 11>,https://github.com/pytorch/pytorch/issues/61839,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2021-07-20T08:36:58Z
issue,16090,build failure with NNAPI enabled (Caffe2 path but under pytorch umbrella),https://github.com/pytorch/pytorch/issues/16090,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2021-07-19T21:36:36Z
issue,61827,Bail out if `-std=c++` setting is detected during build,https://github.com/pytorch/pytorch/issues/61827,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2021-07-19T21:35:49Z
issue,61843,Investigate fairscale's use of storage APIs and find alternatives,https://github.com/pytorch/pytorch/issues/61843,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-07-19T21:34:19Z
issue,60956,Naming convention for variants of operations that never short circuit,https://github.com/pytorch/pytorch/issues/60956,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ux,scan when time allows,2021-07-19T19:06:34Z
issue,61695,miss header file while using python setup.py install to install pytorch with anaconda env,https://github.com/pytorch/pytorch/issues/61695,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2021-07-19T17:31:25Z
issue,61538,The use of epsilon in torch.nn.functional.pairwise_distance leads to incorrect minima of the distances,https://github.com/pytorch/pytorch/issues/61538,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: distance functions, module: nn",scan when time allows,2021-07-19T17:13:47Z
issue,61513,Adding 'differentiable' section to each function docstring,https://github.com/pytorch/pytorch/issues/61513,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: docs",scan when time allows,2021-07-19T17:10:54Z
issue,61825,Improved OpInfo dtype testing,https://github.com/pytorch/pytorch/issues/61825,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tests,scan when time allows,2021-07-19T05:46:54Z
issue,57534,`torch.view_as_complex()` does not work when `storage_offset` is odd,https://github.com/pytorch/pytorch/issues/57534,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: complex,scan when time allows,2021-07-19T03:07:05Z
issue,48832,cudnn cannot be pickled by cloudpickle,https://github.com/pytorch/pytorch/issues/48832,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: cudnn, module: serialization",scan when time allows,2021-07-16T17:59:42Z
issue,61665,Default Collate doesn't work for subtypes of ndarray,https://github.com/pytorch/pytorch/issues/61665,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2021-07-16T16:21:33Z
issue,61718,ATen codegen is always re-running for Makefile generator,https://github.com/pytorch/pytorch/issues/61718,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2021-07-15T19:13:54Z
issue,61658,Move `torch.cholesky_solve` into `torch.linalg`.,https://github.com/pytorch/pytorch/issues/61658,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: linear algebra,scan when time allows,2021-07-15T14:26:45Z
issue,61410,requires grad get lost during transform.,https://github.com/pytorch/pytorch/issues/61410,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: deprecation, module: tensor creation",scan when time allows,2021-07-15T06:24:51Z
issue,61645,`channel_shuffle` output is sometimes aliased with its input,https://github.com/pytorch/pytorch/issues/61645,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: nn",scan when time allows,2021-07-14T15:46:49Z
issue,61619,"When 'trapezoid' is called with an empty tensor input, it does not produce an output with requires_grad",https://github.com/pytorch/pytorch/issues/61619,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2021-07-14T15:20:07Z
issue,61614,Possibly flaky test: TestMultiprocessing.test_fd_sharing,https://github.com/pytorch/pytorch/issues/61614,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: flaky-tests,scan when time allows,2021-07-14T15:18:41Z
issue,61650,"Move `{svd,pca}_lowrank` into `torch.linalg`",https://github.com/pytorch/pytorch/issues/61650,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: linear algebra,scan when time allows,2021-07-14T15:10:17Z
issue,51014,[pipeline] gpu util + peak mem reporting to tune partitions and chunks,https://github.com/pytorch/pytorch/issues/51014,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2021-07-13T19:16:53Z
issue,61149,`pad(mode='reflect')` beyond `input.size`,https://github.com/pytorch/pytorch/issues/61149,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: padding",scan when time allows,2021-07-13T16:38:21Z
issue,61486,Some reduction operators have double signatures,https://github.com/pytorch/pytorch/issues/61486,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: deprecation, module: python array api, module: reductions",scan when time allows,2021-07-13T15:06:12Z
issue,61544,`__torch_function__` always records default kwargs as kwargs,https://github.com/pytorch/pytorch/issues/61544,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: __torch_function__,scan when time allows,2021-07-13T14:25:56Z
issue,61503,UCI Data Sets,https://github.com/pytorch/pytorch/issues/61503,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2021-07-13T13:39:46Z
issue,31943,error when building pytorch 1.1.0 from source,https://github.com/pytorch/pytorch/issues/31943,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2021-07-13T12:12:50Z
issue,61452,New module: Split log softmax with loss,https://github.com/pytorch/pytorch/issues/61452,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2021-07-13T08:01:44Z
issue,61516,MemoryError with pip Install even with --no-cache-dir option,https://github.com/pytorch/pytorch/issues/61516,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: binaries,scan when time allows,2021-07-13T04:39:59Z
issue,61523,F.nll_loss with 16-bit CUDA tensors and reduction=mean produces NaNs,https://github.com/pytorch/pytorch/issues/61523,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: loss, module: nans and infs, module: numerical-stability, module: reductions",scan when time allows,2021-07-12T22:49:16Z
issue,51509,Massive Performance bottlenecks in some of the Reduce operations.,https://github.com/pytorch/pytorch/issues/51509,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: performance, module: reductions, module: tensoriterator",scan when time allows,2021-07-12T21:35:39Z
issue,61515,Buggy page with overlapped tabs,https://github.com/pytorch/pytorch/issues/61515,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2021-07-12T21:29:24Z
issue,61525,"I have graphic card of P4, but torch.cuda.is_available() always shows false",https://github.com/pytorch/pytorch/issues/61525,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2021-07-12T20:46:24Z
issue,61528,Function Request: scipy.ndimage.map_coordinates,https://github.com/pytorch/pytorch/issues/61528,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: interpolation, module: numpy",scan when time allows,2021-07-12T20:42:46Z
issue,61136,rename(**{}) should throw no error,https://github.com/pytorch/pytorch/issues/61136,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: named tensor,scan when time allows,2021-07-12T16:10:53Z
issue,61510,What is find_package(Torch REQUIRED) doing that a manual include/glob doesnt?,https://github.com/pytorch/pytorch/issues/61510,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2021-07-12T15:39:08Z
issue,15812,Possible regression in incremental build,https://github.com/pytorch/pytorch/issues/15812,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2021-07-12T15:38:01Z
issue,61485,CompositeImplicitAutograd ops should not call out= variants of operators,https://github.com/pytorch/pytorch/issues/61485,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2021-07-12T15:13:11Z
issue,15617,Feature request: transposed locally connected layer,https://github.com/pytorch/pytorch/issues/15617,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2021-07-12T13:32:44Z
issue,15607,documentation for adding a new type via C++ extensions,https://github.com/pytorch/pytorch/issues/15607,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bfloat16, module: complex, module: docs",scan when time allows,2021-07-12T13:28:20Z
issue,61411,"I do have the GPU, but the print(torch.cuda.is_available()) always shows False",https://github.com/pytorch/pytorch/issues/61411,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2021-07-11T14:19:31Z
issue,40550,RemoteModule enhancements,https://github.com/pytorch/pytorch/issues/40550,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, module: rpc",scan when time allows,2021-07-10T06:45:41Z
issue,58745,"`torch.(min|max)(..., dim=...)` diverges from array API specification",https://github.com/pytorch/pytorch/issues/58745,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: deprecation, module: python array api, module: reductions",scan when time allows,2021-07-09T22:41:39Z
issue,56926,CUDA 11 tracking issue,https://github.com/pytorch/pytorch/issues/56926,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: cuda",scan when time allows,2021-07-09T17:55:39Z
issue,61453,"Reduce with any(), all(), median() over multiple dimensions",https://github.com/pytorch/pytorch/issues/61453,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: numpy, module: reductions",scan when time allows,2021-07-09T17:02:56Z
issue,55442,More AVX2 vectorization support for half (float16),https://github.com/pytorch/pytorch/issues/55442,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: half, module: vectorization",scan when time allows,2021-07-09T16:15:25Z
issue,61186,possible SummaryWriter memory leak,https://github.com/pytorch/pytorch/issues/61186,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tensorboard,scan when time allows,2021-07-09T15:43:17Z
issue,61289,Zero seed in DistributedSampler,https://github.com/pytorch/pytorch/issues/61289,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2021-07-08T23:47:58Z
issue,54082,Dispatch table for linalg_norm is fishy,https://github.com/pytorch/pytorch/issues/54082,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dispatch, module: norms and normalization",scan when time allows,2021-07-08T14:12:25Z
issue,51456,Optional tensor type,https://github.com/pytorch/pytorch/issues/51456,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-07-07T19:50:10Z
issue,59218,Clarify BLAS configuration option,https://github.com/pytorch/pytorch/issues/59218,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: docs, module: linear algebra",scan when time allows,2021-07-07T17:45:39Z
issue,61261,libtorch_cpu.so is getting too large.,https://github.com/pytorch/pytorch/issues/61261,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: releng,scan when time allows,2021-07-07T04:14:29Z
issue,61260,libtorch_cpu.so: undefined symbol: _ZN3c1010ThreadPool3runESt8functionIFvvEE,https://github.com/pytorch/pytorch/issues/61260,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: abi, module: build",scan when time allows,2021-07-06T21:33:54Z
issue,9678,Remove BatchNorm layers once the training is completed.,https://github.com/pytorch/pytorch/issues/9678,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: norms and normalization",scan when time allows,2021-07-06T20:31:43Z
issue,5157,BCELoss - weight parameter shape incorrect,https://github.com/pytorch/pytorch/issues/5157,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: loss, module: nn",scan when time allows,2021-07-06T20:30:21Z
issue,61213,[feature request] Low pass filtering of FFT results,https://github.com/pytorch/pytorch/issues/61213,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fft,scan when time allows,2021-07-06T20:18:57Z
issue,61211,[feature request] Do zero-padding in high-frequency modes in `ifft`,https://github.com/pytorch/pytorch/issues/61211,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fft,scan when time allows,2021-07-06T20:18:48Z
issue,61233,[typing] ModuleDict ctor has wrong signature,https://github.com/pytorch/pytorch/issues/61233,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: typing,scan when time allows,2021-07-06T19:45:39Z
issue,59099,CSR: Relaxing constraints to s_addmm_out_sparse_dense_cuda_worker,https://github.com/pytorch/pytorch/issues/59099,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2021-07-06T13:22:12Z
issue,49752,IndexError: Input _features.0.weight is undefined!,https://github.com/pytorch/pytorch/issues/49752,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: vision,scan when time allows,2021-07-05T06:02:55Z
issue,56522,[cpp op] TORCH_LIBRARY schema doesn't respect keyword only,https://github.com/pytorch/pytorch/issues/56522,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: cpp-extensions",scan when time allows,2021-07-04T19:10:44Z
issue,22415,Add support for serializing Mkldnn Tensor,https://github.com/pytorch/pytorch/issues/22415,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: mkldnn, module: serialization",scan when time allows,2021-07-04T14:22:36Z
issue,29758,[doc] Tensor.mean: dtype kwarg is not documented,https://github.com/pytorch/pytorch/issues/29758,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: reductions",scan when time allows,2021-07-02T01:33:58Z
issue,27312,torch.Tensor.mean erroneously documented as sometimes returning a tuple,https://github.com/pytorch/pytorch/issues/27312,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: reductions",scan when time allows,2021-07-02T01:33:58Z
issue,18904,[FR] torch.dist along a dimension,https://github.com/pytorch/pytorch/issues/18904,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: reductions,scan when time allows,2021-07-02T01:31:00Z
issue,56586,"Support multi-dim reductions for torch.prod, torch.all, torch.any",https://github.com/pytorch/pytorch/issues/56586,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: reductions,scan when time allows,2021-07-02T01:23:13Z
issue,60381,broken source link / missing anchors,https://github.com/pytorch/pytorch/issues/60381,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2021-07-01T20:05:54Z
issue,61117,flatten renames in place,https://github.com/pytorch/pytorch/issues/61117,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: named tensor,scan when time allows,2021-07-01T16:31:13Z
issue,47027,Fix the way imports are done to be more correct for static type checkers,https://github.com/pytorch/pytorch/issues/47027,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: typing,scan when time allows,2021-07-01T15:57:23Z
issue,28239,support class annotations in __init__,https://github.com/pytorch/pytorch/issues/28239,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2021-07-01T15:40:35Z
issue,45565,Resolving overloads and default arguments from within __torch_function__,https://github.com/pytorch/pytorch/issues/45565,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: __torch_function__,scan when time allows,2021-07-01T14:51:16Z
issue,59936,Can't perform any operation on Vulkan device - macOS M1,https://github.com/pytorch/pytorch/issues/59936,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: arm, module: macos, module: vulkan",scan when time allows,2021-07-01T14:03:34Z
issue,20206,Failed to build with system protobuf,https://github.com/pytorch/pytorch/issues/20206,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: protobuf",scan when time allows,2021-07-01T12:09:32Z
issue,60541,torch.cuda.Event(blocking=True) doesn't work,https://github.com/pytorch/pytorch/issues/60541,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2021-06-30T14:32:39Z
issue,60997,Dispatcher doesn't handle ops with an empty list of tensors (e.g. `torch.cat()`),https://github.com/pytorch/pytorch/issues/60997,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dispatch,scan when time allows,2021-06-30T00:15:55Z
issue,60941,Incorrect dtype cast for binary ops w/ mixed ComplexFloat + Double operands,https://github.com/pytorch/pytorch/issues/60941,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: complex, module: type promotion",scan when time allows,2021-06-30T00:01:29Z
issue,60950,Singular build setting CLANG_CXX_LANGUAGE_STANDARD has different values,https://github.com/pytorch/pytorch/issues/60950,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: macos",scan when time allows,2021-06-29T23:57:37Z
issue,60977,[package] pattern usage can potentially create broken packages,https://github.com/pytorch/pytorch/issues/60977,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: package/deploy,scan when time allows,2021-06-29T22:12:27Z
issue,60991,[NNC] APIs needed to bridge scheduling gap in reductions,https://github.com/pytorch/pytorch/issues/60991,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-06-29T21:26:23Z
issue,56187,AVX512 and Vec512,https://github.com/pytorch/pytorch/issues/56187,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: vectorization,scan when time allows,2021-06-29T03:58:13Z
issue,60911,[elastic launcher] redirects/tee support for global rank,https://github.com/pytorch/pytorch/issues/60911,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: elastic,scan when time allows,2021-06-28T23:57:33Z
issue,53658,torch.distributed with NCCL can hang if the first operation is a barrier,https://github.com/pytorch/pytorch/issues/53658,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: c10d, oncall: distributed",scan when time allows,2021-06-28T20:44:30Z
issue,54591,Various issues in derivatives.yaml,https://github.com/pytorch/pytorch/issues/54591,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2021-06-28T17:47:20Z
issue,60802,Bug in `pytorch/benchmarks/tensorexpr/pt_engine.py`,https://github.com/pytorch/pytorch/issues/60802,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-06-28T16:21:30Z
issue,60796,AccumulateGrad short circuits before calling variable hooks when the incoming grad is None,https://github.com/pytorch/pytorch/issues/60796,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2021-06-28T16:04:32Z
issue,60843,PyTorch's last 1 year update became unfriendly due to the lack of examples,https://github.com/pytorch/pytorch/issues/60843,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2021-06-28T15:38:57Z
issue,60848,Fail to build with gcc11,https://github.com/pytorch/pytorch/issues/60848,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: xnnpack",scan when time allows,2021-06-28T15:29:41Z
issue,60431,Implement simple view methods for TensorAccessor/PackedTensorAccessor,https://github.com/pytorch/pytorch/issues/60431,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: bootcamp,scan when time allows,2021-06-28T15:10:56Z
issue,60834,complie error when i use python setup.py build on arm computer,https://github.com/pytorch/pytorch/issues/60834,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: arm, module: build",scan when time allows,2021-06-28T13:40:18Z
issue,46272,test_distributed_* does not work with run_test.py -i option,https://github.com/pytorch/pytorch/issues/46272,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: tests, oncall: distributed",scan when time allows,2021-06-28T13:34:47Z
issue,47281,Error: invalid use of register,https://github.com/pytorch/pytorch/issues/47281,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2021-06-28T07:40:38Z
issue,60520,[docs] torch.log_softmax is undocumented,https://github.com/pytorch/pytorch/issues/60520,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2021-06-28T03:47:08Z
issue,49810,Undefined reference to `cv::imread`,https://github.com/pytorch/pytorch/issues/49810,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2021-06-27T12:12:31Z
issue,15386,F.grid_sample doesn't respect padding_mode when height of inputs is 1,https://github.com/pytorch/pytorch/issues/15386,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: interpolation",scan when time allows,2021-06-25T21:58:12Z
issue,60766,"_foreach_maximum returns ""incompatible type"" mypy error unexpectedly",https://github.com/pytorch/pytorch/issues/60766,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: typing,scan when time allows,2021-06-25T21:49:12Z
issue,60306,Divergent code is needed to record usage streams on different TensorImpl types,https://github.com/pytorch/pytorch/issues/60306,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: sparse, module: vmap",scan when time allows,2021-06-25T04:54:04Z
issue,53952,[Pipe] supporting None and non-Tensors in forward's input/output,https://github.com/pytorch/pytorch/issues/53952,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-06-25T00:25:04Z
issue,60674,Turn on -Winfinite-recursion in OSS CI,https://github.com/pytorch/pytorch/issues/60674,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2021-06-24T19:39:31Z
issue,37751,Documentation of _CtxMethodMixin: must be tensors?,https://github.com/pytorch/pytorch/issues/37751,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: docs",scan when time allows,2021-06-24T15:35:48Z
issue,39370,Draw quantized tensors in tensorboard,https://github.com/pytorch/pytorch/issues/39370,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tensorboard,scan when time allows,2021-06-24T11:17:10Z
issue,56161,"[rfc] Add a ""core only"" build flag",https://github.com/pytorch/pytorch/issues/56161,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2021-06-24T10:18:22Z
issue,60372,Docs for `Installing C++ Distributions of PyTorch` needs update,https://github.com/pytorch/pytorch/issues/60372,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2021-06-24T05:33:47Z
issue,60459,[Mkldnn] has_bf16 check only works on Linux for tests,https://github.com/pytorch/pytorch/issues/60459,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: mkldnn, module: tests",scan when time allows,2021-06-24T05:24:56Z
issue,60288,provide hashsum for downloads,https://github.com/pytorch/pytorch/issues/60288,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: binaries,scan when time allows,2021-06-24T04:51:24Z
issue,60400,Using inbuilt function name `input` as variable names in examples,https://github.com/pytorch/pytorch/issues/60400,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2021-06-24T02:09:24Z
issue,60561,nn.CrossMapLRN2d is missing from docs,https://github.com/pytorch/pytorch/issues/60561,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: nn",scan when time allows,2021-06-24T01:44:32Z
issue,60146,[torch.profiler] enhancements + corrections,https://github.com/pytorch/pytorch/issues/60146,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: profiler,scan when time allows,2021-06-24T00:26:03Z
issue,47149,Allow all torch.nn modules to accept arbitrary batch dimensions,https://github.com/pytorch/pytorch/issues/47149,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: batching, module: nn, module: ux",scan when time allows,2021-06-23T21:45:42Z
issue,57157,Using register_full_backward_hook with target module for intermediate activation,https://github.com/pytorch/pytorch/issues/57157,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: docs, module: nn",scan when time allows,2021-06-23T15:51:45Z
issue,60277,Sparse CSR tensor should not accept equal column indices in the same row,https://github.com/pytorch/pytorch/issues/60277,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: error checking, module: sparse",scan when time allows,2021-06-23T14:59:46Z
issue,14790,Tensor.copy_() seems to work improperly with numpy/list indices,https://github.com/pytorch/pytorch/issues/14790,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2021-06-23T11:22:24Z
issue,52575,Error: ‘str’ is not a member of ‘c10’; did you mean ‘c10::aten::str’? while using libtorch,https://github.com/pytorch/pytorch/issues/52575,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp,scan when time allows,2021-06-23T11:15:58Z
issue,60448,Enable large file support,https://github.com/pytorch/pytorch/issues/60448,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2021-06-23T02:25:41Z
issue,57773,BFloat16 CUDA GEMM ops unsupported on Nvidia P100 (SM_60) on CUDA 11.3,https://github.com/pytorch/pytorch/issues/57773,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bfloat16, module: cublas, module: cuda",scan when time allows,2021-06-22T19:01:45Z
issue,32288,Allow range in dim argument of reducing operations such as sum,https://github.com/pytorch/pytorch/issues/32288,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: reductions, module: ux",scan when time allows,2021-06-22T18:21:53Z
issue,21760,[FR] faster reduce sum on expanded/unfolded tensors,https://github.com/pytorch/pytorch/issues/21760,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: linear algebra, module: performance",scan when time allows,2021-06-22T18:19:13Z
issue,60343,What changes we need to make in metrics calculation and visualization part when we use Distributed Data Parallel for distributed training,https://github.com/pytorch/pytorch/issues/60343,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ddp, oncall: distributed",scan when time allows,2021-06-22T07:15:26Z
issue,37113,pip install torch==1.4.0 is broken when using CUDA 10.1,https://github.com/pytorch/pytorch/issues/37113,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2021-06-21T21:51:50Z
issue,58814,TensorIterator for sparse layouts,https://github.com/pytorch/pytorch/issues/58814,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2021-06-21T21:25:06Z
issue,60159,Some ops throw runtime errors when called with complex tensors that require_grad but work when requires_grad=False,https://github.com/pytorch/pytorch/issues/60159,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: complex",scan when time allows,2021-06-21T16:22:08Z
issue,57158,Increasing RTT when using rpc_async in a PS architecture setup,https://github.com/pytorch/pytorch/issues/57158,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2021-06-21T15:35:27Z
issue,12659,Differentiation through Module parameters updates,https://github.com/pytorch/pytorch/issues/12659,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: nn, module: optimizer",scan when time allows,2021-06-20T20:56:02Z
issue,60096,Tutorials that require `pip install` not loading correctly in Colab,https://github.com/pytorch/pytorch/issues/60096,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2021-06-20T13:46:06Z
issue,50323,running a job on multiple gpus with qsub,https://github.com/pytorch/pytorch/issues/50323,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2021-06-20T06:49:12Z
issue,45448,CuDNN version not found,https://github.com/pytorch/pytorch/issues/45448,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cudnn",scan when time allows,2021-06-20T01:54:07Z
issue,60264,Subtensor operations like `tril/triu` to have an option to respect the strides of the input,https://github.com/pytorch/pytorch/issues/60264,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: linear algebra, module: memory format",scan when time allows,2021-06-19T22:49:38Z
issue,41852,Difference in inference time between CUDA 10.0 & 10.2,https://github.com/pytorch/pytorch/issues/41852,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: cudnn, module: performance",scan when time allows,2021-06-19T13:25:23Z
issue,60165,torch.load non backwards compatible on Transformer between 1.8.1 and 1.9.0,https://github.com/pytorch/pytorch/issues/60165,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: serialization,scan when time allows,2021-06-19T09:51:26Z
issue,44159,"[Feature Request] Add an option to run GPU tests only, and skip all CPU tests",https://github.com/pytorch/pytorch/issues/44159,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tests,scan when time allows,2021-06-18T18:02:16Z
issue,15298,Momentum problem (1-momentum is correct?) in BatchNorm2d,https://github.com/pytorch/pytorch/issues/15298,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: nn",scan when time allows,2021-06-18T15:26:54Z
issue,15303,CMake Error at CMakeLists.txt:10 (find_package) in C++,https://github.com/pytorch/pytorch/issues/15303,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2021-06-18T15:26:15Z
issue,59935,Error /usr/local/lib/libopenblas.so: error adding symbols: File in wrong format while building pytorch for ppc64le,https://github.com/pytorch/pytorch/issues/59935,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: power",scan when time allows,2021-06-18T06:21:17Z
issue,60153,[oneline docs] adding a source link for non-python functions,https://github.com/pytorch/pytorch/issues/60153,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2021-06-18T05:17:14Z
issue,15259,err:torch.nn.CrossEntropyLoss,https://github.com/pytorch/pytorch/issues/15259,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2021-06-17T22:35:54Z
issue,15260,MultiGPU for gru,https://github.com/pytorch/pytorch/issues/15260,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: data parallel, module: rnn",scan when time allows,2021-06-17T22:34:54Z
issue,15284,index_add_ with scalar values instead of tensors,https://github.com/pytorch/pytorch/issues/15284,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: advanced indexing,scan when time allows,2021-06-17T22:32:41Z
issue,15288,possible unsafety in torch.distributions.kl_divergence for Bernoullis,https://github.com/pytorch/pytorch/issues/15288,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: distributions, module: nans and infs",scan when time allows,2021-06-17T22:31:36Z
issue,59855,pytorch test failed,https://github.com/pytorch/pytorch/issues/59855,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: cuda, module: tests",scan when time allows,2021-06-17T20:18:38Z
issue,60102,Allow packaging of classes defined in a notebook,https://github.com/pytorch/pytorch/issues/60102,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-06-17T05:02:16Z
issue,45428,Use libtorch for online inference: an illegal memory access was encountered,https://github.com/pytorch/pytorch/issues/45428,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2021-06-17T03:32:59Z
issue,59925,Implicit module registration permits silent programming errors in torch.nn.Module,https://github.com/pytorch/pytorch/issues/59925,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2021-06-16T21:23:10Z
issue,60063,Replace native NHWC BN kernel,https://github.com/pytorch/pytorch/issues/60063,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: memory format,scan when time allows,2021-06-16T19:17:42Z
issue,60092,`memory_format` flag for `tensor.new_empty()`.,https://github.com/pytorch/pytorch/issues/60092,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: memory format,scan when time allows,2021-06-16T18:53:07Z
issue,60083,caffe_translator.py Unable to convert Power Layer,https://github.com/pytorch/pytorch/issues/60083,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-06-16T15:23:23Z
issue,59783,[Feature Request] Deterministic implementation for AdaptiveMaxpool1d.,https://github.com/pytorch/pytorch/issues/59783,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: determinism, module: nn, module: pooling",scan when time allows,2021-06-16T09:06:52Z
issue,15142,Update third_party/googletest - Ability to skip tests in GTEST,https://github.com/pytorch/pytorch/issues/15142,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: tests, module: third_party",scan when time allows,2021-06-15T22:24:05Z
issue,17688,Improve save() method in torch.jit.ScriptModule,https://github.com/pytorch/pytorch/issues/17688,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2021-06-15T22:06:00Z
issue,58841,[bug] torch.topk sometimes supports `float16` and sometimes doesn't,https://github.com/pytorch/pytorch/issues/58841,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: half, module: sorting and selection",scan when time allows,2021-06-14T20:10:58Z
issue,59896,Autograd engine current graph_task should be in the TLSState,https://github.com/pytorch/pytorch/issues/59896,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2021-06-14T18:14:52Z
issue,59854,Misreported linting messages,https://github.com/pytorch/pytorch/issues/59854,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: lint,scan when time allows,2021-06-14T16:15:55Z
issue,36066,Add Specific Warning/Error For Unsupported GPU or Systems,https://github.com/pytorch/pytorch/issues/36066,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2021-06-11T23:10:13Z
issue,59778,CUDA Memory Error: PyTorch doen't allocate memory even though it's available,https://github.com/pytorch/pytorch/issues/59778,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: memory usage",scan when time allows,2021-06-11T04:46:43Z
issue,47028,torch.autograd.backward() fails to sync with other stream,https://github.com/pytorch/pytorch/issues/47028,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2021-06-11T00:10:14Z
issue,59822,TransformedDistribution's log_prob gradient is inconsistent w.r.t. cache_size,https://github.com/pytorch/pytorch/issues/59822,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distributions,scan when time allows,2021-06-10T23:09:49Z
issue,56774,Support `expand_dims`,https://github.com/pytorch/pytorch/issues/56774,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: numpy, module: python array api",scan when time allows,2021-06-10T07:55:33Z
issue,59627,Example code for is_storage missing,https://github.com/pytorch/pytorch/issues/59627,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2021-06-10T07:45:32Z
issue,59493,ctc loss bug,https://github.com/pytorch/pytorch/issues/59493,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: loss",scan when time allows,2021-06-10T00:37:50Z
issue,59743,m.fallback(torch::CppFunction::makeFromBoxedFunction<&my_fallback>) gives bad error message,https://github.com/pytorch/pytorch/issues/59743,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: assert failure, module: cpp, module: internals",scan when time allows,2021-06-09T21:09:42Z
issue,54332,Unit testing failures when porting PyTorch wheel,https://github.com/pytorch/pytorch/issues/54332,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: build",scan when time allows,2021-06-09T18:12:41Z
issue,23780,Construction of MultivariateNormal much slower on GPU than CPU,https://github.com/pytorch/pytorch/issues/23780,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: distributions, module: performance",scan when time allows,2021-06-09T09:20:17Z
issue,49405,PyTorch 1.7.1 on (macOS/python 3.9/conda) links libtorch_global_deps.dylib with libomp.dylib instead of libiomp5.dylib,https://github.com/pytorch/pytorch/issues/49405,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: build, module: macos",scan when time allows,2021-06-08T20:05:20Z
issue,59617,torch.split() infer -1 entry from the other split sizes,https://github.com/pytorch/pytorch/issues/59617,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: viewing and reshaping,scan when time allows,2021-06-08T14:41:27Z
issue,59438,"DataLoader worker (pid(s) 18056, 20540, 4512) exited unexpectedly",https://github.com/pytorch/pytorch/issues/59438,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2021-06-08T14:03:41Z
issue,59542,Build Pytorch with Aten?,https://github.com/pytorch/pytorch/issues/59542,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2021-06-07T20:22:51Z
issue,59531,Failed to enable libuv,https://github.com/pytorch/pytorch/issues/59531,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2021-06-07T20:20:26Z
issue,59546,Multiprocessing model evaluation gets stuck,https://github.com/pytorch/pytorch/issues/59546,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: multiprocessing,scan when time allows,2021-06-07T20:18:08Z
issue,59515,Conv1d with large batch size and half precision in cuda returns incorrect result,https://github.com/pytorch/pytorch/issues/59515,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: cuda",scan when time allows,2021-06-07T20:02:13Z
issue,59575,[clang-tidy] check for iteration over unordered data structures,https://github.com/pytorch/pytorch/issues/59575,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-06-07T19:36:28Z
issue,56440,Raise exception when torch.clamp min value underflows the input tensor's dtype,https://github.com/pytorch/pytorch/issues/56440,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: error checking,scan when time allows,2021-06-07T18:37:53Z
issue,58479,Make SavedVariables aware of pyobjects,https://github.com/pytorch/pytorch/issues/58479,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2021-06-07T18:12:36Z
issue,59262,"Possibly alignment issues on NEON vectorized ops, Jetson platforms",https://github.com/pytorch/pytorch/issues/59262,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: arm, module: jetson",scan when time allows,2021-06-07T17:16:27Z
issue,59140,No index type check of torch.tensor in array manner,https://github.com/pytorch/pytorch/issues/59140,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: advanced indexing, module: error checking",scan when time allows,2021-06-07T17:11:42Z
issue,14839,Test OpenCV4 in CI,https://github.com/pytorch/pytorch/issues/14839,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2021-06-07T15:28:33Z
issue,14797,Modern interface for Storage,https://github.com/pytorch/pytorch/issues/14797,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: internals,scan when time allows,2021-06-07T15:25:49Z
issue,59386,Energy-efficiency benchmarks?,https://github.com/pytorch/pytorch/issues/59386,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-06-07T07:59:08Z
issue,12484,CuDNN convolution on some CUDA devices will not preserve NaN weights (upstream bug),https://github.com/pytorch/pytorch/issues/12484,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cudnn, module: dependency bug",scan when time allows,2021-06-07T02:15:20Z
issue,58997,[discussion] Should Optimizers be also Modules?,https://github.com/pytorch/pytorch/issues/58997,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: optimizer",scan when time allows,2021-06-06T11:31:22Z
issue,58973,First-class Java API,https://github.com/pytorch/pytorch/issues/58973,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: java,scan when time allows,2021-06-05T06:17:01Z
issue,59256,ProcessGroupGloo creation crashes when world_size > 150,https://github.com/pytorch/pytorch/issues/59256,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: c10d, module: data parallel, oncall: distributed",scan when time allows,2021-06-04T21:05:39Z
issue,59457,Error compiling PyTorch: `/usr/bin/ld: cannot find -lmagma /usr/bin/ld: cannot find -lnvToolsExt`,https://github.com/pytorch/pytorch/issues/59457,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cuda",scan when time allows,2021-06-04T18:09:52Z
issue,12532,Provide better documentation for torch.Size,https://github.com/pytorch/pytorch/issues/12532,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2021-06-04T14:51:35Z
issue,14702,pytorch_doc_push is racing with itself,https://github.com/pytorch/pytorch/issues/14702,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-06-04T13:41:49Z
issue,59418,Libtorch segfault in packed GRU evaluation with cuda batch_sizes,https://github.com/pytorch/pytorch/issues/59418,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: nn, module: rnn",scan when time allows,2021-06-04T13:39:36Z
issue,55056,[FR] Safe softmax,https://github.com/pytorch/pytorch/issues/55056,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2021-06-03T21:27:49Z
issue,59216,Add a BlendLoss class in torch.nn and a blend_loss function in torch.nn.functional to allow blending of a list of loss functions,https://github.com/pytorch/pytorch/issues/59216,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: loss, module: nn",scan when time allows,2021-06-02T16:23:33Z
issue,59058,Issue: support auto generation of device check for sparse tensors,https://github.com/pytorch/pytorch/issues/59058,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: codegen, module: sparse",scan when time allows,2021-06-02T15:37:01Z
issue,40988,'_mm256_extract_epi64' was not declared in this scope when compiling on Debian 32-bit,https://github.com/pytorch/pytorch/issues/40988,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2021-06-02T14:32:38Z
issue,59247,Support complex numbers in `at::nan_to_num`.,https://github.com/pytorch/pytorch/issues/59247,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: numpy,scan when time allows,2021-06-02T04:49:24Z
issue,58860,[cuDNN v8] Extend current cuDNN convolution v8 API binding to support conv-bias-activation fusion,https://github.com/pytorch/pytorch/issues/58860,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: cudnn",scan when time allows,2021-06-01T23:02:41Z
issue,58847,Dataloader Rerunning with num_workers=0 may give better error trace,https://github.com/pytorch/pytorch/issues/58847,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dataloader, module: ddp, oncall: distributed",scan when time allows,2021-06-01T22:03:02Z
issue,38639,Difference between using a python list and nn.ModuleList,https://github.com/pytorch/pytorch/issues/38639,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: nn",scan when time allows,2021-06-01T21:45:19Z
issue,42689,How can I specify NumPy while building PyTorch?,https://github.com/pytorch/pytorch/issues/42689,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2021-06-01T21:40:50Z
issue,14436,Use of STL templates in cpu/ directory (compiling with different AVX settings) is silently hazardous,https://github.com/pytorch/pytorch/issues/14436,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2021-06-01T21:22:21Z
issue,14467,"Optimizer warning when parameters ""change""",https://github.com/pytorch/pytorch/issues/14467,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2021-06-01T21:13:49Z
issue,14477,"Add a debug mode which is -O0 for framework code, but -O for kernels",https://github.com/pytorch/pytorch/issues/14477,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2021-06-01T21:12:49Z
issue,59251,Ahead of time C++ Extensions should have an option to build with verbose=False,https://github.com/pytorch/pytorch/issues/59251,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp-extensions,scan when time allows,2021-06-01T20:49:56Z
issue,59101,Use of storage_offset is not needed in add_out_dense_sparse_csr_cuda,https://github.com/pytorch/pytorch/issues/59101,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: sparse",scan when time allows,2021-06-01T14:03:35Z
issue,59071,test_bottleneck_cuda fails without error message,https://github.com/pytorch/pytorch/issues/59071,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: tests",scan when time allows,2021-06-01T13:56:13Z
issue,59104,Data access pattern in the loop in add_out_dense_sparse_csr_cuda could be pretty bad,https://github.com/pytorch/pytorch/issues/59104,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2021-06-01T03:06:53Z
issue,49445,[doc] how to prevent pytorch-nightly from being replaced by a released version on pip install,https://github.com/pytorch/pytorch/issues/49445,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: docs, oncall: releng",scan when time allows,2021-05-31T17:06:32Z
issue,44489,for_each doesn't support integer input to float output type promotion,https://github.com/pytorch/pytorch/issues/44489,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-05-31T01:07:51Z
issue,59185,Add supports_nnc metadata to OpInfos,https://github.com/pytorch/pytorch/issues/59185,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tests,scan when time allows,2021-05-30T08:04:24Z
issue,56896,masked_select is x3 slower than reshaping and index_select,https://github.com/pytorch/pytorch/issues/56896,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: advanced indexing, module: performance, module: viewing and reshaping",scan when time allows,2021-05-29T21:58:15Z
issue,58758,precision/consistency issue in `linspace`,https://github.com/pytorch/pytorch/issues/58758,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: numerical-reproducibility, module: numpy, module: tensor creation",scan when time allows,2021-05-29T07:41:35Z
issue,32706,Torch serialization does not restore tensors properly when custom __reduce__ is defined,https://github.com/pytorch/pytorch/issues/32706,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: serialization,scan when time allows,2021-05-27T23:14:06Z
issue,58596,Some cuda objects take very long time to build,https://github.com/pytorch/pytorch/issues/58596,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cuda",scan when time allows,2021-05-27T07:22:53Z
issue,51644,"Segmentation Fault: Garbage collector, cuda memory",https://github.com/pytorch/pytorch/issues/51644,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2021-05-26T22:55:04Z
issue,58765,[codegen] generated inplace/out= wrappers don't have input checks,https://github.com/pytorch/pytorch/issues/58765,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: codegen, module: error checking",scan when time allows,2021-05-26T17:59:44Z
issue,58962,Operating on a shared memory tensor with multiple threads hangs,https://github.com/pytorch/pytorch/issues/58962,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: multiprocessing, oncall: distributed",scan when time allows,2021-05-26T15:33:03Z
issue,58980,libtorch conflict with cxxopts and cause memery leak,https://github.com/pytorch/pytorch/issues/58980,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: memory usage,scan when time allows,2021-05-26T15:29:17Z
issue,54905,Allow CUDA build without requiring a physical GPU device.,https://github.com/pytorch/pytorch/issues/54905,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cuda",scan when time allows,2021-05-26T08:40:48Z
issue,58522,DDP grads dont have parity with local training when grads are undefined,https://github.com/pytorch/pytorch/issues/58522,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ddp, oncall: distributed",scan when time allows,2021-05-26T02:22:24Z
issue,58876,[deploy] Enable `torch.distributed.rpc` Python bindings in deploy,https://github.com/pytorch/pytorch/issues/58876,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: deploy, module: rpc, oncall: distributed",scan when time allows,2021-05-26T00:05:39Z
issue,4145,[Proposal] Consistent `batch_first` effect for RNN modules,https://github.com/pytorch/pytorch/issues/4145,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: nn, module: rnn",scan when time allows,2021-05-25T22:34:49Z
issue,58944,Use PYTHONHASHSEED during pytorch build to avoid nondeterminism,https://github.com/pytorch/pytorch/issues/58944,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: codegen, module: determinism",scan when time allows,2021-05-25T21:31:14Z
issue,58862,[cuDNN v8] Improve cuDNN convolution v8 API error reporting,https://github.com/pytorch/pytorch/issues/58862,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: cudnn",scan when time allows,2021-05-25T19:41:55Z
issue,58859,[cuDNN v8] Extend current cuDNN convolution v8 API binding to support cuDNN benchmark,https://github.com/pytorch/pytorch/issues/58859,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: cudnn",scan when time allows,2021-05-25T19:41:31Z
issue,58858,[cuDNN v8] Extend current cuDNN v8 API binding to support convolution backward and transposed convolution forward,https://github.com/pytorch/pytorch/issues/58858,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: cudnn",scan when time allows,2021-05-25T19:40:16Z
issue,44172,build from scratch failed because of the expansion of macro,https://github.com/pytorch/pytorch/issues/44172,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2021-05-25T10:07:55Z
issue,53891,Check Archive.org for Failed Dataset Downloads,https://github.com/pytorch/pytorch/issues/53891,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dependency bug,scan when time allows,2021-05-24T21:47:08Z
issue,49480,Internal symbols are leaking from torch.nn.functional,https://github.com/pytorch/pytorch/issues/49480,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-05-24T19:24:52Z
issue,58626,GPU 0 context created on GPU 1 worker when using pin_memory=True,https://github.com/pytorch/pytorch/issues/58626,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: dataloader",scan when time allows,2021-05-24T17:04:25Z
issue,58471,Why the same training code running on different GPUs (V100 vs P100) brings a significiant difference in outputs and thus train loss / model accuracy?,https://github.com/pytorch/pytorch/issues/58471,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: determinism,scan when time allows,2021-05-24T16:53:16Z
issue,58447,TSAN issues in at::RecordFunction,https://github.com/pytorch/pytorch/issues/58447,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: profiler,scan when time allows,2021-05-24T15:26:28Z
issue,50068,Magma functions that don't have queue argument create cublas handles for each call,https://github.com/pytorch/pytorch/issues/50068,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: linear algebra, module: performance",scan when time allows,2021-05-24T02:06:58Z
issue,19267,torch.distributed.broadcast should default to current stream,https://github.com/pytorch/pytorch/issues/19267,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2021-05-23T22:38:32Z
issue,58820,[docs] Tensor.bernoulli_ formatting is hard to read and UX inconsistent with the function variant,https://github.com/pytorch/pytorch/issues/58820,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ux,scan when time allows,2021-05-23T21:13:37Z
issue,58770,MKL csr matmul issue,https://github.com/pytorch/pytorch/issues/58770,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: mkl, module: sparse",scan when time allows,2021-05-23T18:14:52Z
issue,55355,simple matrix multiplication yields wrong result on Ampere (3080),https://github.com/pytorch/pytorch/issues/55355,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: cublas, module: tf32",scan when time allows,2021-05-23T05:51:14Z
issue,58657,"Improvement suggestions for the error message ""Expected batch2_sizes[0] == bs && batch2_sizes[1] == contraction_size to be true, but got false""",https://github.com/pytorch/pytorch/issues/58657,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: error checking, module: linear algebra",scan when time allows,2021-05-23T05:09:11Z
issue,58671,torch.load with dill is unable to unserialize from buffer,https://github.com/pytorch/pytorch/issues/58671,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: serialization,scan when time allows,2021-05-23T03:40:31Z
issue,58016,Eliminate potential double device check,https://github.com/pytorch/pytorch/issues/58016,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-05-22T18:54:43Z
issue,58735,CSR construction: safe_get_attr_string suppresses real errors,https://github.com/pytorch/pytorch/issues/58735,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2021-05-22T05:21:43Z
issue,58791,[nnc] Add more optimization for conditional stmts in loopnest transformations,https://github.com/pytorch/pytorch/issues/58791,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-05-22T03:38:05Z
issue,58617,[rfc] Bug developers about release notes for new APIs,https://github.com/pytorch/pytorch/issues/58617,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-05-20T21:26:53Z
issue,35499,Support tensor.cumsum() for 1-dim tensors,https://github.com/pytorch/pytorch/issues/35499,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: numpy,scan when time allows,2021-05-20T19:06:04Z
issue,58672,Werror should include -Wmissing-prototypes,https://github.com/pytorch/pytorch/issues/58672,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: build warnings",scan when time allows,2021-05-20T17:17:13Z
issue,58670,Ability to enabling/disabling cuDNN and cuBLAS API logging in PyTorch API directly,https://github.com/pytorch/pytorch/issues/58670,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cublas, module: cuda, module: cudnn",scan when time allows,2021-05-20T17:11:19Z
issue,58591,"link error on BatchNormImplBase<D, Derived>::pretty_print when using icpc",https://github.com/pytorch/pytorch/issues/58591,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: undefined reference,scan when time allows,2021-05-20T16:40:23Z
issue,58585,Calling benchmark.Timer with default `num_threads=1` disables parallelism permanently,https://github.com/pytorch/pytorch/issues/58585,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: macos, module: multithreading",scan when time allows,2021-05-20T16:33:44Z
issue,50375,Provide a set of C++ foreach APIs that will take tensor pointers as an input,https://github.com/pytorch/pytorch/issues/50375,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp,scan when time allows,2021-05-20T14:59:30Z
issue,52553,Does pytorch have the inverse activation function of nn.LeakyReLU?,https://github.com/pytorch/pytorch/issues/52553,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-05-20T12:29:17Z
issue,58489,Inter-category type promotion has undocumented behavior,https://github.com/pytorch/pytorch/issues/58489,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: type promotion",scan when time allows,2021-05-20T07:13:57Z
issue,58442,TSAN issues in autograd,https://github.com/pytorch/pytorch/issues/58442,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2021-05-20T05:56:13Z
issue,58474,nn.Upsample result mismatch in 1.1.0a0+828a6a3 and 1.9.0,https://github.com/pytorch/pytorch/issues/58474,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: cudnn",scan when time allows,2021-05-20T05:55:36Z
issue,23482,Build error due to unintended include path /usr/include,https://github.com/pytorch/pytorch/issues/23482,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2021-05-20T04:32:07Z
issue,21028,Zero-dim Tensors (scalars) should be printed at full precision,https://github.com/pytorch/pytorch/issues/21028,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: printing,scan when time allows,2021-05-19T18:14:23Z
issue,58395,[DOC] Add info about differentiability of functions,https://github.com/pytorch/pytorch/issues/58395,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2021-05-19T06:48:14Z
issue,58352,Broadcasting multiple tensors to all procs in distributed training,https://github.com/pytorch/pytorch/issues/58352,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2021-05-18T22:47:54Z
issue,56843,Add clipnorm parameter to optimizers,https://github.com/pytorch/pytorch/issues/56843,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2021-05-18T19:22:03Z
issue,58371,Tests do not pass even though build successful,https://github.com/pytorch/pytorch/issues/58371,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2021-05-18T18:15:56Z
issue,58212,`torch.autograd.Function` subclasses *sometimes* throw away custom subclasses,https://github.com/pytorch/pytorch/issues/58212,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2021-05-17T23:15:12Z
issue,58351,need sampled softmax loss function to train model to get user vector and item vector with a large dictionary,https://github.com/pytorch/pytorch/issues/58351,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: loss,scan when time allows,2021-05-17T20:23:52Z
issue,58343,Add the features like Keras,https://github.com/pytorch/pytorch/issues/58343,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: convolution,scan when time allows,2021-05-17T20:23:12Z
issue,58386,Better shared memory allocation under Docker,https://github.com/pytorch/pytorch/issues/58386,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docker, module: memory usage",scan when time allows,2021-05-17T17:41:49Z
issue,58385,"A ""sanitizer"" mode for the CUDA caching allocator?",https://github.com/pytorch/pytorch/issues/58385,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, oncall: profiler",scan when time allows,2021-05-17T17:27:50Z
issue,13818,test_spectral_norm: Backward is not reentrant,https://github.com/pytorch/pytorch/issues/13818,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: data parallel, module: norms and normalization",scan when time allows,2021-05-17T16:43:27Z
issue,13993,[tracking task] FBGEMM guarding AVX2 properly,https://github.com/pytorch/pytorch/issues/13993,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: third_party",scan when time allows,2021-05-17T16:36:53Z
issue,14047,Make torch.multiprocessing.SpawnContext usable,https://github.com/pytorch/pytorch/issues/14047,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: multiprocessing,scan when time allows,2021-05-17T16:35:12Z
issue,20424,pos_weight argument in torch.nn.BCELoss,https://github.com/pytorch/pytorch/issues/20424,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2021-05-17T04:34:56Z
issue,19529,failed to load model which is saved as text format(pickle_protocol=0) instead of binary format,https://github.com/pytorch/pytorch/issues/19529,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: pickle, module: serialization",scan when time allows,2021-05-14T20:57:05Z
issue,53903,[torch.vmap] Support second order grads,https://github.com/pytorch/pytorch/issues/53903,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: vmap",scan when time allows,2021-05-14T06:30:56Z
issue,47993,"build pytorch from source on ubuntu, building error from fbgemm::SparseAdaGradSignature",https://github.com/pytorch/pytorch/issues/47993,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: collect_env.py",scan when time allows,2021-05-14T02:45:31Z
issue,58238,torchdeploy doesn't prevent Obj from being used on wrong interpreter,https://github.com/pytorch/pytorch/issues/58238,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: deploy,scan when time allows,2021-05-13T23:32:09Z
issue,58218,"Torch.nn.DataParallel training model, the output of the model becomes list type, the number of lists is batch size",https://github.com/pytorch/pytorch/issues/58218,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: data parallel,scan when time allows,2021-05-13T21:42:37Z
issue,48439,cudnn convolution modifies the input Tensor metadata inplace when it tries to `.resize_()` it,https://github.com/pytorch/pytorch/issues/48439,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: cudnn, module: memory format",scan when time allows,2021-05-13T10:06:32Z
issue,56356,"Resolved: Only add type promotion support to unary pwise, binary pwise, and reduction operations",https://github.com/pytorch/pytorch/issues/56356,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: type promotion, module: ux",scan when time allows,2021-05-13T04:33:16Z
issue,58167,Allow `Union` as a `Dict` key,https://github.com/pytorch/pytorch/issues/58167,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-05-12T19:25:45Z
issue,58128,the same input with different batchsize got different precision output,https://github.com/pytorch/pytorch/issues/58128,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: batching, module: nn, module: numerical-reproducibility",scan when time allows,2021-05-12T18:55:49Z
issue,58136,scatter_add_ 6000-times slower with int64 compared to int32,https://github.com/pytorch/pytorch/issues/58136,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: docs, module: performance, module: scatter & gather ops",scan when time allows,2021-05-12T17:07:52Z
issue,58127,Save checkpoint error,https://github.com/pytorch/pytorch/issues/58127,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: optimizer, module: serialization",scan when time allows,2021-05-12T15:52:22Z
issue,52457,Add autograd tests to verify correctness for R -> C cases,https://github.com/pytorch/pytorch/issues/52457,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: complex, module: tests",scan when time allows,2021-05-12T14:08:24Z
issue,58037,linalg.eigh and linalg.cholesky UPLO flag breaks in backward,https://github.com/pytorch/pytorch/issues/58037,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: linear algebra",scan when time allows,2021-05-12T09:22:19Z
issue,58124,Python API binding code generation does not need to pack TensorOptions for `xxx_like` generators,https://github.com/pytorch/pytorch/issues/58124,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: codegen,scan when time allows,2021-05-12T09:18:19Z
issue,58091,Follow-up updates to inference mode python bindings,https://github.com/pytorch/pytorch/issues/58091,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-05-12T04:06:01Z
issue,13782,torch.linspace does not check for infinity and nan,https://github.com/pytorch/pytorch/issues/13782,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: error checking, module: tensor creation",scan when time allows,2021-05-12T03:58:20Z
issue,58087,TensorList upfront checks parameter types in argument parsing and throws an error unexpectedly,https://github.com/pytorch/pytorch/issues/58087,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: codegen,scan when time allows,2021-05-12T03:56:47Z
issue,42155,BusError memory-mapped tensor,https://github.com/pytorch/pytorch/issues/42155,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: internals,scan when time allows,2021-05-11T19:50:13Z
issue,57065,Static Linking Pytorch,https://github.com/pytorch/pytorch/issues/57065,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: static linking,scan when time allows,2021-05-11T06:20:37Z
issue,57960,Use the same CUDA stream for all RPCs within the same dist autograd context,https://github.com/pytorch/pytorch/issues/57960,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: rpc, oncall: distributed",scan when time allows,2021-05-10T23:19:34Z
issue,54941,Regression in Python arg parser performance.,https://github.com/pytorch/pytorch/issues/54941,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, module: performance",scan when time allows,2021-05-10T20:54:05Z
issue,57957,Add a few tests to make sure new dispatch keys for backends are added properly.,https://github.com/pytorch/pytorch/issues/57957,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dispatch,scan when time allows,2021-05-10T19:20:09Z
issue,57787,Add CI automation tests for Infiniband for DDP,https://github.com/pytorch/pytorch/issues/57787,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2021-05-10T16:44:49Z
issue,56460,"I meet an error assert key in deserialized_objects when I torch.load(pthname),and pth file is trained on multi gpu",https://github.com/pytorch/pytorch/issues/56460,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ddp, module: serialization, oncall: distributed",scan when time allows,2021-05-10T16:32:48Z
issue,57949,[structured] Clarify if NoNamesGuard is actually needed in impls or not,https://github.com/pytorch/pytorch/issues/57949,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: structured kernels,scan when time allows,2021-05-10T16:22:37Z
issue,54646,Optional enhanced logging for operator calls for backend implementor debugging,https://github.com/pytorch/pytorch/issues/54646,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: backend,scan when time allows,2021-05-10T13:04:41Z
issue,34953,Graphic tool to view the backward(Gradient Graph) and forward graph in Pytorch,https://github.com/pytorch/pytorch/issues/34953,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: visualization,scan when time allows,2021-05-10T11:24:27Z
issue,54232,DistributedDataParallel: `DDP(model)` hangs on non-master node,https://github.com/pytorch/pytorch/issues/54232,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ddp, oncall: distributed",scan when time allows,2021-05-10T09:00:32Z
issue,57617,complex128 autograd failures on PPC,https://github.com/pytorch/pytorch/issues/57617,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: complex, module: power",scan when time allows,2021-05-10T05:24:37Z
issue,57921,Backward pass for a nn.Conv2d with half-precision on Quadro RTX 8000 leads to CUDNN_STATUS_INTERNAL_ERROR,https://github.com/pytorch/pytorch/issues/57921,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cudnn,scan when time allows,2021-05-10T04:32:03Z
issue,57843,Torch.save doesn't make new file,https://github.com/pytorch/pytorch/issues/57843,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: serialization,scan when time allows,2021-05-10T04:14:58Z
issue,57646,"get_future() should be documented once it is enabled for gloo, mpi backends",https://github.com/pytorch/pytorch/issues/57646,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2021-05-07T22:45:16Z
issue,13508,OpenCL support for smartphones,https://github.com/pytorch/pytorch/issues/13508,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-05-07T22:42:54Z
issue,13268,Non-Zero Padding in Convolution Module,https://github.com/pytorch/pytorch/issues/13268,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: nn",scan when time allows,2021-05-07T22:42:29Z
issue,56583,"All thread stack trace dumps interleave output in distributed tests, giving garbled output",https://github.com/pytorch/pytorch/issues/56583,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2021-05-07T22:33:46Z
issue,56772,The multi-fc losses calculating in DistributedDataParallel.,https://github.com/pytorch/pytorch/issues/56772,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2021-05-07T18:46:16Z
issue,57631,Pruners' compute_mask returns tensors with dtypes that are not consistent with each other,https://github.com/pytorch/pytorch/issues/57631,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: pruning",scan when time allows,2021-05-07T17:42:08Z
issue,57712,[Docs] nn.init._calculate_fan_in_and_fan_out(),https://github.com/pytorch/pytorch/issues/57712,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2021-05-07T12:20:44Z
issue,32322,deadlock when using mp.spawn multiprocessing,https://github.com/pytorch/pytorch/issues/32322,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: multiprocessing,scan when time allows,2021-05-07T04:10:49Z
issue,55981,"LazyEmbedding, an embedding layer with a dynamically sized vocabulary",https://github.com/pytorch/pytorch/issues/55981,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: embedding, module: nn",scan when time allows,2021-05-07T01:10:34Z
issue,57759,Show commit hash for master branch docs on pytorch.org,https://github.com/pytorch/pytorch/issues/57759,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: doc infra,scan when time allows,2021-05-06T19:04:20Z
issue,57611,[perf] 10x improvement on element-wise operations with manual broadcast,https://github.com/pytorch/pytorch/issues/57611,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: performance, module: tensoriterator",scan when time allows,2021-05-06T18:17:51Z
issue,46653,Future returned by RPC should print a warning message on destruction if it's not waited,https://github.com/pytorch/pytorch/issues/46653,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, module: rpc, oncall: distributed",scan when time allows,2021-05-06T18:17:25Z
issue,57742,[structured] Meta-impl split implies redundant dtype tests,https://github.com/pytorch/pytorch/issues/57742,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-05-06T18:04:05Z
issue,56698,Convolution2D may have a numerical error.,https://github.com/pytorch/pytorch/issues/56698,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: numerical-stability,scan when time allows,2021-05-06T09:27:55Z
issue,57547,Add distributed testing for CUDA aware MPI,https://github.com/pytorch/pytorch/issues/57547,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: c10d, oncall: distributed",scan when time allows,2021-05-06T07:17:12Z
issue,57691,Unify `std::getenv` usages,https://github.com/pytorch/pytorch/issues/57691,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-05-05T23:04:18Z
issue,55661,"RuntimeError: ""max_cuda"" not implemented for 'ComplexFloat'",https://github.com/pytorch/pytorch/issues/55661,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: complex,scan when time allows,2021-05-05T21:31:54Z
issue,57440,"I have just wasted 3 solid days trying to build PyTorch, without success",https://github.com/pytorch/pytorch/issues/57440,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2021-05-05T03:11:18Z
issue,57533,test_grid_sample (from TestNN) fails on POWER,https://github.com/pytorch/pytorch/issues/57533,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: power, module: tests",scan when time allows,2021-05-05T03:07:21Z
issue,57415,torch 1.8.1+cuda crashes when setting breakpoints.,https://github.com/pytorch/pytorch/issues/57415,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-05-05T01:17:35Z
issue,57503,[new feature] Adaptive gradient clipping,https://github.com/pytorch/pytorch/issues/57503,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: nn",scan when time allows,2021-05-04T13:52:21Z
issue,57495,Please remove this assertion - it triggers on valid use cases,https://github.com/pytorch/pytorch/issues/57495,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-05-04T02:06:12Z
issue,57502,[structured] Fail if CompositeExplicitAutograd is in dispatch table of structured_delegate function,https://github.com/pytorch/pytorch/issues/57502,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2021-05-04T02:05:57Z
issue,57507,Cleanup usage of IS_PYTHON_3_9_PLUS in autograd engine,https://github.com/pytorch/pytorch/issues/57507,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: pybind",scan when time allows,2021-05-03T22:18:09Z
issue,57475,CUDA RPC stream synchronization does not work with @rpc.functions.async_execution,https://github.com/pytorch/pytorch/issues/57475,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: rpc, oncall: distributed",scan when time allows,2021-05-03T17:39:38Z
issue,57370,gen_proto_typestubs_helper.py is slow,https://github.com/pytorch/pytorch/issues/57370,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2021-05-03T17:11:10Z
issue,13053,Generalized Data Class,https://github.com/pytorch/pytorch/issues/13053,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: data, module: dataloader",scan when time allows,2021-05-03T17:04:24Z
issue,57380,Asserts all tensor are defined in dispatch wrapper,https://github.com/pytorch/pytorch/issues/57380,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-05-03T16:08:18Z
issue,57392,Simplified API for custom inplace & view kernel,https://github.com/pytorch/pytorch/issues/57392,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: custom-operators, module: dispatch",scan when time allows,2021-05-03T15:03:26Z
issue,57369,Delay errors of inference tensor to backward pass?,https://github.com/pytorch/pytorch/issues/57369,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dispatch,scan when time allows,2021-05-03T15:01:02Z
issue,57358,Missing complex autograd support for some operators,https://github.com/pytorch/pytorch/issues/57358,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: complex,scan when time allows,2021-05-03T14:58:28Z
issue,57326,Renaming Autograd dispatch keys to ADCreateGraph,https://github.com/pytorch/pytorch/issues/57326,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dispatch,scan when time allows,2021-05-03T14:57:39Z
issue,57359,torch.return_types does not exist,https://github.com/pytorch/pytorch/issues/57359,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-05-03T14:57:03Z
issue,55736,RuntimeError: CUDA error: CUBLAS_STATUS_INTERNAL_ERROR when calling `cublasCreate(handle)`,https://github.com/pytorch/pytorch/issues/55736,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2021-05-03T12:32:37Z
issue,53473,RuntimeError: CUDA error: CUBLAS_STATUS_INTERNAL_ERROR when calling `cublasCreate(handle)`,https://github.com/pytorch/pytorch/issues/53473,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cublas,scan when time allows,2021-05-03T12:32:33Z
issue,57018,Attempting to concatenate scalar tensors throws a runtime error,https://github.com/pytorch/pytorch/issues/57018,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: numpy,scan when time allows,2021-05-02T14:19:05Z
issue,56975,scatter does not accept scalar src=,https://github.com/pytorch/pytorch/issues/56975,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: numpy, module: scatter & gather ops",scan when time allows,2021-05-02T04:52:15Z
issue,56329,Change `other` to `src` in torch.Tensor.scatter_add_,https://github.com/pytorch/pytorch/issues/56329,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2021-05-01T22:23:52Z
issue,56979,"apex-0.1-py3.6-linux-x86_64.egg/apex/amp/wrap.py"", line 28, in wrapper return orig_fn(*new_args, **kwargs) RuntimeError: CUDA error: an illegal memory access was encountered",https://github.com/pytorch/pytorch/issues/56979,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision), module: cuda",scan when time allows,2021-04-30T02:42:43Z
issue,3356,Data sampling seems to be more complicated than necessary,https://github.com/pytorch/pytorch/issues/3356,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2021-04-29T22:26:59Z
issue,13207,[pytorch] [feature request] Error out if the needed GPU device capability is absent in runtime,https://github.com/pytorch/pytorch/issues/13207,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: error checking",scan when time allows,2021-04-29T16:25:12Z
issue,57108,Compilation failure in caffe2/core/plan_executor_test.cc with GCC 5.4,https://github.com/pytorch/pytorch/issues/57108,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2021-04-29T03:23:40Z
issue,44814,LNK2019 error after static compilation of Libtorch,https://github.com/pytorch/pytorch/issues/44814,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2021-04-28T22:09:37Z
issue,57095,[FR] unflatten support empty shape if unflattened dim has size 1,https://github.com/pytorch/pytorch/issues/57095,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: viewing and reshaping,scan when time allows,2021-04-28T17:44:02Z
issue,12962,Come up with a better strategy for noticing BC-breaking attribute additions to serializable classes,https://github.com/pytorch/pytorch/issues/12962,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bc-breaking, module: molly-guard",scan when time allows,2021-04-28T15:13:02Z
issue,12980,Installing pytorch from source on labs.cognitiveclass.ai,https://github.com/pytorch/pytorch/issues/12980,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: power,scan when time allows,2021-04-28T15:12:19Z
issue,13162,circular module reference raises RecursionError,https://github.com/pytorch/pytorch/issues/13162,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2021-04-28T15:03:42Z
issue,13188,cudnn explicit paths and GCC multilib suffixes prevents detection of good cudnn headers,https://github.com/pytorch/pytorch/issues/13188,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2021-04-28T15:00:19Z
issue,13034,[caffe2]How can I export init_net.pb and predict_net.pb files on my own?,https://github.com/pytorch/pytorch/issues/13034,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-04-28T14:57:56Z
issue,57059,Get rid of weak_intrusive_ptr,https://github.com/pytorch/pytorch/issues/57059,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: internals,scan when time allows,2021-04-28T14:52:02Z
issue,57112,:attr: does not behave as expected.,https://github.com/pytorch/pytorch/issues/57112,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: linear algebra",scan when time allows,2021-04-28T14:33:06Z
issue,57123,LibTorch ships with two identical library files: libdnnl and libmlkdnn,https://github.com/pytorch/pytorch/issues/57123,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2021-04-28T14:26:13Z
issue,54906,Out variants for Convolution ops,https://github.com/pytorch/pytorch/issues/54906,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: convolution,scan when time allows,2021-04-28T05:07:09Z
issue,56777,Error when building PyTorch from source in CentOS Linux 7,https://github.com/pytorch/pytorch/issues/56777,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: tensorpipe",scan when time allows,2021-04-28T02:34:23Z
issue,56881,[FR] integral discrete distributions should support dtype=int64,https://github.com/pytorch/pytorch/issues/56881,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distributions,scan when time allows,2021-04-27T23:01:23Z
issue,56809,Autograd engine worker thread initialization fails with python 3.9 debug build,https://github.com/pytorch/pytorch/issues/56809,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-04-27T22:58:37Z
issue,56921,torch.multiprocessing implement SyncManager,https://github.com/pytorch/pytorch/issues/56921,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: multiprocessing,scan when time allows,2021-04-27T22:57:30Z
issue,57005,about torch install,https://github.com/pytorch/pytorch/issues/57005,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-04-27T22:44:58Z
issue,56671,{h / v / d}split methods are missing,https://github.com/pytorch/pytorch/issues/56671,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: numpy,scan when time allows,2021-04-27T18:46:16Z
issue,54499,Caught an integer overflow or wraparound in torch.nn.MaxPool1d,https://github.com/pytorch/pytorch/issues/54499,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: pooling,scan when time allows,2021-04-27T17:59:06Z
issue,27343,[jit] `ScriptFunction`s are loaded as `ScriptModule`s,https://github.com/pytorch/pytorch/issues/27343,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2021-04-27T17:35:30Z
issue,56464,Scatter tests missing when passing a reduction operation.,https://github.com/pytorch/pytorch/issues/56464,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: scatter & gather ops,scan when time allows,2021-04-27T12:13:27Z
issue,54282,test_randperm is failing on CPU-only build,https://github.com/pytorch/pytorch/issues/54282,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tests,scan when time allows,2021-04-27T02:55:48Z
issue,56694,PyTorch 1.8.0 / 1.8.1 CUDA memory corruption on oddly sized tesnor sections of strided tensor,https://github.com/pytorch/pytorch/issues/56694,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2021-04-26T19:50:53Z
issue,56480,shared torch.tensor with multiprocesses using python Queue cause coredump,https://github.com/pytorch/pytorch/issues/56480,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: multiprocessing,scan when time allows,2021-04-26T15:34:12Z
issue,34257,CustomFromMask pruning stores a copy of the user-provided mask,https://github.com/pytorch/pytorch/issues/34257,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: pruning",scan when time allows,2021-04-26T12:07:14Z
issue,18536,Make it easier to figure out what CuDNN convolution algorithm we actually chose,https://github.com/pytorch/pytorch/issues/18536,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cudnn, module: logging",scan when time allows,2021-04-24T00:45:57Z
issue,51896,Can't reassign a child module instance to a lambda,https://github.com/pytorch/pytorch/issues/51896,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2021-04-23T19:01:13Z
issue,56805,Implement random SeedSequence,https://github.com/pytorch/pytorch/issues/56805,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: random,scan when time allows,2021-04-23T18:55:13Z
issue,12895,RelaxedOneHotCategorical not implementing entropy (and other abstract methods),https://github.com/pytorch/pytorch/issues/12895,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distributions,scan when time allows,2021-04-23T18:41:24Z
issue,56781,Named Tensors in C++ Is Undocumented,https://github.com/pytorch/pytorch/issues/56781,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: named tensor",scan when time allows,2021-04-23T15:37:48Z
issue,56764,Change make_reduction to reflect input resizing.,https://github.com/pytorch/pytorch/issues/56764,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: reductions,scan when time allows,2021-04-23T15:31:13Z
issue,47650,DataParallel support for scripted modules in C++,https://github.com/pytorch/pytorch/issues/47650,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2021-04-23T13:51:10Z
issue,15716,"Nn.dataparallel with multiple output, weird gradient result None",https://github.com/pytorch/pytorch/issues/15716,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2021-04-23T10:07:03Z
issue,56483,Add compute_residuals flag for torch.linalg.lstsq,https://github.com/pytorch/pytorch/issues/56483,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: linear algebra, module: numpy",scan when time allows,2021-04-23T05:48:34Z
issue,56710,Hard fail build when there is no CUDA but `USE_CUDA=1`,https://github.com/pytorch/pytorch/issues/56710,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2021-04-23T04:47:37Z
issue,56691,[structured] at::cpu_unchecked functions,https://github.com/pytorch/pytorch/issues/56691,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: structured kernels,scan when time allows,2021-04-23T04:41:28Z
issue,56542,CrossEntropyLoss target shape expectation is inconsistent with BCEWithLogits,https://github.com/pytorch/pytorch/issues/56542,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: shape checking,scan when time allows,2021-04-23T04:27:59Z
issue,56525,Can't deepcopy memory format objects,https://github.com/pytorch/pytorch/issues/56525,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: memory format,scan when time allows,2021-04-23T04:27:13Z
issue,34880,Expanded tensors don't work as registered buffers,https://github.com/pytorch/pytorch/issues/34880,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: error checking, module: nn, module: partial aliasing",scan when time allows,2021-04-22T16:48:14Z
issue,30246,CUDA masked_select uses way too much memory,https://github.com/pytorch/pytorch/issues/30246,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: performance",scan when time allows,2021-04-22T16:38:45Z
issue,18730,Numerical instability KL divergence RelaxedOneHotCategorical,https://github.com/pytorch/pytorch/issues/18730,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: distributions, module: numerical-stability",scan when time allows,2021-04-22T14:09:10Z
issue,56634,[package] Module name reported in error message does not always match what is needed to extern/mock it,https://github.com/pytorch/pytorch/issues/56634,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-04-21T21:43:22Z
issue,49666,Use rlibm for faster and more accurate floating point operations,https://github.com/pytorch/pytorch/issues/49666,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: bfloat16,scan when time allows,2021-04-21T20:41:24Z
issue,56300,torch.where with input Tensor and other Scalar raises type mismatch error,https://github.com/pytorch/pytorch/issues/56300,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: type promotion,scan when time allows,2021-04-21T17:28:18Z
issue,56370,"Acquiring ""is_grad_enabled()"" inside an autograd function",https://github.com/pytorch/pytorch/issues/56370,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: custom-operators",scan when time allows,2021-04-21T02:52:51Z
issue,56465,Provide a pkg-config file,https://github.com/pytorch/pytorch/issues/56465,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2021-04-20T19:56:05Z
issue,56482,unsafe_reclaim_from_nonowning is not that unsafe,https://github.com/pytorch/pytorch/issues/56482,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, module: internals",scan when time allows,2021-04-20T19:46:54Z
issue,12715,pack_padded_sequence throws IndexError when only kwargs are specified,https://github.com/pytorch/pytorch/issues/12715,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rnn,scan when time allows,2021-04-20T18:19:28Z
issue,12716,cdf in torch.distributions.bernoulli throws NotImplementedError,https://github.com/pytorch/pytorch/issues/12716,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distributions,scan when time allows,2021-04-20T18:18:47Z
issue,12797,Port dragon4_scientific for pretty float tensor print.,https://github.com/pytorch/pytorch/issues/12797,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: printing,scan when time allows,2021-04-20T18:18:18Z
issue,12812,Pytorch BatchNorm2D Unstable,https://github.com/pytorch/pytorch/issues/12812,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: norms and normalization,scan when time allows,2021-04-20T18:17:33Z
issue,12828,BUILD_BINARY is a lie,https://github.com/pytorch/pytorch/issues/12828,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2021-04-20T18:15:41Z
issue,12830,Test that (cd build && ninja) immediately after build is no-op in CI,https://github.com/pytorch/pytorch/issues/12830,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: ci",scan when time allows,2021-04-20T18:14:41Z
issue,56183,"scripting LSTM module generates a script that has no .code, no .graph, and cannot be executed.",https://github.com/pytorch/pytorch/issues/56183,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2021-04-20T17:46:03Z
issue,12868,[CAPI] Increase of memory usage when exporting a Adam optimzer,https://github.com/pytorch/pytorch/issues/12868,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: memory usage, module: optimizer",scan when time allows,2021-04-20T15:55:42Z
issue,54655,qsize not implemented error,https://github.com/pytorch/pytorch/issues/54655,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: macos, module: multiprocessing",scan when time allows,2021-04-20T01:54:21Z
issue,56100,[package] implicit externing can miss cases where the stdlib depends on further stdlib modules,https://github.com/pytorch/pytorch/issues/56100,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-04-20T01:45:42Z
issue,56399,Figure out what mojo FB common/process/StackTrace.h has that we don't,https://github.com/pytorch/pytorch/issues/56399,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: internals,scan when time allows,2021-04-19T20:06:55Z
issue,56350,logcumsumexp (and maybe other cum* ops) has divergent CUDA and CPU out= behavior,https://github.com/pytorch/pytorch/issues/56350,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: safe resize",scan when time allows,2021-04-19T18:23:44Z
issue,56330,Type checking doesn't match actual type in error message,https://github.com/pytorch/pytorch/issues/56330,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: error checking,scan when time allows,2021-04-19T15:24:43Z
issue,56328,Why does output_padding have constraints in ConvTranspose1d?,https://github.com/pytorch/pytorch/issues/56328,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2021-04-19T15:22:07Z
issue,56297,Audit destructors of classes bound using pybind11::class_ to see if they can block; such cases can deadlock,https://github.com/pytorch/pytorch/issues/56297,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: pybind,scan when time allows,2021-04-19T14:37:39Z
issue,56357,[NNC] All NNC APIs should return bool flag,https://github.com/pytorch/pytorch/issues/56357,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-04-19T06:58:40Z
issue,24401,"When running model forward with large batch size, it reports the error: THCudaTensor sizes too large for THCDeviceTensor conversion",https://github.com/pytorch/pytorch/issues/24401,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2021-04-18T22:38:17Z
issue,27336,Allow one inferred axis in torch.split,https://github.com/pytorch/pytorch/issues/27336,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-04-18T08:18:55Z
issue,11213,Ubuntu 16.04 setup.py error - undefined reference to elfLink_Get_FatBinary_From_Object' /usr/lib/x86_64-linux-gnu/libcuda.so: undefined reference to elf32_section_header',https://github.com/pytorch/pytorch/issues/11213,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: undefined reference",scan when time allows,2021-04-17T09:33:09Z
issue,54871,Not possible to save dataloader in C++,https://github.com/pytorch/pytorch/issues/54871,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2021-04-17T09:31:50Z
issue,56237,optim.Adadelta: local variable 'lr' referenced before assignment,https://github.com/pytorch/pytorch/issues/56237,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2021-04-17T09:13:31Z
issue,12672,Move collate_fn functionality / responsibility into Dataset object,https://github.com/pytorch/pytorch/issues/12672,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2021-04-16T20:17:02Z
issue,12690,How can I build caffe2_gtest_main under pytorch/caffe2/test/ folder?,https://github.com/pytorch/pytorch/issues/12690,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-04-16T20:14:06Z
issue,12675,[feature request] ignore_index and size_average in nn.AdaptiveLogSoftMaxWithLoss,https://github.com/pytorch/pytorch/issues/12675,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2021-04-16T20:07:23Z
issue,55933,[NNC] Replace ComputeAt,https://github.com/pytorch/pytorch/issues/55933,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-04-16T20:01:17Z
issue,56285,qnnpack uses deprecated pthreadpool APIs,https://github.com/pytorch/pytorch/issues/56285,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-04-16T18:36:48Z
issue,56030,Add Dirichlet Multinomial to PyTorch Distributions,https://github.com/pytorch/pytorch/issues/56030,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distributions,scan when time allows,2021-04-16T18:01:27Z
issue,55757,Synchronize RRef.to_here() CUDA Streams properly when the profiler is enabled,https://github.com/pytorch/pytorch/issues/55757,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rpc,scan when time allows,2021-04-16T17:08:38Z
issue,55941,[Tracking] Remove unneeded BC duplicates from native_functions.yaml,https://github.com/pytorch/pytorch/issues/55941,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-04-16T16:02:56Z
issue,56263,Reflect padding_mode should be supported for Conv3d,https://github.com/pytorch/pytorch/issues/56263,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: docs",scan when time allows,2021-04-16T13:15:14Z
issue,56246,Silent incorrect running with zero padding for Conv1d,https://github.com/pytorch/pytorch/issues/56246,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: error checking, module: padding",scan when time allows,2021-04-16T13:05:17Z
issue,56231,Half precision support for torch.sparse.mm,https://github.com/pytorch/pytorch/issues/56231,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: half,scan when time allows,2021-04-16T12:55:36Z
issue,56224,Error message regarding Padding of Conv2d needs improving,https://github.com/pytorch/pytorch/issues/56224,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: padding",scan when time allows,2021-04-16T12:54:03Z
issue,56177,torch.gradient not throwing error when spacing to be equal to 0,https://github.com/pytorch/pytorch/issues/56177,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2021-04-16T12:46:05Z
issue,27094,[jit] Document what types can be traced,https://github.com/pytorch/pytorch/issues/27094,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2021-04-15T21:19:33Z
issue,26744,[jit] Default args don't work with TorchScript classes,https://github.com/pytorch/pytorch/issues/26744,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2021-04-15T21:19:27Z
issue,52694,[TensorExpr] Loop Transformations: `compute_inline` should use `Let` for inlined expressions,https://github.com/pytorch/pytorch/issues/52694,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-04-15T20:40:50Z
issue,29378,Support out= parameters with autograd,https://github.com/pytorch/pytorch/issues/29378,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: nn",scan when time allows,2021-04-15T17:00:06Z
issue,56090,Converting float16->bool causes an internal compiler error with llvm,https://github.com/pytorch/pytorch/issues/56090,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-04-15T16:54:08Z
issue,50332,torch.Tensor.random_ is divergent from NumPy's np.random.random,https://github.com/pytorch/pytorch/issues/50332,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: deprecation, module: numpy, module: random",scan when time allows,2021-04-15T16:34:55Z
issue,39443,PyTorch multiprocessing.spawn seems slow with list of tensors,https://github.com/pytorch/pytorch/issues/39443,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: multiprocessing, module: performance",scan when time allows,2021-04-15T11:01:27Z
issue,56064,Optionally include padding_idx items in the EmbeddingBag reduction,https://github.com/pytorch/pytorch/issues/56064,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: embedding, module: nn",scan when time allows,2021-04-15T10:29:05Z
issue,56067,A more flexible torch.hub search strategy,https://github.com/pytorch/pytorch/issues/56067,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: hub,scan when time allows,2021-04-15T10:28:24Z
issue,12248,.cuda() changes a module's behavior when there are registered buffers with requires_grad=True,https://github.com/pytorch/pytorch/issues/12248,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: nn",scan when time allows,2021-04-14T19:56:35Z
issue,56012,Failure in complex CUDA numerics tests for sigmoid,https://github.com/pytorch/pytorch/issues/56012,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: complex,scan when time allows,2021-04-14T17:55:06Z
issue,56023,[Opinfo] Better ErrorMsg for test_out with wrong shape,https://github.com/pytorch/pytorch/issues/56023,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tests,scan when time allows,2021-04-14T17:48:28Z
issue,52310,Implementation of many complex functions is fast but inaccurate in libc++,https://github.com/pytorch/pytorch/issues/52310,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: complex, module: numerical-stability, module: numpy",scan when time allows,2021-04-14T15:48:51Z
issue,55756,Cuda OOM after several steps for DataParallel model,https://github.com/pytorch/pytorch/issues/55756,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: memory usage",scan when time allows,2021-04-14T08:39:06Z
issue,55356,torch.allclose does not allow different types for comparison,https://github.com/pytorch/pytorch/issues/55356,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: numpy, module: type promotion",scan when time allows,2021-04-14T07:51:22Z
issue,55869,We should perhaps prevent linking against LibTorch DLLs with the wrong configuration (debug vs release),https://github.com/pytorch/pytorch/issues/55869,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: windows,scan when time allows,2021-04-14T07:16:39Z
issue,52206,from_numpy function rounds float values on Jetson NX,https://github.com/pytorch/pytorch/issues/52206,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: arm, module: correctness (silent), module: jetson",scan when time allows,2021-04-14T06:30:24Z
issue,55944,Sparse-sparse matrix multiplication only works with torch.sparse.mm(),https://github.com/pytorch/pytorch/issues/55944,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: sparse, module: ux",scan when time allows,2021-04-14T05:44:16Z
issue,55804,Add torch.nn.Conv2D correctness test,https://github.com/pytorch/pytorch/issues/55804,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: tests",scan when time allows,2021-04-14T04:59:10Z
issue,54800,[package] Detection of dependencies introduced by torch.ops.load_library,https://github.com/pytorch/pytorch/issues/54800,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-04-14T00:34:12Z
issue,55784,pybind11 Tensor type caster forces reference count bump,https://github.com/pytorch/pytorch/issues/55784,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: pybind,scan when time allows,2021-04-14T00:23:47Z
issue,55643,at::globalContext().hasCUDA()'s forked tongue tells lies,https://github.com/pytorch/pytorch/issues/55643,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: multiprocessing",scan when time allows,2021-04-13T22:27:47Z
issue,55964,Finish deprecating torch.range,https://github.com/pytorch/pytorch/issues/55964,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: deprecation,scan when time allows,2021-04-13T22:17:40Z
issue,55453,C++ sparse_coo_tensor ignores TensorOptions argument,https://github.com/pytorch/pytorch/issues/55453,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: sparse, module: tensor creation",scan when time allows,2021-04-13T21:50:12Z
issue,55876,[rfc] Trigger callback when backwards begins for DDP with custom autograd function,https://github.com/pytorch/pytorch/issues/55876,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ddp, oncall: distributed",scan when time allows,2021-04-13T20:43:27Z
issue,55951,Update linspace and logspace to throw an error when steps is not provided,https://github.com/pytorch/pytorch/issues/55951,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: deprecation, module: tensor creation",scan when time allows,2021-04-13T20:40:43Z
issue,55095,Python dispatch key and backend fallback,https://github.com/pytorch/pytorch/issues/55095,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: __torch_function__,scan when time allows,2021-04-13T19:17:20Z
issue,54622,torch.pow returns incorrect value for 0^0j,https://github.com/pytorch/pytorch/issues/54622,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: complex, module: numpy",scan when time allows,2021-04-13T17:47:13Z
issue,55331,Tests should be runnable without run_test.py,https://github.com/pytorch/pytorch/issues/55331,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: testing,scan when time allows,2021-04-13T15:24:28Z
issue,12609,Request for stripped down / inference only pytorch wheels,https://github.com/pytorch/pytorch/issues/12609,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2021-04-13T14:49:15Z
issue,41177,The new version of the libtorch become slow,https://github.com/pytorch/pytorch/issues/41177,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: performance, oncall: jit",scan when time allows,2021-04-13T12:10:43Z
issue,12460,"input_device, output_device, devices_used properties",https://github.com/pytorch/pytorch/issues/12460,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2021-04-13T01:23:27Z
issue,12482,[Feature request]: add `LayerNormLSTMCell`,https://github.com/pytorch/pytorch/issues/12482,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rnn,scan when time allows,2021-04-13T01:12:08Z
issue,55751,[docs] algolia docsearch in pytorch docs,https://github.com/pytorch/pytorch/issues/55751,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: doc infra,scan when time allows,2021-04-13T00:58:00Z
issue,55764,clang format on OS X ssl verification failure,https://github.com/pytorch/pytorch/issues/55764,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: lint,scan when time allows,2021-04-13T00:55:30Z
issue,55450,[DOC] Incorrect image for the confusion matrix,https://github.com/pytorch/pytorch/issues/55450,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: rnn",scan when time allows,2021-04-12T23:27:12Z
issue,55286,endpoint=False for torch.linspace and torch.logspace,https://github.com/pytorch/pytorch/issues/55286,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: numpy, module: tensor creation",scan when time allows,2021-04-12T23:22:42Z
issue,55755,"Some float16 inputs to CPU matmul are supported, others aren't",https://github.com/pytorch/pytorch/issues/55755,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: half, module: linear algebra",scan when time allows,2021-04-12T07:27:44Z
issue,54588,Make searchsorted and bucketize API consistent,https://github.com/pytorch/pytorch/issues/54588,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: deprecation, module: docs, module: numpy, module: ux",scan when time allows,2021-04-11T08:32:16Z
issue,54766,Add `start_dim` and `end_dim` functionality for common reduction operations.,https://github.com/pytorch/pytorch/issues/54766,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: reductions,scan when time allows,2021-04-09T21:12:28Z
issue,55629,torch.arctanh not implemented for torch.float16,https://github.com/pytorch/pytorch/issues/55629,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: half,scan when time allows,2021-04-09T19:45:40Z
issue,55696,[TensorExpr] IRSimplifier fails to optimize useless stores,https://github.com/pytorch/pytorch/issues/55696,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-04-09T19:31:49Z
issue,55611,ATen/native/sparse headers not in LibTorch or pypi package,https://github.com/pytorch/pytorch/issues/55611,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: binaries,scan when time allows,2021-04-08T22:51:19Z
issue,55585,"[amp scaler] unable to prevent ""scheduler before optimizer step"" warning",https://github.com/pytorch/pytorch/issues/55585,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision), module: optimizer",scan when time allows,2021-04-08T22:49:57Z
issue,55583,Improve test runtime of distributed tests.,https://github.com/pytorch/pytorch/issues/55583,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: tests, oncall: distributed",scan when time allows,2021-04-08T20:59:18Z
issue,51933,Performance bugs of transpose2d on A100,https://github.com/pytorch/pytorch/issues/51933,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cudnn, module: performance",scan when time allows,2021-04-08T17:55:43Z
issue,55557,Gradcheck failure for `torchaudio.functional.phase_vocoder`,https://github.com/pytorch/pytorch/issues/55557,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: correctness (silent)",scan when time allows,2021-04-08T14:51:00Z
issue,35290,Why C++ version libtorch so slow,https://github.com/pytorch/pytorch/issues/35290,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2021-04-08T14:33:14Z
issue,55571,Update TCPStore.wait() error message to be more friendly,https://github.com/pytorch/pytorch/issues/55571,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2021-04-07T23:32:39Z
issue,55440,Have the ability to disable `__torch_function__` dispatch for torch.nn.functional functions,https://github.com/pytorch/pytorch/issues/55440,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: __torch_function__, module: nn",scan when time allows,2021-04-07T20:03:52Z
issue,53263,Different gradients in torch.matmul depending on input shape,https://github.com/pytorch/pytorch/issues/53263,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: numerical-stability,scan when time allows,2021-04-07T19:47:14Z
issue,4958,[Feature Request] Extract glimpses from a batch of images (as in tf.image.extract_glimpse),https://github.com/pytorch/pytorch/issues/4958,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: vision,scan when time allows,2021-04-07T18:35:51Z
issue,55329,Split test/cpp_extensions/setup.py per orthogonal extension module and/or do Python build from ninja too,https://github.com/pytorch/pytorch/issues/55329,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cpp-extensions",scan when time allows,2021-04-07T16:29:53Z
issue,55330,CI not surfacing some failures (-Werror?) on PR,https://github.com/pytorch/pytorch/issues/55330,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2021-04-07T16:19:39Z
issue,55448,"Errors occur when operating broadcast with error log: Got completion with error 12, opcode 1, len 32547, vendor err 129",https://github.com/pytorch/pytorch/issues/55448,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2021-04-07T16:13:57Z
issue,55461,Cross-compiling for cortexa9 processor,https://github.com/pytorch/pytorch/issues/55461,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2021-04-07T16:03:48Z
issue,55304,nn.DataParallel should raise error when provided with list of tensors,https://github.com/pytorch/pytorch/issues/55304,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: data parallel,scan when time allows,2021-04-07T14:59:38Z
issue,55297,SIGSEGV at at::is_vulkan_available() invocation on Android,https://github.com/pytorch/pytorch/issues/55297,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: android,scan when time allows,2021-04-07T14:57:32Z
issue,55274,[Feature Request] PowerNorm,https://github.com/pytorch/pytorch/issues/55274,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-04-07T14:54:13Z
issue,55260,Cuda memory leak check is somehow unstable with repeat_test_for_types,https://github.com/pytorch/pytorch/issues/55260,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: memory usage, module: testing",scan when time allows,2021-04-07T14:52:49Z
issue,37002,[RFC] Modularize DistributedDataParallel,https://github.com/pytorch/pytorch/issues/37002,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2021-04-07T08:39:41Z
issue,55122,Cusolver handle may decrease MAGMA performance on GPU,https://github.com/pytorch/pytorch/issues/55122,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: linear algebra, module: magma, module: performance",scan when time allows,2021-04-07T00:34:26Z
issue,5790,Add hookable weights,https://github.com/pytorch/pytorch/issues/5790,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2021-04-05T21:09:40Z
issue,28761,torch.tensor() is very slow when it is passed an h5py Dataset.,https://github.com/pytorch/pytorch/issues/28761,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: performance, module: tensor creation",scan when time allows,2021-04-05T17:27:21Z
issue,55299,[feature] `torch.polygamma` : Support Tensor for argument `n`,https://github.com/pytorch/pytorch/issues/55299,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: special,scan when time allows,2021-04-05T16:59:04Z
issue,55030,Add Minimal Gated Unit (MGU),https://github.com/pytorch/pytorch/issues/55030,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2021-04-05T14:47:42Z
issue,54938,[tests] `cumprod` OpInfo tests take long time to run (around 1min),https://github.com/pytorch/pytorch/issues/54938,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ci, module: tests",scan when time allows,2021-04-05T07:36:12Z
issue,54774,"RuntimeError: ""log2"" ""_vml_cpu"" not implemented for 'Half'",https://github.com/pytorch/pytorch/issues/54774,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: half, module: numpy",scan when time allows,2021-04-05T07:29:25Z
issue,55174,Isolate CPU tests from GPU tests,https://github.com/pytorch/pytorch/issues/55174,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tests,scan when time allows,2021-04-03T21:57:41Z
issue,55135,LR scheduler step() behaviour with and without epoch parameter,https://github.com/pytorch/pytorch/issues/55135,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2021-04-02T17:37:37Z
issue,55161,Internal assert failed: `iter.device(arg).is_cuda()`,https://github.com/pytorch/pytorch/issues/55161,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: error checking,scan when time allows,2021-04-02T06:44:57Z
issue,25979,[C++] `Module::pretty_print` is broken,https://github.com/pytorch/pytorch/issues/25979,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: nn",scan when time allows,2021-04-02T02:33:55Z
issue,55156,multiprocessing function cannot pass cuda objects use when calling inside from a DDP process,https://github.com/pytorch/pytorch/issues/55156,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ddp, oncall: distributed",scan when time allows,2021-04-01T23:40:32Z
issue,55143,Add suport to torch.gather for negative indices,https://github.com/pytorch/pytorch/issues/55143,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: advanced indexing,scan when time allows,2021-04-01T21:34:17Z
issue,55104,Alternative backend fallback like mechanism which has higher precedence than operator-specific composite implementations,https://github.com/pytorch/pytorch/issues/55104,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: internals,scan when time allows,2021-04-01T21:30:22Z
issue,55120,Why isn't pip finding the correct pytorch dependency?,https://github.com/pytorch/pytorch/issues/55120,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: binaries,scan when time allows,2021-04-01T21:26:46Z
issue,54030,[FR] Configurable gain value for kaiming_normal_ / kaiming_uniform_ in torch.nn.init,https://github.com/pytorch/pytorch/issues/54030,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: initialization, module: nn",scan when time allows,2021-04-01T17:54:01Z
issue,54055,Wrong initialisation gain for SNNs/SELU,https://github.com/pytorch/pytorch/issues/54055,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: initialization, module: nn",scan when time allows,2021-04-01T14:14:04Z
issue,54942,Optimizations to TORCH_CHECK change inlining behavior.,https://github.com/pytorch/pytorch/issues/54942,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, module: performance",scan when time allows,2021-04-01T04:20:10Z
issue,54874,Get a thread safe copy of torch::nn::Sequential object,https://github.com/pytorch/pytorch/issues/54874,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: nn",scan when time allows,2021-03-31T21:07:47Z
issue,54104,TorchVision and TorchAudio wheels for AArch64 absent from https://download.pytorch.org/whl/torch_stable.html,https://github.com/pytorch/pytorch/issues/54104,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: arm, module: build, module: vision",scan when time allows,2021-03-31T19:56:55Z
issue,55083,[JIT][Testing] Split up test_jit.py and test_jit_py3.jit,https://github.com/pytorch/pytorch/issues/55083,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2021-03-31T18:44:42Z
issue,55045,16 bit accumulation for Convolution,https://github.com/pytorch/pytorch/issues/55045,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: convolution,scan when time allows,2021-03-31T17:51:02Z
issue,12060,cuda test hangs if GPUs in Exclusive Process mode,https://github.com/pytorch/pytorch/issues/12060,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2021-03-31T17:47:48Z
issue,12069,Stop using make_intrusive directly; provide some make_tensor,https://github.com/pytorch/pytorch/issues/12069,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: internals,scan when time allows,2021-03-31T17:47:17Z
issue,12159,clip_grad_norm_ does not work on grads of different types,https://github.com/pytorch/pytorch/issues/12159,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2021-03-31T17:43:22Z
issue,12189,[Feature] Support Adaptive Max Gradient Norm / Clipping,https://github.com/pytorch/pytorch/issues/12189,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: norms and normalization,scan when time allows,2021-03-31T17:42:15Z
issue,54564,why single pytorch process is displayed on each GPUs with memory 0MB,https://github.com/pytorch/pytorch/issues/54564,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2021-03-31T13:50:13Z
issue,54983,Accept objects with `__float__` wherever regular `float`s are accepted,https://github.com/pytorch/pytorch/issues/54983,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: half,scan when time allows,2021-03-31T05:00:52Z
issue,54897,Default generated meta functions for inplace operations don't report errors,https://github.com/pytorch/pytorch/issues/54897,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-03-31T04:12:30Z
issue,54947,Results surprisingly differ on different backends,https://github.com/pytorch/pytorch/issues/54947,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: numerical-reproducibility,scan when time allows,2021-03-30T23:39:25Z
issue,54843,Inconsistent behaviour with `weight` argument in CrossEntropyLoss and BCEWithLogitsLoss,https://github.com/pytorch/pytorch/issues/54843,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: loss,scan when time allows,2021-03-30T23:21:31Z
issue,55005,[TensorExpr] Provide a safe API to request the N-th loop for a given Buf.,https://github.com/pytorch/pytorch/issues/55005,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: bootcamp,scan when time allows,2021-03-30T22:58:10Z
issue,54861,Occured error in loss.backward() when using sparse=True in Embedding layer,https://github.com/pytorch/pytorch/issues/54861,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: embedding, module: sparse",scan when time allows,2021-03-30T21:49:20Z
issue,54815,TensorExpr `LoopNest.get_loops_for` misbehaved after loop distribution transformation,https://github.com/pytorch/pytorch/issues/54815,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp,scan when time allows,2021-03-30T21:32:52Z
issue,54829,c++ libtorch config mismatch with python version,https://github.com/pytorch/pytorch/issues/54829,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2021-03-30T16:15:10Z
issue,54802,[nnc] simplify is sometimes broken after computeAt,https://github.com/pytorch/pytorch/issues/54802,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-03-30T15:56:23Z
issue,54943,Structured kernels have increased TensorIterator overhead.,https://github.com/pytorch/pytorch/issues/54943,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, module: performance",scan when time allows,2021-03-30T07:01:00Z
issue,54677,[docs] use sphinx link/reference checking features,https://github.com/pytorch/pytorch/issues/54677,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: doc infra, module: docs",scan when time allows,2021-03-30T05:01:26Z
issue,54823,"Fixing pytorch distributed training (DDP) code send a SIGKILL signal on its own on most version of pytorch 1.6.0, 1.7.0, 1.7.1?",https://github.com/pytorch/pytorch/issues/54823,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2021-03-29T20:24:26Z
issue,54614,Dispatcher TLS to bypass loads of dispatch key from tensor arguments,https://github.com/pytorch/pytorch/issues/54614,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dispatch, module: performance",scan when time allows,2021-03-29T15:51:32Z
issue,46321,Add distributed examples into PyTorch CI tests,https://github.com/pytorch/pytorch/issues/46321,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, module: tests, oncall: distributed",scan when time allows,2021-03-29T14:16:03Z
issue,54753,torch.sqrt for negative values should either return complex tensors or clearly throw a domain error/warning,https://github.com/pytorch/pytorch/issues/54753,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: complex, module: numpy",scan when time allows,2021-03-29T10:39:56Z
issue,54713,Test test_large_cumprod_cuda_float16 gets killed due to (probably) OOM,https://github.com/pytorch/pytorch/issues/54713,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: memory usage, module: tests",scan when time allows,2021-03-29T02:49:55Z
issue,52552,Cumulative integration?,https://github.com/pytorch/pytorch/issues/52552,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: numpy,scan when time allows,2021-03-28T11:03:27Z
issue,54841,[testing] test_reference_numerics_extremal_clamp_cpu_bfloat16 fails on ci build with GCC5.4 and Python 3.6,https://github.com/pytorch/pytorch/issues/54841,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bfloat16, module: tests",scan when time allows,2021-03-28T06:30:06Z
issue,49726,Wrong error is raised for property of nn.Module (again),https://github.com/pytorch/pytorch/issues/49726,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2021-03-27T23:40:33Z
issue,53707,[RFC] Add common device runtime abstraction,https://github.com/pytorch/pytorch/issues/53707,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-03-27T10:06:47Z
issue,54757,[nnc] computeAt and index flatten broken in presence of if statements,https://github.com/pytorch/pytorch/issues/54757,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-03-26T22:20:15Z
issue,11980,[Enhancement] Increase user-friendliness of dataset.random_split,https://github.com/pytorch/pytorch/issues/11980,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: data,scan when time allows,2021-03-26T18:00:48Z
issue,11974,Specify out= argument to convolution,https://github.com/pytorch/pytorch/issues/11974,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: convolution,scan when time allows,2021-03-26T16:40:13Z
issue,3877,Wrap Cephes library for mathematical special functions,https://github.com/pytorch/pytorch/issues/3877,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: numpy,scan when time allows,2021-03-26T04:57:15Z
issue,53642,torch.unsafe_chunk and torch.unsafe_split 's documentation does not render on the website.,https://github.com/pytorch/pytorch/issues/53642,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: viewing and reshaping",scan when time allows,2021-03-25T18:58:30Z
issue,45996,Backpropagation for sparse matrix indexing is problematic (colab provided),https://github.com/pytorch/pytorch/issues/45996,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2021-03-25T07:52:32Z
issue,54629,"""Skipped!"" tests should have more descriptive skipped reason",https://github.com/pytorch/pytorch/issues/54629,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tests,scan when time allows,2021-03-24T21:52:35Z
issue,54620,"torch.pow(tensor, tensor) throws RuntimeError for dtype bool",https://github.com/pytorch/pytorch/issues/54620,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: numpy,scan when time allows,2021-03-24T20:36:30Z
issue,54566,Support docstring definition in torch/library.h operator definition,https://github.com/pytorch/pytorch/issues/54566,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-03-24T20:34:25Z
issue,53957,Is pytorch 1.8.0 incompatible with cuda 11.2 or what is the reason for this error?,https://github.com/pytorch/pytorch/issues/53957,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2021-03-24T14:13:31Z
issue,54546,Let backends specify a schema version when registering kernels,https://github.com/pytorch/pytorch/issues/54546,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: backend, module: codegen, module: internals",scan when time allows,2021-03-23T21:44:03Z
issue,52940,[JIT] Provide a way to access a flattened optimized graph from Profiling Executor,https://github.com/pytorch/pytorch/issues/52940,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, oncall: jit",scan when time allows,2021-03-23T21:23:54Z
issue,54407,test_clamp fails on ppc64le,https://github.com/pytorch/pytorch/issues/54407,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: power, oncall: jit",scan when time allows,2021-03-23T17:27:33Z
issue,54445,add tests to cover DDP on mixed dense-sparse models,https://github.com/pytorch/pytorch/issues/54445,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ddp, oncall: distributed",scan when time allows,2021-03-23T02:01:56Z
issue,54283,Split VariableTypeManual.cpp,https://github.com/pytorch/pytorch/issues/54283,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dispatch,scan when time allows,2021-03-22T23:33:21Z
issue,54375,[proposal] Integrate autograd graph visualizer for displaying crashed locations during detect_anomaly,https://github.com/pytorch/pytorch/issues/54375,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2021-03-22T22:38:43Z
issue,11786,Mysterious error due to num_workers: 1,https://github.com/pytorch/pytorch/issues/11786,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: multiprocessing,scan when time allows,2021-03-22T21:01:21Z
issue,11790,set num_workers on the dataloader make the jupyter kernel crash at the almost end of the epoch,https://github.com/pytorch/pytorch/issues/11790,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2021-03-22T21:00:06Z
issue,11854,error: ‘array_size’ is not a class template,https://github.com/pytorch/pytorch/issues/11854,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cpp",scan when time allows,2021-03-22T20:53:04Z
issue,53623,test/distributed/test_c10d.py::DistributedDataParallelTest only passes if run before other test cases,https://github.com/pytorch/pytorch/issues/53623,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2021-03-22T18:01:44Z
issue,54138,Support Array Interface (__array_interface__ attribute),https://github.com/pytorch/pytorch/issues/54138,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: numpy,scan when time allows,2021-03-22T17:28:31Z
issue,54387,[docs] torch.lerp has unmatching signature and explained parameter names,https://github.com/pytorch/pytorch/issues/54387,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: interpolation",scan when time allows,2021-03-22T17:24:05Z
issue,54368,RMSProp documentation is confusing,https://github.com/pytorch/pytorch/issues/54368,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: optimizer",scan when time allows,2021-03-22T16:30:11Z
issue,54356,Generate delegates for all non-structured kernels,https://github.com/pytorch/pytorch/issues/54356,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dispatch,scan when time allows,2021-03-22T16:27:58Z
issue,54351,Numpy.float64 vs native python float breaks DDP,https://github.com/pytorch/pytorch/issues/54351,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ddp,scan when time allows,2021-03-22T16:26:31Z
issue,54213,"[docs] torch.clamp / torch.clip do support None as placeholder of min/max, but this is not documented",https://github.com/pytorch/pytorch/issues/54213,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2021-03-22T16:23:03Z
issue,54340,RuntimeError: polar does not support automatic differentiation for outputs with complex dtype.,https://github.com/pytorch/pytorch/issues/54340,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-03-22T16:22:24Z
issue,47175,[feature request] Show warning if optimizer.zero_grad() was not called before optimizer.step(),https://github.com/pytorch/pytorch/issues/47175,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: optimizer, module: ux",scan when time allows,2021-03-22T02:41:42Z
issue,54307,cuDNN error upon the backward method with a sliced tensor output from nn.GRU.,https://github.com/pytorch/pytorch/issues/54307,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: cudnn, module: rnn",scan when time allows,2021-03-20T02:36:25Z
issue,54366,[package] Heuristics for extern-ing common dependencies,https://github.com/pytorch/pytorch/issues/54366,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-03-20T01:06:46Z
issue,54362,[package] Buck integration for dependency analysis,https://github.com/pytorch/pytorch/issues/54362,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-03-20T00:54:21Z
issue,54364,[package] Clear and minimal interface for what a package provides,https://github.com/pytorch/pytorch/issues/54364,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-03-20T00:52:50Z
issue,11727,Semaphore leaks in dataloader,https://github.com/pytorch/pytorch/issues/11727,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2021-03-19T15:45:12Z
issue,54279,[Static Runtime] static_runtime_benchmark compile-time error,https://github.com/pytorch/pytorch/issues/54279,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2021-03-19T13:24:35Z
issue,54278,[Static Runtime] StaticRuntime.EmbeddingBag test case broken,https://github.com/pytorch/pytorch/issues/54278,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tests,scan when time allows,2021-03-19T13:23:59Z
issue,54269,SummaryWriter add_image() docs do not state that it is expecting images in a certain format,https://github.com/pytorch/pytorch/issues/54269,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: tensorboard, oncall: visualization",scan when time allows,2021-03-18T21:29:13Z
issue,54201,Provide half-away-from-zero rounding mode on Tensor::round,https://github.com/pytorch/pytorch/issues/54201,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp,scan when time allows,2021-03-18T13:09:10Z
issue,54192,OSS Mobile Errors Point to fburls,https://github.com/pytorch/pytorch/issues/54192,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-03-18T12:50:42Z
issue,54149,OSS build doesn't hard fail on nodiscard,https://github.com/pytorch/pytorch/issues/54149,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build warnings,scan when time allows,2021-03-18T12:47:29Z
issue,54139,as_tensor returns CPU copy of a CUDA buffer object implementing CUDA Array Interface,https://github.com/pytorch/pytorch/issues/54139,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: numba",scan when time allows,2021-03-18T01:56:36Z
issue,53124,Deprecate and remove torch.set_default_tensor_type,https://github.com/pytorch/pytorch/issues/53124,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: deprecation,scan when time allows,2021-03-18T00:09:28Z
issue,46187,Better support for operators that return (named) tuples of tensors,https://github.com/pytorch/pytorch/issues/46187,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: deprecation, module: numpy, module: ux",scan when time allows,2021-03-17T23:10:58Z
issue,49456,Impossible to run tests target with LibTorch as dependency inside a cocoapods.,https://github.com/pytorch/pytorch/issues/49456,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ios,scan when time allows,2021-03-17T20:53:42Z
issue,54116,torch.put is divergent from np.put,https://github.com/pytorch/pytorch/issues/54116,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: numpy,scan when time allows,2021-03-17T09:27:15Z
issue,54058,c++ convert from std::vector<Tensor> to c10::List<optional<Tensor>>,https://github.com/pytorch/pytorch/issues/54058,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: docs",scan when time allows,2021-03-17T06:21:43Z
issue,54062,Many advanced indexing operations have untested large tensor branches,https://github.com/pytorch/pytorch/issues/54062,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: advanced indexing, module: tests",scan when time allows,2021-03-17T05:49:47Z
issue,54077,[cuda extension] tutorial should mention device guard,https://github.com/pytorch/pytorch/issues/54077,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp-extensions, module: docs",scan when time allows,2021-03-17T05:48:48Z
issue,54096,cdist skip for manual backward is kind of janky,https://github.com/pytorch/pytorch/issues/54096,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dispatch, module: distance functions",scan when time allows,2021-03-17T05:42:21Z
issue,22343,New Weight Scheduler Concept for Weight Decay,https://github.com/pytorch/pytorch/issues/22343,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2021-03-17T01:43:13Z
issue,37809,"Add Numpy-like ""order"" argument to reshape",https://github.com/pytorch/pytorch/issues/37809,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: numpy,scan when time allows,2021-03-17T00:10:30Z
issue,49440,[RFC] DataLoader architecture updates and TarDataset implementation,https://github.com/pytorch/pytorch/issues/49440,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2021-03-16T02:05:24Z
issue,53982,[Feature Request] Optionally specify batch dimension in DataLoader and collate_fn,https://github.com/pytorch/pytorch/issues/53982,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2021-03-15T21:50:14Z
issue,54021,Support scatter/gather for NCCL backend,https://github.com/pytorch/pytorch/issues/54021,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: c10d,scan when time allows,2021-03-15T20:11:08Z
issue,49481,rank-0 gpu consume too high memory,https://github.com/pytorch/pytorch/issues/49481,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2021-03-15T17:50:09Z
issue,53407,"torch.matmul doesn't handle zero-sized inputs in some cases, leading to batched grad failures",https://github.com/pytorch/pytorch/issues/53407,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: tests, module: vmap",scan when time allows,2021-03-15T17:05:34Z
issue,52465,[nnc][perf] Performance decrease with CPU fusion on `freeze(script(pytorch_mobilenet_v3))`,https://github.com/pytorch/pytorch/issues/52465,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: performance,scan when time allows,2021-03-15T16:53:02Z
issue,53386,Default pytest is truncating stacktrace (somehow!!!),https://github.com/pytorch/pytorch/issues/53386,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-03-15T16:45:42Z
issue,11514,[feature request] - Allow sequences lengths to be 0 in PackSequence,https://github.com/pytorch/pytorch/issues/11514,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: rnn",scan when time allows,2021-03-15T16:02:14Z
issue,11537,"[CLEANUP] Context functions should return TypeExtendedInterface, not Type",https://github.com/pytorch/pytorch/issues/11537,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-03-15T15:48:32Z
issue,11516,at::Device makes it very easy to write buggy code,https://github.com/pytorch/pytorch/issues/11516,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-03-15T15:44:42Z
issue,11551,"CrossEntropyLoss, ignore_index does not prevent back-prop if the logits are -inf",https://github.com/pytorch/pytorch/issues/11551,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: loss, module: nn",scan when time allows,2021-03-15T15:40:11Z
issue,53983,"hope to support something like ""torch::manual_seed_for_mulit_thread""",https://github.com/pytorch/pytorch/issues/53983,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: random",scan when time allows,2021-03-15T14:19:32Z
issue,54001,[codegen] Change public C++ to accept out/mut arguments by const&,https://github.com/pytorch/pytorch/issues/54001,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: codegen,scan when time allows,2021-03-15T14:16:39Z
issue,30786,C++ / JIT Parity for ops in `torch/functional.py` and `torch/tensor.py`,https://github.com/pytorch/pytorch/issues/30786,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2021-03-15T13:58:28Z
issue,11624,Add min mode to embedding bags,https://github.com/pytorch/pytorch/issues/11624,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2021-03-15T13:49:10Z
issue,11635,[Feature request] Advanced indexing in functions like `expand`,https://github.com/pytorch/pytorch/issues/11635,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: advanced indexing,scan when time allows,2021-03-15T13:46:59Z
issue,11645,One GPU is more memory efficient than Multiple GPUs,https://github.com/pytorch/pytorch/issues/11645,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: data parallel, module: multi-gpu",scan when time allows,2021-03-15T13:44:20Z
issue,53890,GoogleNet pretrained/scratch bug?,https://github.com/pytorch/pytorch/issues/53890,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: vision,scan when time allows,2021-03-15T01:30:54Z
issue,42849,[RFC] DeepSpeed + PT Distributed Integration,https://github.com/pytorch/pytorch/issues/42849,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2021-03-13T22:44:24Z
issue,53900,Have the possibility to reduce a tensor with median on more than one specified dimension,https://github.com/pytorch/pytorch/issues/53900,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: numpy, module: reductions",scan when time allows,2021-03-13T10:55:47Z
issue,49252,Methods that solve systems of linear equations are memory-inefficient (batch-wise broadcasting),https://github.com/pytorch/pytorch/issues/49252,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: linear algebra, module: memory usage",scan when time allows,2021-03-13T07:47:37Z
issue,11439,High leverage TH operations to port to ATen,https://github.com/pytorch/pytorch/issues/11439,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: porting, module: tensor creation",scan when time allows,2021-03-12T22:49:01Z
issue,11426,[Feature request] Intuitive error message when input to Linear is not cudarized,https://github.com/pytorch/pytorch/issues/11426,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: error checking,scan when time allows,2021-03-12T22:13:25Z
issue,11389,[distributions] Torch distribution samplers slow on expanded parameters,https://github.com/pytorch/pytorch/issues/11389,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distributions,scan when time allows,2021-03-12T22:10:32Z
issue,53910,There should be a CI build with OpenBLAS in addition to MKL,https://github.com/pytorch/pytorch/issues/53910,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2021-03-12T20:30:58Z
issue,52274,Port unary elementwise ops to structured kernels,https://github.com/pytorch/pytorch/issues/52274,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: structured kernels,scan when time allows,2021-03-12T16:18:58Z
issue,46822,C++ Optimizer: remove warning on Optimizer::size method,https://github.com/pytorch/pytorch/issues/46822,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2021-03-12T15:36:54Z
issue,53477,Show methods of the class in the right sidebar,https://github.com/pytorch/pytorch/issues/53477,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2021-03-12T10:51:05Z
issue,53866,Add comprehensive subgroup testing to torch.distributed,https://github.com/pytorch/pytorch/issues/53866,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: c10d, module: ddp, oncall: distributed",scan when time allows,2021-03-12T02:00:53Z
issue,53427,Backward compute time in DDP logging may not be accurate,https://github.com/pytorch/pytorch/issues/53427,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2021-03-11T21:37:58Z
issue,53666,torch.autograd.functional.jacobian/hessian allow tensor in addition to function,https://github.com/pytorch/pytorch/issues/53666,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2021-03-11T14:16:25Z
issue,53708,torch.load with Exception,https://github.com/pytorch/pytorch/issues/53708,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: serialization,scan when time allows,2021-03-11T05:56:37Z
issue,53584,How to delete Module from GPU? (libtorch C++),https://github.com/pytorch/pytorch/issues/53584,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: cpp-extensions",scan when time allows,2021-03-11T03:11:09Z
issue,53491,Very slow backward speed when using gather with small-range indices,https://github.com/pytorch/pytorch/issues/53491,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: performance,scan when time allows,2021-03-10T18:03:45Z
issue,53193,[nnc] Prevent user from scheduling after prepareForCodegen,https://github.com/pytorch/pytorch/issues/53193,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-03-10T18:00:06Z
issue,52767,[nnc] Grouped convolutions,https://github.com/pytorch/pytorch/issues/52767,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-03-10T17:53:50Z
issue,53463,[Feature Request] return type support for `torch.norm` and tensor iteration in for loop,https://github.com/pytorch/pytorch/issues/53463,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-03-10T17:51:26Z
issue,53710,Ninja recompiles certain unchanged source files in cpp extension since torch 1.8.0,https://github.com/pytorch/pytorch/issues/53710,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2021-03-10T17:43:24Z
issue,53442,IRSimplifier performs unsafe transformation on mod: m*n%m -> 0,https://github.com/pytorch/pytorch/issues/53442,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-03-10T06:56:30Z
issue,47502,Hackable python-only autograd engine,https://github.com/pytorch/pytorch/issues/47502,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2021-03-10T01:44:27Z
issue,53622,Properly design manual_cpp_binding (make it less error prone),https://github.com/pytorch/pytorch/issues/53622,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: language binding",scan when time allows,2021-03-10T00:07:31Z
issue,53627,test_metal.py must skip when not compiled with metal support,https://github.com/pytorch/pytorch/issues/53627,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tests,scan when time allows,2021-03-10T00:03:25Z
issue,40034,Doing optimizing compilations in a separate thread,https://github.com/pytorch/pytorch/issues/40034,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-03-09T22:55:47Z
issue,53649,vmap gradgradcheck test fails for unfold operation,https://github.com/pytorch/pytorch/issues/53649,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: vmap",scan when time allows,2021-03-09T21:36:27Z
issue,50503,Huge file sizes for libtorch and dependencies?,https://github.com/pytorch/pytorch/issues/50503,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: build",scan when time allows,2021-03-09T14:30:44Z
issue,19177,8 tests in test_c10d fail when running all tests in one command,https://github.com/pytorch/pytorch/issues/19177,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: tests, oncall: distributed",scan when time allows,2021-03-09T08:12:00Z
issue,46225,torch.mode when input has nans,https://github.com/pytorch/pytorch/issues/46225,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: nans and infs, module: numpy, module: reductions",scan when time allows,2021-03-08T16:58:18Z
issue,53257,Inconsistent `out=` behaviour on advanced indexing operations.,https://github.com/pytorch/pytorch/issues/53257,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: advanced indexing, module: complex",scan when time allows,2021-03-08T15:47:03Z
issue,47378,torch.vmap giving INTERNAL ASSERT FAILED error,https://github.com/pytorch/pytorch/issues/47378,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: vmap,scan when time allows,2021-03-08T15:33:18Z
issue,53337,Should we remind users not to use the dataset on GPU when it's the argument of DataLoader?,https://github.com/pytorch/pytorch/issues/53337,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2021-03-08T04:24:44Z
issue,52265,'TypeError: expected CPU (got CUDA)' when placing a CUDA tensor on a class that is inheriting torch.Tensor,https://github.com/pytorch/pytorch/issues/52265,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tensor creation,scan when time allows,2021-03-07T22:14:12Z
issue,53405,When I ran the 'Get Started with Data Data Parallel' tutorial the code got 'RuntimeError '!There are no hints,https://github.com/pytorch/pytorch/issues/53405,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: data parallel,scan when time allows,2021-03-05T21:27:41Z
issue,53391,torch.empty_strided doesn't test if strides are negative,https://github.com/pytorch/pytorch/issues/53391,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: error checking,scan when time allows,2021-03-05T19:38:58Z
issue,53360,[bug] torch.cumsum: functional and method variant promotes all ints to Long but inplace don't,https://github.com/pytorch/pytorch/issues/53360,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: type promotion",scan when time allows,2021-03-05T19:21:35Z
issue,53345,CUDA error: an illegal memory access was encountered when updating model weights using GradScaler,https://github.com/pytorch/pytorch/issues/53345,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision), module: cuda, module: nn",scan when time allows,2021-03-05T18:01:05Z
issue,53090,'Tensor' object has no attribute 'astype',https://github.com/pytorch/pytorch/issues/53090,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: numpy,scan when time allows,2021-03-05T10:34:42Z
issue,53284,Find a good namespace home for torch._assert_async,https://github.com/pytorch/pytorch/issues/53284,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: numpy, module: testing",scan when time allows,2021-03-05T10:32:13Z
issue,52745,Sort torch.topk() output following input index order,https://github.com/pytorch/pytorch/issues/52745,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-03-05T00:49:13Z
issue,53157,doxygen and pytorch documentation,https://github.com/pytorch/pytorch/issues/53157,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: docs",scan when time allows,2021-03-04T20:46:12Z
issue,53256,`index_copy_` test fails on PyTorch/XLA,https://github.com/pytorch/pytorch/issues/53256,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: tests, module: xla",scan when time allows,2021-03-04T18:10:57Z
issue,53185,[NIT] Unpickler persistent_load monkey patch should fall back to original persistent_load,https://github.com/pytorch/pytorch/issues/53185,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: serialization,scan when time allows,2021-03-04T16:56:33Z
issue,53194,Relative performance of histc vs bincount,https://github.com/pytorch/pytorch/issues/53194,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: performance, module: sorting and selection",scan when time allows,2021-03-04T16:56:00Z
issue,53130,Lint rule to forbid bare assert() in cuda,https://github.com/pytorch/pytorch/issues/53130,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: lint",scan when time allows,2021-03-04T16:44:21Z
issue,53141,Structured kernels for operators that don't have out= variants,https://github.com/pytorch/pytorch/issues/53141,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: structured kernels,scan when time allows,2021-03-04T16:43:42Z
issue,41337,Distributed tests fail with pytest,https://github.com/pytorch/pytorch/issues/41337,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2021-03-04T10:40:14Z
issue,41256,Add nn.functional.swish alias for nn.functional.silu,https://github.com/pytorch/pytorch/issues/41256,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2021-03-04T07:53:25Z
issue,52762,Basic feature missing,https://github.com/pytorch/pytorch/issues/52762,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2021-03-03T14:37:44Z
issue,52913,Create a standalone python execution viewer like livepython,https://github.com/pytorch/pytorch/issues/52913,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: logging,scan when time allows,2021-03-03T11:52:54Z
issue,35534,Fail to load torch script model,https://github.com/pytorch/pytorch/issues/35534,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2021-03-03T07:22:57Z
issue,53094,group conv in amp too slower,https://github.com/pytorch/pytorch/issues/53094,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision), module: cuda, module: performance",scan when time allows,2021-03-03T04:21:40Z
issue,52583,[TensorExpr] Add IR Verifier,https://github.com/pytorch/pytorch/issues/52583,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: bootcamp,scan when time allows,2021-03-02T17:40:41Z
issue,53072,[NNC] Simplify after each transformation,https://github.com/pytorch/pytorch/issues/53072,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-03-02T03:49:49Z
issue,11113,use_system_nccl flag does not work?,https://github.com/pytorch/pytorch/issues/11113,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2021-03-01T21:41:58Z
issue,11072,[...] operator for masked select does not broadcast anymore,https://github.com/pytorch/pytorch/issues/11072,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sorting and selection,scan when time allows,2021-03-01T21:27:56Z
issue,11071,undefined reference to 'caffe2::Caffe2FlagsRegistry[abi:cxx11]()',https://github.com/pytorch/pytorch/issues/11071,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2021-03-01T21:18:21Z
issue,11062,"non-shuffling data loaders can affect random states, thus the results of shuffling data loaders.",https://github.com/pytorch/pytorch/issues/11062,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2021-03-01T21:17:24Z
issue,10913,Cannot find operator schema for 'ATen' Caffe2 Ios,https://github.com/pytorch/pytorch/issues/10913,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-03-01T21:06:50Z
issue,52680,Add Quantized{CPU|CUDA} support to Structured Kernels,https://github.com/pytorch/pytorch/issues/52680,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: internals, module: structured kernels",scan when time allows,2021-03-01T21:03:11Z
issue,53023,[codegen] Make it easier to codegen call to API,https://github.com/pytorch/pytorch/issues/53023,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: codegen,scan when time allows,2021-03-01T19:23:09Z
issue,52673,Add custom types to PyTorch,https://github.com/pytorch/pytorch/issues/52673,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-03-01T18:29:29Z
issue,52911,LR Scheduler load_state_dict does not properly update scaling,https://github.com/pytorch/pytorch/issues/52911,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2021-03-01T16:02:47Z
issue,39503,pytest suppresses stderr from Python startup by default,https://github.com/pytorch/pytorch/issues/39503,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: logging, module: tests",scan when time allows,2021-02-28T19:59:50Z
issue,14864,[discussion] Recommend a different file extension for models (.PTH is a special extension for Python),https://github.com/pytorch/pytorch/issues/14864,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-02-28T12:29:49Z
issue,32552,MultivariateNormal.rsample: use eigen-decomposition when Cholesky fails,https://github.com/pytorch/pytorch/issues/32552,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: distributions, module: linear algebra",scan when time allows,2021-02-26T16:17:59Z
issue,10719,size mismatch when trying to reconstruct predifined network,https://github.com/pytorch/pytorch/issues/10719,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: vision,scan when time allows,2021-02-25T22:30:40Z
issue,52850,How to skip the images in a custom dataset and deal with None values?,https://github.com/pytorch/pytorch/issues/52850,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2021-02-25T22:04:56Z
issue,52800,Create CI test checker to ensure # of test ran is the same as # of test reported in xmlrunner,https://github.com/pytorch/pytorch/issues/52800,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2021-02-25T21:45:40Z
issue,52845,[feature request] F.interpolate_as,https://github.com/pytorch/pytorch/issues/52845,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: interpolation, module: nn",scan when time allows,2021-02-25T20:38:42Z
issue,52722,Creating torch tensor as a function of index value,https://github.com/pytorch/pytorch/issues/52722,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: numpy,scan when time allows,2021-02-25T17:48:05Z
issue,52005,Implementation of Normalized Euclidean Distance (NED) in pytorch,https://github.com/pytorch/pytorch/issues/52005,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: distance functions, module: nn",scan when time allows,2021-02-25T17:44:22Z
issue,52211,Conv3D error : CUDNN_STATUS_INTERNAL_ERROR,https://github.com/pytorch/pytorch/issues/52211,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: cudnn",scan when time allows,2021-02-25T05:10:18Z
issue,52688,[BE] Consolidating scripts/ and tools/ folder,https://github.com/pytorch/pytorch/issues/52688,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2021-02-25T02:05:56Z
issue,52712,Refactor `torch.linalg.norm`,https://github.com/pytorch/pytorch/issues/52712,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: linear algebra,scan when time allows,2021-02-25T02:03:27Z
issue,52717,Fix `torch.norm`'s backward to match forward for ord=+/-inf,https://github.com/pytorch/pytorch/issues/52717,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: linear algebra",scan when time allows,2021-02-25T02:02:38Z
issue,52744,Gradient checkpointing support in C++ API,https://github.com/pytorch/pytorch/issues/52744,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp,scan when time allows,2021-02-25T02:00:22Z
issue,52334,Normal_like operator,https://github.com/pytorch/pytorch/issues/52334,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: numpy,scan when time allows,2021-02-24T18:16:37Z
issue,3920,[docs] Tensor.new is not documented,https://github.com/pytorch/pytorch/issues/3920,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2021-02-24T13:57:50Z
issue,52205,Allow F.pad(mode = 'reflect') when shape == pad,https://github.com/pytorch/pytorch/issues/52205,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: numpy, module: padding",scan when time allows,2021-02-24T04:07:46Z
issue,52578,version1.7.0 is ~1.3x slower than 1.4.0 on ResNet18,https://github.com/pytorch/pytorch/issues/52578,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: performance,scan when time allows,2021-02-24T01:50:03Z
issue,52648,Fix `torch.linalg.vector_norm`'s backward to avoid needing `at::abs()` call in forward for ord=+/-inf,https://github.com/pytorch/pytorch/issues/52648,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: linear algebra,scan when time allows,2021-02-24T01:49:13Z
issue,51794,"optim.Optimizer should copy ""params"" before modifying them",https://github.com/pytorch/pytorch/issues/51794,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2021-02-24T01:08:54Z
issue,46886,Support deleting a parameter/buffer by name,https://github.com/pytorch/pytorch/issues/46886,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: ux",scan when time allows,2021-02-23T23:30:48Z
issue,22755,[RFC] InstanceNorm default affine value,https://github.com/pytorch/pytorch/issues/22755,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2021-02-23T10:24:59Z
issue,52168,CMake errors building project that uses PyTorch,https://github.com/pytorch/pytorch/issues/52168,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2021-02-23T08:49:47Z
issue,52605,Torchaudio 0.8.0.dev nightly requires torch 1.9.0.dev which is missing,https://github.com/pytorch/pytorch/issues/52605,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2021-02-23T06:36:42Z
issue,52644,[FR] torch.load's map_location supports int,https://github.com/pytorch/pytorch/issues/52644,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2021-02-23T02:25:29Z
issue,52598,CUDA error: no kernel image is available for execution on the device When convert cpu tensor to gpu tensor,https://github.com/pytorch/pytorch/issues/52598,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cuda",scan when time allows,2021-02-23T01:11:25Z
issue,52570,"RTX 2080s performs better than RTX 3080 in Semantic Segmentation inference process(Libtorch,win10),why?",https://github.com/pytorch/pytorch/issues/52570,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: performance",scan when time allows,2021-02-22T23:07:27Z
issue,52597,Custom scatter function for DataParallel,https://github.com/pytorch/pytorch/issues/52597,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: data parallel,scan when time allows,2021-02-22T22:06:27Z
issue,52569,Omitting mutability annotations in op schemas can lead to subtle errors,https://github.com/pytorch/pytorch/issues/52569,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: internals,scan when time allows,2021-02-22T21:52:59Z
issue,51038,The pooling code is ignoring sliding windows that start within the right padding,https://github.com/pytorch/pytorch/issues/51038,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: pooling,scan when time allows,2021-02-22T19:26:19Z
issue,52208,C++ standard library confusion,https://github.com/pytorch/pytorch/issues/52208,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2021-02-22T18:31:27Z
issue,51849,.numpy() array failes to keep original storage around,https://github.com/pytorch/pytorch/issues/51849,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: numpy, module: tensor creation",scan when time allows,2021-02-22T18:21:28Z
issue,48245,"RuntimeError: ""threshold_cpu"" not implemented for 'Half'",https://github.com/pytorch/pytorch/issues/48245,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: half",scan when time allows,2021-02-22T08:24:52Z
issue,4906,"Rebuild from no-CUDA to CUDA leads to: error: #error ""Expected GLOO_USE_CUDA to be defined""",https://github.com/pytorch/pytorch/issues/4906,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2021-02-21T18:34:36Z
issue,39634,Installation from source fails on macOS (No CUDA),https://github.com/pytorch/pytorch/issues/39634,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cuda, module: macos",scan when time allows,2021-02-20T22:40:28Z
issue,52318,"TorchConfig.cmake entries of if(ON) are not accepted by CMake as ""true"", to be replaced by if(1)",https://github.com/pytorch/pytorch/issues/52318,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2021-02-20T13:27:45Z
issue,10683,Tensor.register_hook is not passing the tensor object to the hook function,https://github.com/pytorch/pytorch/issues/10683,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: bc-breaking,scan when time allows,2021-02-19T11:54:55Z
issue,10691,Build system doesn't prevent ATen/core from including non-core files,https://github.com/pytorch/pytorch/issues/10691,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2021-02-19T11:50:02Z
issue,10685,[feature request] convtbc with group convolution,https://github.com/pytorch/pytorch/issues/10685,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: convolution,scan when time allows,2021-02-19T11:49:11Z
issue,52478,"Could you please increase the wheel compiled Linux (aarch64) GPU, thank you very much!",https://github.com/pytorch/pytorch/issues/52478,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: arm,scan when time allows,2021-02-19T11:08:34Z
issue,50206,Build Fail with vulkan on ARM64 and wayland drivers,https://github.com/pytorch/pytorch/issues/50206,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2021-02-17T23:35:48Z
issue,52390,autograd.functional.vjp with function that mutates inputs in-place can cause confusion,https://github.com/pytorch/pytorch/issues/52390,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2021-02-17T22:38:40Z
issue,52298,[FR} Implement Skellam distribution,https://github.com/pytorch/pytorch/issues/52298,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distributions,scan when time allows,2021-02-17T19:04:44Z
issue,44809,Simple functions shouldn't go through dispatcher,https://github.com/pytorch/pytorch/issues/44809,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dispatch,scan when time allows,2021-02-17T14:40:17Z
issue,52266,GPU (+CPU/RAM?) self-check script or instructions in core,https://github.com/pytorch/pytorch/issues/52266,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: cudnn",scan when time allows,2021-02-17T14:29:00Z
issue,51622,"`fractional_max_pool{2, 3}d` inconsistent between CUDA and CPU",https://github.com/pytorch/pytorch/issues/51622,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: pooling,scan when time allows,2021-02-16T18:46:21Z
issue,10482,Reduce code duplication in interpolate and make it more generic,https://github.com/pytorch/pytorch/issues/10482,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: interpolation,scan when time allows,2021-02-16T01:21:48Z
issue,51777,Analytical metric package,https://github.com/pytorch/pytorch/issues/51777,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: data, module: distributions",scan when time allows,2021-02-15T20:16:57Z
issue,52011,CPU eval BatchNorm2d is not threaded,https://github.com/pytorch/pytorch/issues/52011,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: norms and normalization, module: performance",scan when time allows,2021-02-13T17:00:45Z
issue,32826,MKLDNN doesnt work and is slower than normal cpu mode,https://github.com/pytorch/pytorch/issues/32826,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: mkldnn,scan when time allows,2021-02-13T13:32:40Z
issue,50360,error: conversion from ‘std::vector<at::Tensor>’ to non-scalar type ‘at::Tensor’ requested,https://github.com/pytorch/pytorch/issues/50360,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: numpy,scan when time allows,2021-02-12T13:57:32Z
issue,52096,"Unable to build: subprocess.CalledProcessError: Command '['cmake', '--build', '.', '--target', 'install', '--config', 'Release', '--', '-j', '4']' returned non-zero exit status 1.",https://github.com/pytorch/pytorch/issues/52096,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2021-02-11T22:11:44Z
issue,52145,Trying to initialise CUDA twice in a process with no visible devices hangs the process and terminal permanently,https://github.com/pytorch/pytorch/issues/52145,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: deadlock",scan when time allows,2021-02-11T17:55:19Z
issue,52041,device (+dtype) arguments for torch.stack / torch.cat,https://github.com/pytorch/pytorch/issues/52041,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: performance",scan when time allows,2021-02-10T23:36:52Z
issue,51911,torch.nn.functional.grid_sample outputs NaN,https://github.com/pytorch/pytorch/issues/51911,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nans and infs, module: nn",scan when time allows,2021-02-10T18:35:39Z
issue,51231,macOS error building 1.7.1 from source on Catalina 10.15.7,https://github.com/pytorch/pytorch/issues/51231,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: macos",scan when time allows,2021-02-10T18:21:07Z
issue,51868,discrepancies between TorchConfigVersion.cmake to git branch,https://github.com/pytorch/pytorch/issues/51868,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2021-02-10T18:18:16Z
issue,43876,Insert at specific index/key in nn.Sequential,https://github.com/pytorch/pytorch/issues/43876,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2021-02-10T08:21:09Z
issue,40275,Conv3d with specific kernel size outputs inconsistent results between FP16 and FP32 in V100 GPU,https://github.com/pytorch/pytorch/issues/40275,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: cuda, module: cudnn, module: dependency bug",scan when time allows,2021-02-10T02:49:44Z
issue,45764,resize_(0) is very expensive,https://github.com/pytorch/pytorch/issues/45764,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: performance,scan when time allows,2021-02-09T23:46:34Z
issue,49101,[RFC] RemoteTensor: Use Remote Devices as Local Devices,https://github.com/pytorch/pytorch/issues/49101,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: rpc, oncall: distributed",scan when time allows,2021-02-09T22:44:21Z
issue,44738,[tools.codegen] Remove byte-for-byte compatibility code,https://github.com/pytorch/pytorch/issues/44738,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: codegen,scan when time allows,2021-02-09T21:05:59Z
issue,51493,Add branch predictor hints to prefer `Context::deterministicAlgorithms() == false`,https://github.com/pytorch/pytorch/issues/51493,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: determinism, module: performance",scan when time allows,2021-02-09T20:29:38Z
issue,51117,TestLinalgCPU.test_cholesky_solve_autograd_cpu_float64 fails on different random seed,https://github.com/pytorch/pytorch/issues/51117,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: correctness (silent), module: linear algebra",scan when time allows,2021-02-09T20:26:21Z
issue,49426,[doc] missing torch.pdist entry,https://github.com/pytorch/pytorch/issues/49426,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: distance functions, module: docs, module: nn",scan when time allows,2021-02-09T07:43:09Z
issue,51320,"Pytorch not working properly (I don't know how to summarize it, see below)",https://github.com/pytorch/pytorch/issues/51320,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2021-02-08T23:52:23Z
issue,10165,torch.utils.data.random_split() returns dataset index as tensor,https://github.com/pytorch/pytorch/issues/10165,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dataloader, module: docs",scan when time allows,2021-02-08T18:14:05Z
issue,51872,Tensor.nonzero tries to allocate huge amount of memory for tensors on GPU with num_elements close to INT_MAX,https://github.com/pytorch/pytorch/issues/51872,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: dependency bug, module: memory usage",scan when time allows,2021-02-08T17:43:41Z
issue,10386,[feature request] batch_first option in torch.utils.data,https://github.com/pytorch/pytorch/issues/10386,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2021-02-08T16:24:25Z
issue,51855,"Attempted to read a PyTorch file with version 3, but the maximum supported version for reading is 2. Your PyTorch installation may be too old. (init at /pytorch/caffe2/serialize/inline_container.cc:132)",https://github.com/pytorch/pytorch/issues/51855,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: assert failure,scan when time allows,2021-02-08T16:19:00Z
issue,48465,anchored direct links to functions disappeared in 1.7.0 docs,https://github.com/pytorch/pytorch/issues/48465,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: doc infra, module: docs",scan when time allows,2021-02-08T15:12:38Z
issue,41186,test_autograd failures on Power,https://github.com/pytorch/pytorch/issues/41186,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: cuda, module: linear algebra, module: power",scan when time allows,2021-02-08T04:01:51Z
issue,50835,Insufficient shared memory (shm) while training,https://github.com/pytorch/pytorch/issues/50835,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: memory usage,scan when time allows,2021-02-08T03:43:57Z
issue,51776,BatchNorm3d error : CUDNN_STATUS_NOT_SUPPORTED,https://github.com/pytorch/pytorch/issues/51776,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cudnn,scan when time allows,2021-02-07T00:07:44Z
issue,50382,out variant of many loss functions are not consistent with non-out variant when reduction is not none,https://github.com/pytorch/pytorch/issues/50382,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: loss, module: nn, module: reductions",scan when time allows,2021-02-05T21:12:29Z
issue,51781,Libtorch lock thread,https://github.com/pytorch/pytorch/issues/51781,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: abi,scan when time allows,2021-02-05T19:43:07Z
issue,39674,[BatchNorm] Add boolean flags to choose the stats for normalization,https://github.com/pytorch/pytorch/issues/39674,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: norms and normalization",scan when time allows,2021-02-05T18:34:01Z
issue,51711,Weird behavior in Conv2d padding when changed to 'reflect',https://github.com/pytorch/pytorch/issues/51711,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: docs, module: nn, module: padding",scan when time allows,2021-02-05T17:26:38Z
issue,51663,Torch DataLoader fails to reraise error from sqlalchemy.,https://github.com/pytorch/pytorch/issues/51663,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2021-02-05T14:42:11Z
issue,49110,[RFC] Integrate DeepSpeed CUDA kernels,https://github.com/pytorch/pytorch/issues/49110,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-02-04T22:19:27Z
issue,51224,`torch.cuda.device` should look like it is inherited from `torch.device`,https://github.com/pytorch/pytorch/issues/51224,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2021-02-04T16:41:20Z
issue,51256,[search] duplicated entries in results,https://github.com/pytorch/pytorch/issues/51256,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2021-02-04T14:42:25Z
issue,31178,torch::nn::functional::interpolate crash,https://github.com/pytorch/pytorch/issues/31178,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp,scan when time allows,2021-02-04T12:43:20Z
issue,40107,[RFC] Add a RPC context manager that collects/waits for all RPC futures created in scope,https://github.com/pytorch/pytorch/issues/40107,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, module: rpc",scan when time allows,2021-02-04T02:53:46Z
issue,51197,"[docs] use `versionadded`, `versionchanged` and `deprecated` directive",https://github.com/pytorch/pytorch/issues/51197,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2021-02-03T20:46:54Z
issue,51471,Make TorchBind object appear in state_dict,https://github.com/pytorch/pytorch/issues/51471,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: serialization",scan when time allows,2021-02-03T03:28:45Z
issue,51426,Different results on RTX 2060s vs GTX 1060 with gpt model,https://github.com/pytorch/pytorch/issues/51426,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: numerical-reproducibility,scan when time allows,2021-02-02T23:15:34Z
issue,51451,Add support for conan package manager,https://github.com/pytorch/pytorch/issues/51451,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-02-02T18:25:32Z
issue,51418,Pytorch num_worker>0 code worked first time and then it never worked with same setting again,https://github.com/pytorch/pytorch/issues/51418,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2021-02-02T17:57:54Z
issue,51406,[numpy] `round` and `trunc` not supported for Integral Type while Python and NumPy supports them,https://github.com/pytorch/pytorch/issues/51406,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: numpy,scan when time allows,2021-02-02T17:07:41Z
issue,51410,Caffe2 RNN Cell Names not rectified,https://github.com/pytorch/pytorch/issues/51410,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-02-02T17:03:15Z
issue,51407,Additon of levenberg-marquardt optimizer in TORCH.OPTIM,https://github.com/pytorch/pytorch/issues/51407,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2021-02-02T17:02:21Z
issue,51074,error: ‘PyFrameObject {aka struct _frame}’ has no member named ‘f_lasti’,https://github.com/pytorch/pytorch/issues/51074,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2021-02-02T16:49:54Z
issue,28619,Channels Last (NHWC) support plan.,https://github.com/pytorch/pytorch/issues/28619,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: memory format,scan when time allows,2021-02-02T13:52:10Z
issue,51207,Initialize NCCL backend with MPI,https://github.com/pytorch/pytorch/issues/51207,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: mpi, module: nccl, oncall: distributed",scan when time allows,2021-02-02T01:53:45Z
issue,51385,DataParallel copies the model onto GPUs sequentially,https://github.com/pytorch/pytorch/issues/51385,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: data parallel, module: performance",scan when time allows,2021-02-01T18:52:56Z
issue,51440,Support unity 3d,https://github.com/pytorch/pytorch/issues/51440,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-02-01T18:28:49Z
issue,48485,Add wheels for all cuda versions on pypi,https://github.com/pytorch/pytorch/issues/48485,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: binaries,scan when time allows,2021-02-01T17:06:29Z
issue,50650,build from source fail in jeston tx2 Python3.8,https://github.com/pytorch/pytorch/issues/50650,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: arm, module: build",scan when time allows,2021-02-01T15:44:49Z
issue,51044,CuDNN 8 with benchmark=True takes minutes to execute for certain configurations,https://github.com/pytorch/pytorch/issues/51044,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cudnn, module: performance",scan when time allows,2021-01-30T22:31:29Z
issue,38193,Can't Build Pytorch0.4.1 From Source,https://github.com/pytorch/pytorch/issues/38193,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2021-01-30T18:09:30Z
issue,37995,unsqueeze support for named tensors,https://github.com/pytorch/pytorch/issues/37995,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: named tensor,scan when time allows,2021-01-30T18:02:59Z
issue,51393,max_pool2d CPU forward performance is poor,https://github.com/pytorch/pytorch/issues/51393,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: performance, module: pooling",scan when time allows,2021-01-30T02:27:30Z
issue,49409,Add global gradcheck setting,https://github.com/pytorch/pytorch/issues/49409,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2021-01-30T00:08:21Z
issue,51302,"""make install"" skips some required libraries for static build",https://github.com/pytorch/pytorch/issues/51302,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2021-01-29T05:01:19Z
issue,51229,std::type_index doesn't provide reliable equality for getCustomClassTypeMap,https://github.com/pytorch/pytorch/issues/51229,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: internals,scan when time allows,2021-01-29T05:00:35Z
issue,51282,Bind tensor.unflatten to torch.unflatten namespace (torch.flatten exists but not torch.unflatten),https://github.com/pytorch/pytorch/issues/51282,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-01-29T05:00:06Z
issue,32306,[FR] bincount along arbitrary dimension,https://github.com/pytorch/pytorch/issues/32306,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sorting and selection,scan when time allows,2021-01-29T01:12:15Z
issue,43867,torch.bincount beyond 1d arrays,https://github.com/pytorch/pytorch/issues/43867,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-01-29T00:29:47Z
issue,41232,Cleanup git branches,https://github.com/pytorch/pytorch/issues/41232,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-01-28T16:13:58Z
issue,50829,pin_memory leads to GPU memory being used but it doesn't initialize CUDA,https://github.com/pytorch/pytorch/issues/50829,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2021-01-28T10:34:28Z
issue,51003,Proposing new features: Developing Echo State Network layers,https://github.com/pytorch/pytorch/issues/51003,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: embedding,scan when time allows,2021-01-27T02:45:28Z
issue,17902,Conjugate gradient method,https://github.com/pytorch/pytorch/issues/17902,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: derivatives,scan when time allows,2021-01-27T00:17:31Z
issue,9882,Unintuitive reduction of mini-batch loss for NLLLoss,https://github.com/pytorch/pytorch/issues/9882,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: loss, module: nn",scan when time allows,2021-01-26T23:01:46Z
issue,50852,[TorchScript Performance Deep-dive] problems discovered in TS performance deep-dive,https://github.com/pytorch/pytorch/issues/50852,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: performance,scan when time allows,2021-01-26T22:27:45Z
issue,51084,dyld: Symbol not found: _PyBaseObject_Type Referenced from,https://github.com/pytorch/pytorch/issues/51084,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: macos, module: static linking",scan when time allows,2021-01-26T22:20:33Z
issue,39365,Attributes not removed by freeze module,https://github.com/pytorch/pytorch/issues/39365,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2021-01-26T21:36:36Z
issue,50987,Scalar/Tensor arg type for op schemas,https://github.com/pytorch/pytorch/issues/50987,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: codegen,scan when time allows,2021-01-26T18:44:44Z
issue,9753,dataloader stuck at sched_yield =0,https://github.com/pytorch/pytorch/issues/9753,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2021-01-26T18:13:54Z
issue,9893,interaction with FindCUDA causes spurious re-cmakes,https://github.com/pytorch/pytorch/issues/9893,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2021-01-26T17:58:18Z
issue,9912,Stop ifdef'ing out scatter/gather (comm) in libtorch,https://github.com/pytorch/pytorch/issues/9912,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: cuda",scan when time allows,2021-01-26T17:47:17Z
issue,9950,Stop passing inplace/out arguments as (non-const) Tensor& to functions,https://github.com/pytorch/pytorch/issues/9950,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: internals",scan when time allows,2021-01-26T17:14:23Z
issue,9952,[feature request] Add COCOB Optimizer,https://github.com/pytorch/pytorch/issues/9952,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2021-01-26T17:10:27Z
issue,47669,Pytorch 1.7.0 with cuda 11.1.1 and cudnn 8.0.5,https://github.com/pytorch/pytorch/issues/47669,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ci, module: cuda, oncall: releng",scan when time allows,2021-01-26T13:39:18Z
issue,51054,Class-based structured kernels instruction count regression,https://github.com/pytorch/pytorch/issues/51054,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: internals, module: performance, module: structured kernels",scan when time allows,2021-01-25T21:52:00Z
issue,51015,/usr/local/include/google/protobuf/stubs/pbconfig.h:3:37: note: expanded from macro 'GOOGLE_PROTOBUF_HASH_MAP_H',https://github.com/pytorch/pytorch/issues/51015,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: macos",scan when time allows,2021-01-25T15:48:07Z
issue,51008,Suggestion: Assimilate activation images to PyTorch brand color,https://github.com/pytorch/pytorch/issues/51008,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2021-01-25T15:40:26Z
issue,50925,THCCachingAllocator::cuda_free_mutex has no effect,https://github.com/pytorch/pytorch/issues/50925,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: nccl",scan when time allows,2021-01-25T15:22:04Z
issue,50888,Does pytorch support cuda 11.1,https://github.com/pytorch/pytorch/issues/50888,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2021-01-25T14:47:04Z
issue,50219,`requires_grad_` in `no_grad` context returns incorrect value with tensor subclasses,https://github.com/pytorch/pytorch/issues/50219,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2021-01-25T14:25:52Z
issue,50890,"torch.unique(x, dim=1, return_inverse=True) returns inverse for only the last sub-tensor along dim",https://github.com/pytorch/pytorch/issues/50890,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: numpy, module: sorting and selection",scan when time allows,2021-01-25T13:51:39Z
issue,50737,Adding visualizations for indexing and other functions,https://github.com/pytorch/pytorch/issues/50737,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: advanced indexing, module: docs, module: scatter & gather ops",scan when time allows,2021-01-25T09:52:58Z
issue,47109,Pytorch 1.7/Cuda 11.1 binaries,https://github.com/pytorch/pytorch/issues/47109,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: cuda",scan when time allows,2021-01-23T19:09:33Z
issue,50939,dataloader bug need help,https://github.com/pytorch/pytorch/issues/50939,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2021-01-23T00:19:33Z
issue,50944,libpytorch macos build: static library eigen_blas_LIBRARY-NOTFOUND not found,https://github.com/pytorch/pytorch/issues/50944,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2021-01-22T17:53:47Z
issue,4731,Consider disallowing Variables that require grad in NCCL/comm functions,https://github.com/pytorch/pytorch/issues/4731,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: nccl",scan when time allows,2021-01-22T16:40:23Z
issue,35666,[discussion] Generic solutions for too-small-epsilon in FP16 training,https://github.com/pytorch/pytorch/issues/35666,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: docs, module: half",scan when time allows,2021-01-22T15:57:28Z
issue,35797,Backward function causes device error in C++ when changing module's device repeatly.,https://github.com/pytorch/pytorch/issues/35797,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: cpp",scan when time allows,2021-01-22T15:54:01Z
issue,40480,torch.autograd.functional.* for models,https://github.com/pytorch/pytorch/issues/40480,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2021-01-22T15:39:50Z
issue,20058,Implement noise_shape keyword for Dropout layers,https://github.com/pytorch/pytorch/issues/20058,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2021-01-21T22:03:51Z
issue,43556,FP16 gives NaN loss when using pre-trained model,https://github.com/pytorch/pytorch/issues/43556,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision), module: cuda",scan when time allows,2021-01-21T10:00:51Z
issue,50831,Dataloader Prefetch data to GPU by cudaMemPrefetchAsync,https://github.com/pytorch/pytorch/issues/50831,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2021-01-21T04:12:49Z
issue,50303,Way to solve GPU Host Thread Contention?,https://github.com/pytorch/pytorch/issues/50303,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2021-01-21T00:10:13Z
issue,50830,Pull upstream cmake's change in FindCUDA/select_compute_arch,https://github.com/pytorch/pytorch/issues/50830,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cuda",scan when time allows,2021-01-20T20:42:16Z
issue,49949,"RFC: Add a ""see also"" section to operators' documentation",https://github.com/pytorch/pytorch/issues/49949,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2021-01-20T14:27:24Z
issue,9681,[feature request] Support for 0-length sequences in packed_sequences,https://github.com/pytorch/pytorch/issues/9681,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: rnn",scan when time allows,2021-01-20T13:28:05Z
issue,50354,for CNN in fp16 execution time depends on input scale,https://github.com/pytorch/pytorch/issues/50354,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: half, module: nn, module: performance",scan when time allows,2021-01-19T15:23:44Z
issue,14945,[Feature request] create sparse coo matrix w/o index check,https://github.com/pytorch/pytorch/issues/14945,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: performance, module: sparse",scan when time allows,2021-01-19T06:20:02Z
issue,50694,CosineAnnealingWarmRestarts lacks verbose functionality,https://github.com/pytorch/pytorch/issues/50694,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2021-01-18T23:55:08Z
issue,50468,[RFC] Make RRef proxy APIs non-blocking,https://github.com/pytorch/pytorch/issues/50468,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rpc,scan when time allows,2021-01-18T11:01:58Z
issue,9559,[feature request] Rename `Subset` -> `Resample` to reflect wider use,https://github.com/pytorch/pytorch/issues/9559,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2021-01-18T02:01:54Z
issue,9560,Incorrect term in _LRScheduler.,https://github.com/pytorch/pytorch/issues/9560,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2021-01-18T02:00:38Z
issue,50616,Use unified type for distributions.constraint API,https://github.com/pytorch/pytorch/issues/50616,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distributions,scan when time allows,2021-01-18T01:44:51Z
issue,46644,Trying to compile PyTorch for sm_30 fails with `error: identifier “__ldg” is undefined`,https://github.com/pytorch/pytorch/issues/46644,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cuda",scan when time allows,2021-01-16T03:28:19Z
issue,50613,Discourage slow gradchecks,https://github.com/pytorch/pytorch/issues/50613,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: tests",scan when time allows,2021-01-15T21:43:07Z
issue,50473,Make use of eps for numerical stability more consistent,https://github.com/pytorch/pytorch/issues/50473,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: numerical-stability,scan when time allows,2021-01-15T02:44:31Z
issue,50552,test_fn_grad_fft_fftn_cpu_complex128 and test_fn_grad_fft_rfftn_cpu_float64 are failing under TSAN,https://github.com/pytorch/pytorch/issues/50552,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: fft, module: sanitizers, module: tests",scan when time allows,2021-01-14T21:13:06Z
issue,47627,Allow add_embedding to have dict for metadata,https://github.com/pytorch/pytorch/issues/47627,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tensorboard,scan when time allows,2021-01-14T10:15:45Z
issue,49180,torch.utils.data.DistributedSampler allow uneven inputs,https://github.com/pytorch/pytorch/issues/49180,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: data, module: dataloader",scan when time allows,2021-01-14T03:51:09Z
issue,49911,[RFC] Speed up python function and arg serialization in RPC APIs,https://github.com/pytorch/pytorch/issues/49911,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: rpc, oncall: distributed",scan when time allows,2021-01-13T21:48:38Z
issue,48702,provide example for distributed training with iterative dataloaders,https://github.com/pytorch/pytorch/issues/48702,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dataloader, oncall: distributed",scan when time allows,2021-01-13T14:58:22Z
issue,39680,CUDA cannot be found,https://github.com/pytorch/pytorch/issues/39680,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: arm, module: build, module: cuda",scan when time allows,2021-01-13T09:55:28Z
issue,50462,Standalone OSS RPC benchmark,https://github.com/pytorch/pytorch/issues/50462,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: rpc, oncall: distributed",scan when time allows,2021-01-13T07:00:58Z
issue,9381,[doc] many losses still mention size_average in formula,https://github.com/pytorch/pytorch/issues/9381,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: deprecation, module: docs, module: loss, module: nn",scan when time allows,2021-01-13T00:17:37Z
issue,9417,"[feature request] allow forward_pre_hook to ""preprocess"" input",https://github.com/pytorch/pytorch/issues/9417,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-01-12T22:12:25Z
issue,48605,[NNC] Bugs Exposed in Binary Op Testing,https://github.com/pytorch/pytorch/issues/48605,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: bootcamp,scan when time allows,2021-01-12T20:49:15Z
issue,37766,CMake Documentation Issue,https://github.com/pytorch/pytorch/issues/37766,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2021-01-12T08:39:35Z
issue,47476,Reciprocal and reciprocal square root instructions are too inaccurate on ARM64,https://github.com/pytorch/pytorch/issues/47476,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: arm, module: correctness (silent)",scan when time allows,2021-01-12T04:55:42Z
issue,39279,Differentiable Optimizers,https://github.com/pytorch/pytorch/issues/39279,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: optimizer",scan when time allows,2021-01-11T22:30:06Z
issue,35312,Method/constructor which takes as input angle and magnitude and returns a complex tensor,https://github.com/pytorch/pytorch/issues/35312,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: complex,scan when time allows,2021-01-11T21:50:49Z
issue,50366,Optimizer Closure: Enable skip by returning None loss.,https://github.com/pytorch/pytorch/issues/50366,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2021-01-11T16:46:40Z
issue,49724,cure unnecessary NaN values that arise from ±∞ arguments,https://github.com/pytorch/pytorch/issues/49724,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: bootcamp, module: nans and infs, module: numerical-stability",scan when time allows,2021-01-11T13:19:25Z
issue,7359,[feature request] [PyTorch] Dynamic Samplers.,https://github.com/pytorch/pytorch/issues/7359,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2021-01-11T04:39:48Z
issue,50352,torch.tril_indices is incompatible with np.tril_indices,https://github.com/pytorch/pytorch/issues/50352,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: deprecation, module: numpy, module: ux",scan when time allows,2021-01-11T02:50:49Z
issue,50268,CUDA error when using `binary_cross_entropy_with_logits`,https://github.com/pytorch/pytorch/issues/50268,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: memory usage",scan when time allows,2021-01-10T18:19:57Z
issue,50333,Function Request: scipy.stats.pearsonr,https://github.com/pytorch/pytorch/issues/50333,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: numpy,scan when time allows,2021-01-10T11:31:34Z
issue,50336,Function Request: scipy.ndimage.zoom,https://github.com/pytorch/pytorch/issues/50336,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: interpolation, module: numpy",scan when time allows,2021-01-10T11:28:03Z
issue,9186,[pytorch] Make dtype second positional argument of tensor factory methods,https://github.com/pytorch/pytorch/issues/9186,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: numpy, module: pybind, module: ux",scan when time allows,2021-01-10T10:51:02Z
issue,7745,[Caffe2] Align Element-Wise Ops Broadcasting to Numpy,https://github.com/pytorch/pytorch/issues/7745,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-01-10T08:48:32Z
issue,8062,CrossEntropyLoss mishandles weights,https://github.com/pytorch/pytorch/issues/8062,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: loss, module: nn",scan when time allows,2021-01-10T08:30:07Z
issue,50340,Function request: scipy.interpolate.InterpolatedUnivariateSpline,https://github.com/pytorch/pytorch/issues/50340,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: interpolation, module: numpy",scan when time allows,2021-01-10T07:01:08Z
issue,50338,Function request: scipy.interpolate.RegularGridInterpolator,https://github.com/pytorch/pytorch/issues/50338,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: interpolation, module: numpy",scan when time allows,2021-01-10T06:57:42Z
issue,50337,Function request: scipy.interpolate.RectBivariateSpline,https://github.com/pytorch/pytorch/issues/50337,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: interpolation, module: numpy",scan when time allows,2021-01-10T06:55:46Z
issue,46222,Use better tempfile creation mechanism to avoid skip windows test,https://github.com/pytorch/pytorch/issues/46222,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tests,scan when time allows,2021-01-10T06:44:25Z
issue,28258,[jit] TorchScript classes don't work in notebooks,https://github.com/pytorch/pytorch/issues/28258,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2021-01-08T23:29:14Z
issue,9286,Accumulate into accreal instead of real for CPU loss functions,https://github.com/pytorch/pytorch/issues/9286,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: loss, module: nn",scan when time allows,2021-01-08T22:27:17Z
issue,9250,[feature request] SSIM-based cost function as part of the standard set of loss functions,https://github.com/pytorch/pytorch/issues/9250,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: loss,scan when time allows,2021-01-08T18:28:31Z
issue,9193,[Feature Request] Additional torch.nn.LSTM functionality,https://github.com/pytorch/pytorch/issues/9193,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rnn,scan when time allows,2021-01-08T18:27:53Z
issue,49431,clip_grad_norm_ performance regression,https://github.com/pytorch/pytorch/issues/49431,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: norms and normalization, module: performance",scan when time allows,2021-01-08T06:07:14Z
issue,32113,Inconsistent linking flags result in error when building lib/libc10_cuda.so,https://github.com/pytorch/pytorch/issues/32113,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cuda",scan when time allows,2021-01-08T04:15:16Z
issue,47291,ProcessGroupNCCL NCCL lib version mismatch,https://github.com/pytorch/pytorch/issues/47291,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, oncall: distributed",scan when time allows,2021-01-08T00:49:13Z
issue,50213,torch.float_power out= and inplace variant errors on non-matching output dtype instead of casting,https://github.com/pytorch/pytorch/issues/50213,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: type promotion,scan when time allows,2021-01-08T00:28:35Z
issue,50185,CUDACachingAllocator is not GC-aware,https://github.com/pytorch/pytorch/issues/50185,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: memory usage",scan when time allows,2021-01-07T17:59:57Z
issue,50118,torch.where scalar/tensor documentation is unclear and not formatted,https://github.com/pytorch/pytorch/issues/50118,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: sorting and selection",scan when time allows,2021-01-07T17:14:35Z
issue,50190,Calling emptyCache in Conv_v7.cpp causes performance degradation,https://github.com/pytorch/pytorch/issues/50190,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: cudnn",scan when time allows,2021-01-07T16:54:15Z
issue,49865,torch.solve on Jetson is slower than humans,https://github.com/pytorch/pytorch/issues/49865,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: linear algebra",scan when time allows,2021-01-07T08:37:35Z
issue,50090,Torch native_layer_norm OP out-of-bounds access,https://github.com/pytorch/pytorch/issues/50090,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: norms and normalization,scan when time allows,2021-01-07T05:44:15Z
issue,50170,Automated Mixed Precision not documented to work with nn.DataParallel,https://github.com/pytorch/pytorch/issues/50170,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: data parallel,scan when time allows,2021-01-07T04:48:40Z
issue,50136,Inconsistent Variable Naming in FindTorch.cmake,https://github.com/pytorch/pytorch/issues/50136,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cpp",scan when time allows,2021-01-07T01:16:06Z
issue,49849,[dataloader] RuntimeError: Too many open files when yielding integers,https://github.com/pytorch/pytorch/issues/49849,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2021-01-06T21:17:40Z
issue,49992,"[docs] nn.modules pages should mention corresponding functional versions, e.g. for nn.Hardshrink",https://github.com/pytorch/pytorch/issues/49992,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: nn",scan when time allows,2021-01-06T13:00:30Z
issue,49885,LR scheduler `get_lr()` bug,https://github.com/pytorch/pytorch/issues/49885,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2021-01-06T12:12:26Z
issue,50094,Broken LAPACK links in the documentation,https://github.com/pytorch/pytorch/issues/50094,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: linear algebra",scan when time allows,2021-01-06T04:22:36Z
issue,49207,Provide a mechanism to limit the workspace size of cudnn convolution,https://github.com/pytorch/pytorch/issues/49207,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: cudnn",scan when time allows,2021-01-05T23:26:55Z
issue,50076,resize_ documentation does not match implementation when memory_format is given,https://github.com/pytorch/pytorch/issues/50076,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2021-01-05T19:36:04Z
issue,9171,Mismatch in behaviour of WeightedRandomSampler and other samplers,https://github.com/pytorch/pytorch/issues/9171,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2021-01-05T15:43:13Z
issue,9133,[Caffe2] Error running net train when running resnet50,https://github.com/pytorch/pytorch/issues/9133,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-01-05T15:42:34Z
issue,8821,Storages still use legacy printing,https://github.com/pytorch/pytorch/issues/8821,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: printing,scan when time allows,2021-01-04T18:34:58Z
issue,16703,split_with_sizes should accept a LongTensor as the split_sizes parameter,https://github.com/pytorch/pytorch/issues/16703,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: numpy,scan when time allows,2021-01-04T18:05:52Z
issue,49673,Is there any upper-level API to disable the detection of the weight’s version,https://github.com/pytorch/pytorch/issues/49673,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2021-01-04T17:00:55Z
issue,50031,Inconsistent function name between stub and implementation in `torch.optim.swa_utils`,https://github.com/pytorch/pytorch/issues/50031,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: typing,scan when time allows,2021-01-04T16:38:51Z
issue,37713,Builtin FusedLayerNorm is slower than apex one,https://github.com/pytorch/pytorch/issues/37713,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: performance",scan when time allows,2021-01-02T23:46:13Z
issue,34064,pip --requirement installs incorrect CUDA version,https://github.com/pytorch/pytorch/issues/34064,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: dependency bug, oncall: releng",scan when time allows,2020-12-31T18:17:37Z
issue,8874,Issues with dynamically created grad_fn for views,https://github.com/pytorch/pytorch/issues/8874,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: viewing and reshaping",scan when time allows,2020-12-30T13:13:04Z
issue,49784,Compatibility with Raspberry Pi,https://github.com/pytorch/pytorch/issues/49784,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2020-12-29T21:43:35Z
issue,49874,Accessibility Coding Needed,https://github.com/pytorch/pytorch/issues/49874,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: doc infra,scan when time allows,2020-12-29T20:31:03Z
issue,49873,Accessibility of Website for Low Vision Users,https://github.com/pytorch/pytorch/issues/49873,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: doc infra,scan when time allows,2020-12-29T20:28:52Z
issue,49476,Use METH_FASTCALL protocol in Python bindings,https://github.com/pytorch/pytorch/issues/49476,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: pybind,scan when time allows,2020-12-29T20:21:40Z
issue,49891,Support DDP find_unused_parameters=True mode when combined with Pipe,https://github.com/pytorch/pytorch/issues/49891,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ddp, module: rpc, oncall: distributed",scan when time allows,2020-12-29T19:41:21Z
issue,8871,Use target_compile_options to set warning flags,https://github.com/pytorch/pytorch/issues/8871,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2020-12-29T18:17:44Z
issue,9032,Lint check for non-Unicode characters in diffs / Unicode characters without coding,https://github.com/pytorch/pytorch/issues/9032,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: lint,scan when time allows,2020-12-29T18:04:25Z
issue,38684,Add Plackett-Luce distribution,https://github.com/pytorch/pytorch/issues/38684,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-12-29T00:15:40Z
issue,46971,Yolov5 detect.py(ingerence) Error:Torch.nn.modules.module.ModuleAttributeError: 'Hardswish' object has no attribute 'inplace',https://github.com/pytorch/pytorch/issues/46971,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: pickle",scan when time allows,2020-12-28T18:06:31Z
issue,49416,A few years after #701 and PyTorch is still using implicit __all__ imports.,https://github.com/pytorch/pytorch/issues/49416,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: codegen, module: pybind, module: typing",scan when time allows,2020-12-28T16:20:28Z
issue,49224,Torch abs op crashes on -128 int8 tensor with ASAN,https://github.com/pytorch/pytorch/issues/49224,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: error checking,scan when time allows,2020-12-28T16:17:39Z
issue,43672,Make `torch.Generator` picklable,https://github.com/pytorch/pytorch/issues/43672,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: serialization,scan when time allows,2020-12-28T02:49:20Z
issue,49539,Improve AssertionError for torch.nn.functional.pad 'replicate',https://github.com/pytorch/pytorch/issues/49539,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, module: nn",scan when time allows,2020-12-25T00:20:35Z
issue,49727,RandomSampler generator created in every iteration,https://github.com/pytorch/pytorch/issues/49727,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2020-12-23T21:07:08Z
issue,49723,Flattening nn.Parameters while maintaining gradients from neural network forward pass,https://github.com/pytorch/pytorch/issues/49723,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: optimizer",scan when time allows,2020-12-23T03:35:15Z
issue,49677,"File ""setup.py"", line 773:subprocess.CalledProcessError: Command '['cmake', '--build', '.', '--target', 'install', '--config', 'Release', '--', '-j', '8']' returned non-zero exit status 2",https://github.com/pytorch/pytorch/issues/49677,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cuda",scan when time allows,2020-12-22T22:07:49Z
issue,49672,Model passed through queue hangs while loading in sub process,https://github.com/pytorch/pytorch/issues/49672,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: multiprocessing,scan when time allows,2020-12-22T19:47:09Z
issue,49716,Two independent future chains in PowerSGD cannot be kicked off asynchronously,https://github.com/pytorch/pytorch/issues/49716,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2020-12-22T18:54:42Z
issue,42459,[JIT] Runtime error when backpropagating through input gradient,https://github.com/pytorch/pytorch/issues/42459,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-12-22T03:27:21Z
issue,8602,Come with a better strategy for TensorArg (error reporting),https://github.com/pytorch/pytorch/issues/8602,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: internals, module: performance",scan when time allows,2020-12-21T20:02:28Z
issue,49642,Turn deprecation warnings into errors in CI,https://github.com/pytorch/pytorch/issues/49642,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ci, module: tests",scan when time allows,2020-12-21T18:28:46Z
issue,30388,Increasing memory usage on CPU,https://github.com/pytorch/pytorch/issues/30388,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: memory usage, module: performance",scan when time allows,2020-12-21T17:09:24Z
issue,49398,Can we add try-except for list/slice indices even when auto_collation is True?,https://github.com/pytorch/pytorch/issues/49398,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: data,scan when time allows,2020-12-21T07:27:05Z
issue,49653,"torch.cuda.amp, example with 20% memory increase compared to apex/amp",https://github.com/pytorch/pytorch/issues/49653,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision),scan when time allows,2020-12-20T16:54:08Z
issue,25814,(PyTorch1.1 and 1.2) RuntimeError: Can't detach views in-place. Use detach() instead,https://github.com/pytorch/pytorch/issues/25814,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: optimizer",scan when time allows,2020-12-20T12:08:42Z
issue,49282,Bug when combining AMP with channels_last memory format,https://github.com/pytorch/pytorch/issues/49282,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision), module: cuda, module: memory format",scan when time allows,2020-12-20T10:47:28Z
issue,49427,Inserting named tensor into other fails,https://github.com/pytorch/pytorch/issues/49427,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: named tensor,scan when time allows,2020-12-19T10:13:01Z
issue,40659,[typing] overly restrictive List[int],https://github.com/pytorch/pytorch/issues/40659,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: typing,scan when time allows,2020-12-18T20:24:53Z
issue,42460,Training with AMP gives overflow warning,https://github.com/pytorch/pytorch/issues/42460,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision), module: cuda",scan when time allows,2020-12-18T10:07:35Z
issue,49023,Automatically rerun tests with CUDA_LAUNCH_BLOCKING=1 when they fail with CUDA errors in CI,https://github.com/pytorch/pytorch/issues/49023,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: tests",scan when time allows,2020-12-18T07:28:42Z
issue,49475,Write link in README.md,https://github.com/pytorch/pytorch/issues/49475,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2020-12-18T06:17:09Z
issue,35180,hasSpecialCase INTERNAL ASSERT FAILED: We don't have an op for aten::to but it isn't a special case.,https://github.com/pytorch/pytorch/issues/35180,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-12-17T09:04:24Z
issue,49321,Backward hangs with DDP during training.,https://github.com/pytorch/pytorch/issues/49321,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ddp, oncall: distributed",scan when time allows,2020-12-17T00:32:55Z
issue,49477,Callbacks may not be automatically synchronized in a single NCCL future chain,https://github.com/pytorch/pytorch/issues/49477,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2020-12-16T19:15:07Z
issue,49320,how could I print the log in source code,https://github.com/pytorch/pytorch/issues/49320,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: logging",scan when time allows,2020-12-16T16:04:01Z
issue,8561,cleanup BLAS detection,https://github.com/pytorch/pytorch/issues/8561,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: linear algebra",scan when time allows,2020-12-16T15:27:30Z
issue,8386,Properly release NCCL resources,https://github.com/pytorch/pytorch/issues/8386,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nccl,scan when time allows,2020-12-16T15:22:16Z
issue,44182,Support keep stride for neg with requires_grad=False,https://github.com/pytorch/pytorch/issues/44182,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: memory usage, module: numpy",scan when time allows,2020-12-15T20:29:55Z
issue,49332,global_pruning costs memory after model is trained,https://github.com/pytorch/pytorch/issues/49332,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: memory usage, module: pruning",scan when time allows,2020-12-15T15:21:14Z
issue,49347,[pre-commit hook] fails on `BLK100 Black would make changes.`,https://github.com/pytorch/pytorch/issues/49347,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2020-12-15T05:40:08Z
issue,49300,is this a typo in optimizer.pyi ? it says `statue` instead of `state`,https://github.com/pytorch/pytorch/issues/49300,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2020-12-15T04:35:59Z
issue,8097,manager.cpp:64: undefined reference to `shm_open (when building with GCC 5.x (sic)),https://github.com/pytorch/pytorch/issues/8097,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2020-12-11T16:38:34Z
issue,8107,Export CC is ignored when I build pytorch,https://github.com/pytorch/pytorch/issues/8107,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2020-12-11T16:37:51Z
issue,17920,Build compact libtorch from source with cmake,https://github.com/pytorch/pytorch/issues/17920,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cpp",scan when time allows,2020-12-11T13:24:05Z
issue,49217,Bug in profiler on CI machines when profiling NCCL distributed calls,https://github.com/pytorch/pytorch/issues/49217,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: c10d, oncall: distributed",scan when time allows,2020-12-11T04:42:33Z
issue,49076,Add List Type Data through add_hparam method in torch.utils.tensorboard.SummaryWriter,https://github.com/pytorch/pytorch/issues/49076,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-12-11T02:02:48Z
issue,49178,[NNC] Some ops have type promotion logic which adds extra casts & does compute in different dtype than eager,https://github.com/pytorch/pytorch/issues/49178,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: bootcamp,scan when time allows,2020-12-10T22:53:44Z
issue,49177,Investigate `torch.linalg.norm` performance,https://github.com/pytorch/pytorch/issues/49177,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: linear algebra, module: performance",scan when time allows,2020-12-10T22:41:06Z
issue,49166,Implement fill_out on complex,https://github.com/pytorch/pytorch/issues/49166,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-12-10T20:23:07Z
issue,49082,swa_utils.bn_update is too opinionated in how it calls the model,https://github.com/pytorch/pytorch/issues/49082,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2020-12-10T20:15:25Z
issue,49139,"Hello everyone, why does libtorch half precision float 16 leak memory?大家好，为什么libtorch半精度float16会内存泄漏？",https://github.com/pytorch/pytorch/issues/49139,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: memory usage,scan when time allows,2020-12-10T20:11:32Z
issue,45493,Build LibTorch for cpu using OpenBLAS: Had to manually remove a path in Caffe2Targets.cmake,https://github.com/pytorch/pytorch/issues/45493,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2020-12-10T15:23:58Z
issue,49133,Unknown failure for test_reference_numerics_sinc_cpu_complex64,https://github.com/pytorch/pytorch/issues/49133,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: complex, module: trigonometric functions",scan when time allows,2020-12-10T06:24:53Z
issue,48526,CARU: A Content-Adaptive Recurrent Unit for the Transition of Hidden State in NLP,https://github.com/pytorch/pytorch/issues/48526,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: rnn",scan when time allows,2020-12-09T08:30:45Z
issue,48945,CUDA error: illegal memory access Conv3d,https://github.com/pytorch/pytorch/issues/48945,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: cudnn",scan when time allows,2020-12-08T17:30:20Z
issue,48998,Torch1.2 can't be downloaded!,https://github.com/pytorch/pytorch/issues/48998,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-12-08T17:29:17Z
issue,48645,Can sublist a nn.Sequential subclass,https://github.com/pytorch/pytorch/issues/48645,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2020-12-08T17:06:49Z
issue,48641,"[bug] `torch.{sinh, cosh}`: Incorrect values for vectorized path",https://github.com/pytorch/pytorch/issues/48641,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: nans and infs, module: sleef, module: trigonometric functions",scan when time allows,2020-12-08T11:09:05Z
issue,9282,[gradcheck] warn about the case that mulitple inputs share storage,https://github.com/pytorch/pytorch/issues/9282,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2020-12-07T20:07:18Z
issue,48878,Libtorch: nvrtc: error: invalid value for --gpu-architecture (-arch),https://github.com/pytorch/pytorch/issues/48878,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-12-07T19:59:08Z
issue,48929,Domain Specific Batch Normalization,https://github.com/pytorch/pytorch/issues/48929,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-12-07T19:28:53Z
issue,28245,PyTorch RPC should expose critical metrics to the application.,https://github.com/pytorch/pytorch/issues/28245,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rpc,scan when time allows,2020-12-07T02:32:19Z
issue,48892,is_non_overlapping_and_dense() does not error out for sparse tensors,https://github.com/pytorch/pytorch/issues/48892,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2020-12-06T03:42:03Z
issue,46564,Clean up request_callback_no_python.cpp,https://github.com/pytorch/pytorch/issues/46564,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rpc,scan when time allows,2020-12-04T23:45:36Z
issue,48291,pytorch is not linked with support for cuda devices,https://github.com/pytorch/pytorch/issues/48291,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cuda",scan when time allows,2020-12-04T12:41:35Z
issue,48706,test_inverse setup is flaky using MKL>=2020.1 on certain CPUs and fails on CUDA,https://github.com/pytorch/pytorch/issues/48706,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: mkl",scan when time allows,2020-12-03T17:51:31Z
issue,7944,Better error message in DataChannelTCP::_receive,https://github.com/pytorch/pytorch/issues/7944,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: backend,scan when time allows,2020-12-03T16:30:46Z
issue,8035,[PyTorch] EmbeddingBag comparison vs Embedding fails w/ small max_norm on CUDA,https://github.com/pytorch/pytorch/issues/8035,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: norms and normalization,scan when time allows,2020-12-03T16:21:34Z
issue,26218,Default adam epsilon to 1e-7 when on fp16,https://github.com/pytorch/pytorch/issues/26218,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: half, module: numerical-stability, module: optimizer",scan when time allows,2020-12-03T07:53:28Z
issue,48466,[master] new default signature: op=<ReduceOp.SUM: 0>from `op=<ReduceOp.SUM>,https://github.com/pytorch/pytorch/issues/48466,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, oncall: distributed",scan when time allows,2020-12-03T01:03:58Z
issue,48684,Remove remaining native:: references from non-native ATen,https://github.com/pytorch/pytorch/issues/48684,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: internals,scan when time allows,2020-12-02T14:54:06Z
issue,7834,torch.Tensor.new() disappeared in 0.4 doc,https://github.com/pytorch/pytorch/issues/7834,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: deprecation, module: docs, module: tensor creation",scan when time allows,2020-12-02T07:14:08Z
issue,48649,[docs] A single additional summary page for diag* methods,https://github.com/pytorch/pytorch/issues/48649,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2020-12-01T22:44:16Z
issue,48621,[FR] loss reduction argument accepts None,https://github.com/pytorch/pytorch/issues/48621,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: loss,scan when time allows,2020-12-01T22:37:47Z
issue,48648,Confused on project website pointer.,https://github.com/pytorch/pytorch/issues/48648,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2020-12-01T22:33:36Z
issue,48650,[feature request] Keyword-only device argument (and maybe dtype) for torch.meshgrid,https://github.com/pytorch/pytorch/issues/48650,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: internals,scan when time allows,2020-12-01T22:31:47Z
issue,48667,Operator registration doesn't work with noexcept functions on some compilers (c10::guts::is_function_type rejects noexcept),https://github.com/pytorch/pytorch/issues/48667,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: internals,scan when time allows,2020-12-01T22:28:17Z
issue,48153,More robust list comprehension,https://github.com/pytorch/pytorch/issues/48153,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-12-01T18:49:04Z
issue,48548,at::size documentation conflict,https://github.com/pytorch/pytorch/issues/48548,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: docs",scan when time allows,2020-12-01T16:35:55Z
issue,46464,[META] Assert that expected libraries appear in libtorch,https://github.com/pytorch/pytorch/issues/46464,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: static linking,scan when time allows,2020-12-01T15:54:54Z
issue,48083,CPU Tensor with Python MP Freezing in Docker Container,https://github.com/pytorch/pytorch/issues/48083,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docker, module: multiprocessing",scan when time allows,2020-12-01T15:41:48Z
issue,48587,[feature request] Give `torch.cholesky` an optional fallback to test whether the tensor is positive definite,https://github.com/pytorch/pytorch/issues/48587,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: linear algebra,scan when time allows,2020-12-01T07:56:05Z
issue,48353,[FR] bool tensor should support basic arithmetics,https://github.com/pytorch/pytorch/issues/48353,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: boolean tensor, module: numpy, module: type promotion",scan when time allows,2020-11-30T18:09:25Z
issue,48235,Softplus differs depending on number of elements in tensor,https://github.com/pytorch/pytorch/issues/48235,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2020-11-30T18:04:45Z
issue,48281,Batched inplace mm changes stride when out size is correct,https://github.com/pytorch/pytorch/issues/48281,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: linear algebra, module: memory format",scan when time allows,2020-11-30T17:52:49Z
issue,44533,“doxygenfunction: Unable to resolve multiple matches for function ...” in C++ documentation,https://github.com/pytorch/pytorch/issues/44533,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: docs",scan when time allows,2020-11-30T17:43:43Z
issue,48518,C++ optimizer check for duplicate parameters,https://github.com/pytorch/pytorch/issues/48518,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2020-11-30T15:40:11Z
issue,48377,Management of Gpu memory to avoid memory errors,https://github.com/pytorch/pytorch/issues/48377,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: memory usage,scan when time allows,2020-11-27T16:05:02Z
issue,48366,Add Go to high level interface,https://github.com/pytorch/pytorch/issues/48366,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-11-27T07:27:03Z
issue,48450,"Qt 5.14.2, libtorch1.7.0 cuda10.2，Error after using header file",https://github.com/pytorch/pytorch/issues/48450,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: static linking,scan when time allows,2020-11-27T07:13:50Z
issue,48511,Ambiguous RuntimeError raised when device of input PackedSequence does not match with nn.LSTM's device,https://github.com/pytorch/pytorch/issues/48511,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: rnn",scan when time allows,2020-11-27T07:01:05Z
issue,7900,Deprecate torch.Tensor,https://github.com/pytorch/pytorch/issues/7900,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: deprecation, module: tensor creation",scan when time allows,2020-11-27T06:58:46Z
issue,48306,[doc] example for pairwise distance matrix,https://github.com/pytorch/pytorch/issues/48306,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: distance functions, module: docs, module: nn",scan when time allows,2020-11-26T23:17:22Z
issue,48305,[RFC] CUDA-aware future for distributed,https://github.com/pytorch/pytorch/issues/48305,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: c10d, module: rpc, oncall: distributed",scan when time allows,2020-11-26T19:51:26Z
issue,48457,compile error immintrin.h,https://github.com/pytorch/pytorch/issues/48457,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2020-11-26T15:44:25Z
issue,46202,pytorch not uploading data to tensorboard page,https://github.com/pytorch/pytorch/issues/46202,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tensorboard,scan when time allows,2020-11-26T07:31:09Z
issue,28218,"index_sub, index_mul and index_div",https://github.com/pytorch/pytorch/issues/28218,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-11-26T06:30:15Z
issue,48338,torch.float128 datatype,https://github.com/pytorch/pytorch/issues/48338,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: numpy,scan when time allows,2020-11-26T00:01:29Z
issue,7773,[feature request] Add Local Contrast Normalization,https://github.com/pytorch/pytorch/issues/7773,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2020-11-25T23:51:21Z
issue,7865,[PyTorch] weight tensor dimension assumption,https://github.com/pytorch/pytorch/issues/7865,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2020-11-25T23:47:20Z
issue,44248,Additional clutter memory allocated on GPU 0 when training on GPU 1+,https://github.com/pytorch/pytorch/issues/44248,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2020-11-25T22:11:27Z
issue,34799,Is it possible to run an object detection android app by using Pytorch Mobile?,https://github.com/pytorch/pytorch/issues/34799,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: android,scan when time allows,2020-11-25T14:48:48Z
issue,22454,Tracing an RNN does not support torch.nn.utils.rnn.PackedSequence as input,https://github.com/pytorch/pytorch/issues/22454,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-11-25T11:40:13Z
issue,33463,torch._C.Node.scopeName() missing in pytorch 1.4,https://github.com/pytorch/pytorch/issues/33463,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-11-25T08:26:02Z
issue,25004,[C++] Call find_package(Torch REQUIRED) more than one time in downstream project causes CMake configuration error,https://github.com/pytorch/pytorch/issues/25004,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2020-11-25T08:19:25Z
issue,22791,How to use mpi backend without CUDA_aware,https://github.com/pytorch/pytorch/issues/22791,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: mpi,scan when time allows,2020-11-25T06:05:04Z
issue,48202,"""Found no NVIDIA driver"" produces too much log",https://github.com/pytorch/pytorch/issues/48202,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2020-11-23T17:05:28Z
issue,7740,Conv3D can be optimized for cases when kernel is spatial (probably),https://github.com/pytorch/pytorch/issues/7740,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: performance",scan when time allows,2020-11-20T16:53:07Z
issue,45459,Backward for sparse tensor item select does not work,https://github.com/pytorch/pytorch/issues/45459,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: sparse",scan when time allows,2020-11-20T14:48:34Z
issue,47307,[Feature request] Stochastic Frank-Wolfe optimizer,https://github.com/pytorch/pytorch/issues/47307,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2020-11-20T08:21:57Z
issue,33785,Deepcopy fails with nn.parallel.replicate,https://github.com/pytorch/pytorch/issues/33785,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: data parallel,scan when time allows,2020-11-20T02:12:30Z
issue,48167,[libtorch] Build shared library with libtorch and compile a static library,https://github.com/pytorch/pytorch/issues/48167,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: static linking,scan when time allows,2020-11-20T01:59:49Z
issue,48227,Modifying values() tensor of COO tensor requiring grad throws an odd error message,https://github.com/pytorch/pytorch/issues/48227,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: sparse",scan when time allows,2020-11-19T22:33:18Z
issue,33628,Remove .data subset1 for fixathon,https://github.com/pytorch/pytorch/issues/33628,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2020-11-19T03:36:25Z
issue,28742,torch.as_tensor fails to create named tensors,https://github.com/pytorch/pytorch/issues/28742,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: named tensor, module: tensor creation",scan when time allows,2020-11-18T15:49:40Z
issue,38765,LibTorch 1.5.0 not supporting GLIBC < 2.23,https://github.com/pytorch/pytorch/issues/38765,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cpp",scan when time allows,2020-11-18T11:58:29Z
issue,47906,amp for custom op,https://github.com/pytorch/pytorch/issues/47906,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision), module: custom-operators",scan when time allows,2020-11-18T01:01:16Z
issue,7610,Inconsistent interactions of PyTorch tensors and NumPy ops,https://github.com/pytorch/pytorch/issues/7610,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: numpy,scan when time allows,2020-11-17T19:46:16Z
issue,48064,FAILED: test_api/CMakeFiles/test_api.dir/dataloader.cpp.o,https://github.com/pytorch/pytorch/issues/48064,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2020-11-17T16:24:01Z
issue,48088,[dataloader] Worker threads to print the signal they received before they die,https://github.com/pytorch/pytorch/issues/48088,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2020-11-17T15:45:28Z
issue,47588,Can we apply Weight normalization hook to a method other than `forward`?,https://github.com/pytorch/pytorch/issues/47588,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: norms and normalization",scan when time allows,2020-11-17T08:12:43Z
issue,47841,Factorial & Binomial Coefficient,https://github.com/pytorch/pytorch/issues/47841,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: numpy,scan when time allows,2020-11-17T06:48:17Z
issue,47954,[Feature] torch.bitwise_where,https://github.com/pytorch/pytorch/issues/47954,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-11-17T01:56:13Z
issue,48033,Missing -fopenmp when used in torchvision,https://github.com/pytorch/pytorch/issues/48033,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: build warnings, module: cpp-extensions, module: openmp",scan when time allows,2020-11-16T20:16:02Z
issue,23434,einsum equation with conditional mask works in numpy but not in PyTorch,https://github.com/pytorch/pytorch/issues/23434,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: linear algebra, module: type promotion",scan when time allows,2020-11-16T18:18:20Z
issue,45447,Retire usages of CUDA_tensor_apply helpers in ATen,https://github.com/pytorch/pytorch/issues/45447,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: internals",scan when time allows,2020-11-16T18:00:07Z
issue,45578,Why is RTX3080 slower than RTX2020-Ti?,https://github.com/pytorch/pytorch/issues/45578,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: performance",scan when time allows,2020-11-16T17:54:21Z
issue,47912,Code becomes more than x20 slower after upgrading torch version,https://github.com/pytorch/pytorch/issues/47912,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: performance,scan when time allows,2020-11-16T17:10:51Z
issue,47901,"Get forward output, grad, hessian all at once",https://github.com/pytorch/pytorch/issues/47901,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2020-11-16T16:25:46Z
issue,7801,Checkpointing is slow on nn.DataParallel models,https://github.com/pytorch/pytorch/issues/7801,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: activation checkpointing, module: performance",scan when time allows,2020-11-15T04:54:57Z
issue,25714,Backtrace prints many <unknown function>,https://github.com/pytorch/pytorch/issues/25714,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2020-11-14T15:13:47Z
issue,15378,Projective Transformation grid generator,https://github.com/pytorch/pytorch/issues/15378,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-11-14T05:39:45Z
issue,47824,"DDP error, when I use ""if"" branch in forward function.",https://github.com/pytorch/pytorch/issues/47824,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: ddp",scan when time allows,2020-11-14T04:55:55Z
issue,47904,Memory leak in nn.MaxPool2d layer when run on iOS,https://github.com/pytorch/pytorch/issues/47904,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ios, module: memory usage",scan when time allows,2020-11-14T00:13:33Z
issue,46918,Jacobians computed by autograd.functional.jacobian with compute_graph sometimes set requires_grad True,https://github.com/pytorch/pytorch/issues/46918,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2020-11-13T17:37:02Z
issue,47751,Make libtorch modular,https://github.com/pytorch/pytorch/issues/47751,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: build, module: selective build",scan when time allows,2020-11-13T17:28:17Z
issue,7353,"Windows MAGMA binary requires explicit linking against MKL LAPACK, or it will silently give incorrect results",https://github.com/pytorch/pytorch/issues/7353,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cuda",scan when time allows,2020-11-13T16:07:06Z
issue,7352,Check for F2C convention (for blas) at runtime,https://github.com/pytorch/pytorch/issues/7352,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2020-11-13T16:06:34Z
issue,9375,[Caffe2] Not handle attribute 'spatial' of ONNX operator 'BatchNormalization',https://github.com/pytorch/pytorch/issues/9375,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-11-13T07:16:03Z
issue,30867,save model definition file,https://github.com/pytorch/pytorch/issues/30867,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: serialization,scan when time allows,2020-11-13T02:42:38Z
issue,47753,"Attempted to read a PyTorch file with version 3, but the maximum supported version for reading is 2",https://github.com/pytorch/pytorch/issues/47753,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: serialization,scan when time allows,2020-11-12T15:21:21Z
issue,47649,error: reference to __host__ function 'parallel_for<thrust::cuda_cub::for_each_f...' in __host__ __device__ function,https://github.com/pytorch/pytorch/issues/47649,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cuda",scan when time allows,2020-11-12T07:05:42Z
issue,47699,DCNv2 Layers,https://github.com/pytorch/pytorch/issues/47699,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2020-11-12T00:24:38Z
issue,47610,Input dimension check for `torch.gather`,https://github.com/pytorch/pytorch/issues/47610,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: error checking,scan when time allows,2020-11-11T22:43:12Z
issue,47609,Error while trying to load a pretrained ResNet34 VTN (Video Transformer Network) model in Android,https://github.com/pytorch/pytorch/issues/47609,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: mobile,scan when time allows,2020-11-11T22:07:05Z
issue,47739,Conda package should install libtorch in the standard location,https://github.com/pytorch/pytorch/issues/47739,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: build",scan when time allows,2020-11-11T21:54:12Z
issue,47681,How to install Pytorch on AIX7.2 without internet access?,https://github.com/pytorch/pytorch/issues/47681,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2020-11-11T02:05:42Z
issue,47591,PyTorch1.3.1 Can not using namespace torch::indexing,https://github.com/pytorch/pytorch/issues/47591,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp,scan when time allows,2020-11-10T05:59:28Z
issue,7342,[feature request] [PyTorch] More flexible optimizer API,https://github.com/pytorch/pytorch/issues/7342,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2020-11-10T05:25:34Z
issue,7330,Issue when importing both retro (from OpenAI) and torch,https://github.com/pytorch/pytorch/issues/7330,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: pybind,scan when time allows,2020-11-10T05:24:58Z
issue,46902,How to use clang as a cuda compiler instead of nvcc?,https://github.com/pytorch/pytorch/issues/46902,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2020-11-10T03:48:41Z
issue,47557,Ninja-built CUDAExtension build process ignores changed #include dependencies,https://github.com/pytorch/pytorch/issues/47557,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2020-11-09T22:52:49Z
issue,7423,Multi queue for dataloader when workers > 1,https://github.com/pytorch/pytorch/issues/7423,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2020-11-09T22:33:00Z
issue,7358,DataParallel on list inputs,https://github.com/pytorch/pytorch/issues/7358,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: data parallel,scan when time allows,2020-11-09T22:31:39Z
issue,7365,nn.DataParallel fills None grads with 0,https://github.com/pytorch/pytorch/issues/7365,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: data parallel,scan when time allows,2020-11-09T22:31:26Z
issue,47563,Handling multiple large-scale datasets efficiently,https://github.com/pytorch/pytorch/issues/47563,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2020-11-09T18:21:16Z
issue,47560,Minumul LR is never reached in,https://github.com/pytorch/pytorch/issues/47560,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2020-11-09T16:11:14Z
issue,42173,Exclusive flag in torch.cumsum,https://github.com/pytorch/pytorch/issues/42173,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-11-08T19:05:12Z
issue,7546,[Caffe2] Build broken on macOS High Sierra: can't find sys headers in /usr/local/include.,https://github.com/pytorch/pytorch/issues/7546,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-11-08T04:58:38Z
issue,3818,Memory leak when doing backward with grad as yourself,https://github.com/pytorch/pytorch/issues/3818,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: memory usage",scan when time allows,2020-11-07T19:33:24Z
issue,38630,Torchvision error TypeError: _resolve_type_from_object(),https://github.com/pytorch/pytorch/issues/38630,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: vision,scan when time allows,2020-11-07T08:46:06Z
issue,47447,"thread blocked when moving a tensor from GPU to CPU, by calling the function .cpu() in pytorch. This kind of block can be stop by any window event like mouse moving/clicking or keyboard pressing.",https://github.com/pytorch/pytorch/issues/47447,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: cuda",scan when time allows,2020-11-06T05:38:18Z
issue,47357,Conv1D formula in docs is wrong,https://github.com/pytorch/pytorch/issues/47357,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: nn",scan when time allows,2020-11-06T05:36:02Z
issue,47341,ConvTranspose1d groups=channels is very slow!!!,https://github.com/pytorch/pytorch/issues/47341,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: cudnn",scan when time allows,2020-11-05T16:26:57Z
issue,46807,[cudatoolkit 11.0] segfaults,https://github.com/pytorch/pytorch/issues/46807,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cuda",scan when time allows,2020-11-05T16:15:34Z
issue,41142,Support for More Loss Functions,https://github.com/pytorch/pytorch/issues/41142,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: loss,scan when time allows,2020-11-04T19:04:29Z
issue,26097,[RFC] TensorBoard extensions and improvements for PyTorch,https://github.com/pytorch/pytorch/issues/26097,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tensorboard,scan when time allows,2020-11-04T16:26:39Z
issue,40968,Pytorch 1.4 compilation hangs on AMD Epyc,https://github.com/pytorch/pytorch/issues/40968,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: vectorization",scan when time allows,2020-11-03T18:54:57Z
issue,35116,JIT does not support class instance attribute type annotation,https://github.com/pytorch/pytorch/issues/35116,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-11-03T16:01:09Z
issue,15918,Broadcasting and Additional Dimensions for pairwise_distance,https://github.com/pytorch/pytorch/issues/15918,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: distance functions, module: nn",scan when time allows,2020-11-03T05:39:42Z
issue,7214,Do not put system paths in RPATH,https://github.com/pytorch/pytorch/issues/7214,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2020-11-03T04:22:59Z
issue,47208,[RFC] Add test execution time analysis CI workflow,https://github.com/pytorch/pytorch/issues/47208,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2020-11-02T18:43:06Z
issue,25150,Problems with install python from source,https://github.com/pytorch/pytorch/issues/25150,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2020-11-02T18:30:21Z
issue,46248,"""distributed"" NCCL tests fail when having more than 3 GPUs",https://github.com/pytorch/pytorch/issues/46248,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ddp, module: nccl, module: tests, oncall: distributed",scan when time allows,2020-11-02T15:14:35Z
issue,32531,Batched torch.eig() and gradient of torch.eig() for real eigenvalues,https://github.com/pytorch/pytorch/issues/32531,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: batching, module: linear algebra",scan when time allows,2020-11-02T14:34:18Z
issue,35535,Apple Review Rejected. ITMS-90338: Non-public API usage In Pytorch For iOS,https://github.com/pytorch/pytorch/issues/35535,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ios,scan when time allows,2020-11-02T05:30:57Z
issue,46947,Add Optimistic Mirror Descent (OMD),https://github.com/pytorch/pytorch/issues/46947,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2020-11-02T03:47:24Z
issue,7032,[feature request] norm argument for RNNCells,https://github.com/pytorch/pytorch/issues/7032,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rnn,scan when time allows,2020-10-30T20:20:34Z
issue,3823,Fuse bias to CuDNN convolution,https://github.com/pytorch/pytorch/issues/3823,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-10-30T19:30:17Z
issue,47111,THCudaShutdown should be called before THCState_free,https://github.com/pytorch/pytorch/issues/47111,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: memory usage",scan when time allows,2020-10-30T13:30:56Z
issue,46428,send_object/recv_object APIs for c10d,https://github.com/pytorch/pytorch/issues/46428,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: c10d, oncall: distributed",scan when time allows,2020-10-29T17:16:55Z
issue,46999,"torch.sum(tensor(2.), dim=0) (and probably other reduction functions) doesn't make sense",https://github.com/pytorch/pytorch/issues/46999,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: reductions",scan when time allows,2020-10-29T16:06:16Z
issue,46522,ModuleNotFoundError when installing PyTorch via pip on aarch64 environment,https://github.com/pytorch/pytorch/issues/46522,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: arm, module: binaries",scan when time allows,2020-10-29T14:44:48Z
issue,46980,torch.acos not supported for sparse layout,https://github.com/pytorch/pytorch/issues/46980,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2020-10-28T17:28:09Z
issue,43352,"Remove warning, and update documentation.",https://github.com/pytorch/pytorch/issues/43352,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: optimizer",scan when time allows,2020-10-28T14:56:35Z
issue,46929,Pooling code does not allow sliding window starting in right padded region,https://github.com/pytorch/pytorch/issues/46929,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: pooling,scan when time allows,2020-10-27T20:19:07Z
issue,46820,simple v *= v_scale error,https://github.com/pytorch/pytorch/issues/46820,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2020-10-27T17:45:12Z
issue,46899,CUDA out of memory when using torch.load,https://github.com/pytorch/pytorch/issues/46899,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2020-10-27T15:18:17Z
issue,46869,[META] DO NOT DELETE THIS LABEL,https://github.com/pytorch/pytorch/issues/46869,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-10-26T19:34:51Z
issue,46819,Specify input dimensions where they are not obvious,https://github.com/pytorch/pytorch/issues/46819,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2020-10-26T16:56:01Z
issue,46787,Clearer error messages for 'different devices' error in the RNN module,https://github.com/pytorch/pytorch/issues/46787,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-10-26T16:41:37Z
issue,39662,[RFC][distributed] RFC: c10d ProcessGroup extension and C++ API change,https://github.com/pytorch/pytorch/issues/39662,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2020-10-26T16:16:24Z
issue,46760,How to define a new data type in native_functions.yaml?,https://github.com/pytorch/pytorch/issues/46760,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: internals,scan when time allows,2020-10-26T15:16:34Z
issue,4241,single-gpu works but multi-gpu hangs,https://github.com/pytorch/pytorch/issues/4241,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cudnn, module: data parallel",scan when time allows,2020-10-26T05:54:13Z
issue,32909,Some questions Concerning Intel's DNNL(MKLDNN) support in Pytorch (adding support for Intel Processors GPUs) by transitioning to DNNL,https://github.com/pytorch/pytorch/issues/32909,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: mkldnn,scan when time allows,2020-10-25T13:34:05Z
issue,46766,Unable to compile from code v1.6.0,https://github.com/pytorch/pytorch/issues/46766,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2020-10-24T06:18:55Z
issue,46724,PROBLEM WITH INSTALLATION PYTORCH FROM SOURCE,https://github.com/pytorch/pytorch/issues/46724,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2020-10-24T05:41:35Z
issue,46753,Memory leak when creating new tensors inside nn.DataParallel on multiple GPUs,https://github.com/pytorch/pytorch/issues/46753,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2020-10-24T01:09:36Z
issue,44026,[Feature] Einsum like ShapeGuard,https://github.com/pytorch/pytorch/issues/44026,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-10-23T20:06:54Z
issue,46712,The documentation for c10::Dict is completely empty.,https://github.com/pytorch/pytorch/issues/46712,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: docs, module: internals",scan when time allows,2020-10-23T15:06:03Z
issue,46711,A random split function that return the datasets following specific target (label) distribution.,https://github.com/pytorch/pytorch/issues/46711,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: data,scan when time allows,2020-10-22T17:47:51Z
issue,46659,Excluding image processing operators due to no opencv,https://github.com/pytorch/pytorch/issues/46659,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2020-10-22T17:36:36Z
issue,46382,"Code asserts when register new aten operations' implementation thru ""c10::RegisterOperators::op"" API.",https://github.com/pytorch/pytorch/issues/46382,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: internals,scan when time allows,2020-10-22T01:25:49Z
issue,46643,Autograd support for the tensor multiplication of sparse tensors,https://github.com/pytorch/pytorch/issues/46643,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: sparse",scan when time allows,2020-10-21T17:02:27Z
issue,46597,Suggestion: link individual APIs in docs with tutorials that use them,https://github.com/pytorch/pytorch/issues/46597,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2020-10-21T14:54:04Z
issue,34451,Mac build from source error,https://github.com/pytorch/pytorch/issues/34451,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: xnnpack, oncall: mobile",scan when time allows,2020-10-21T10:01:04Z
issue,45855,Do we have plan to offer C++ binding for prune related features.,https://github.com/pytorch/pytorch/issues/45855,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: pruning",scan when time allows,2020-10-21T04:02:14Z
issue,46530,[Request] Batched Dataset->DataLoader interface,https://github.com/pytorch/pytorch/issues/46530,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2020-10-20T15:24:11Z
issue,46559,Exception raised in rpc_async context is silently handled,https://github.com/pytorch/pytorch/issues/46559,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: rpc, oncall: distributed",scan when time allows,2020-10-19T21:07:24Z
issue,46311,"Compile error when use c10::RegisterOperators::Options::kernel with the return type ""std::tuple<Tensor&, Tensor&>""",https://github.com/pytorch/pytorch/issues/46311,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: internals,scan when time allows,2020-10-19T17:42:08Z
issue,46178,"torch.cuda.amp! when I use @autocast() on DCN(DeformConv), the error ""RuntimeError:expect scalar type Float but Half """,https://github.com/pytorch/pytorch/issues/46178,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision), module: cuda",scan when time allows,2020-10-19T02:30:20Z
issue,37409,Compatibility of subset dataset with disabled batch sampling,https://github.com/pytorch/pytorch/issues/37409,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2020-10-17T11:03:24Z
issue,46367,[chore] Test pybind11 2.6.0 (RC 2?),https://github.com/pytorch/pytorch/issues/46367,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: ci",scan when time allows,2020-10-17T00:34:18Z
issue,46350,torch.distributions.half_normal.HalfNormal.cdf returns negative values,https://github.com/pytorch/pytorch/issues/46350,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distributions,scan when time allows,2020-10-16T22:18:37Z
issue,42882,Support for defining/saving custom operators with dynamic schema,https://github.com/pytorch/pytorch/issues/42882,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-10-16T21:15:46Z
issue,46379,"Reduce ""reserved"" memory by PyTorch.",https://github.com/pytorch/pytorch/issues/46379,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: memory usage",scan when time allows,2020-10-16T04:55:00Z
issue,43429,Inference performance regression caused by hacky_wrapper_for_legacy_signatures,https://github.com/pytorch/pytorch/issues/43429,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dispatch, module: internals, module: performance",scan when time allows,2020-10-16T01:45:42Z
issue,46381,[Feature] Imbalanced batch scattering in th.nn.DataParallel,https://github.com/pytorch/pytorch/issues/46381,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: data parallel,scan when time allows,2020-10-15T22:01:11Z
issue,25045,[Distance functions] F.pdist backward CUDA invalid configuration,https://github.com/pytorch/pytorch/issues/25045,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: distance functions",scan when time allows,2020-10-15T20:36:50Z
issue,15253,[Feature Request] cdist: pairwise distances between two sets of tensors with batch mode,https://github.com/pytorch/pytorch/issues/15253,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: batching, module: distance functions",scan when time allows,2020-10-15T20:36:14Z
issue,29528,Half precision cdist,https://github.com/pytorch/pytorch/issues/29528,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: distance functions, module: half",scan when time allows,2020-10-15T20:35:44Z
issue,37363,"cdist docs imply only one batch dimension is possible, but actually arbitrarily many are allowed",https://github.com/pytorch/pytorch/issues/37363,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: distance functions, module: docs",scan when time allows,2020-10-15T20:35:31Z
issue,37408,torch.cdist() implementation without using contiguous() calls,https://github.com/pytorch/pytorch/issues/37408,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: distance functions, module: performance",scan when time allows,2020-10-15T20:35:23Z
issue,37734,torch.cdist returns inconsistent result,https://github.com/pytorch/pytorch/issues/37734,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: distance functions, module: docs, module: numerical-stability",scan when time allows,2020-10-15T20:35:19Z
issue,35363,NCCL version upgrade for PyTorch,https://github.com/pytorch/pytorch/issues/35363,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2020-10-15T20:26:12Z
issue,46400,Make Undefined reprensentable in DispatchKeySet.,https://github.com/pytorch/pytorch/issues/46400,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-10-15T18:51:47Z
issue,21411,Implementation of Group equivariant convolutions,https://github.com/pytorch/pytorch/issues/21411,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2020-10-15T00:50:48Z
issue,46276,test_distributed_* does not show error details from the subprocess,https://github.com/pytorch/pytorch/issues/46276,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: tests, oncall: distributed",scan when time allows,2020-10-14T19:37:50Z
issue,46102,test_cat_cuda (__main__.TestTensorExprFuser) fails,https://github.com/pytorch/pytorch/issues/46102,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tests,scan when time allows,2020-10-14T17:55:46Z
issue,46065,TestDataLoader.test_proper_exit takes 2.5min to finish,https://github.com/pytorch/pytorch/issues/46065,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ci, module: dataloader, module: tests",scan when time allows,2020-10-14T16:01:44Z
issue,46319,Remove header declarations for CPUType/TypeDefault,https://github.com/pytorch/pytorch/issues/46319,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: internals,scan when time allows,2020-10-14T15:09:58Z
issue,46084,"In PyTorch Tutorials, RuntimeError: CUDA error: out of memory happen",https://github.com/pytorch/pytorch/issues/46084,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: memory usage,scan when time allows,2020-10-14T04:11:32Z
issue,37104,[JIT] Use of global value creates confusing error message,https://github.com/pytorch/pytorch/issues/37104,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-10-14T03:34:40Z
issue,46241,ubuntu16.04+vscode compile libtorch error,https://github.com/pytorch/pytorch/issues/46241,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2020-10-14T01:50:12Z
issue,32589,User-friendly handling of types and devices,https://github.com/pytorch/pytorch/issues/32589,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: numpy, module: ux",scan when time allows,2020-10-13T22:53:35Z
issue,46254,Enable named tensor inputs to `torch.linalg.norm`,https://github.com/pytorch/pytorch/issues/46254,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: named tensor,scan when time allows,2020-10-13T21:39:26Z
issue,22024,Consolidate definition of operators/gradients where possible,https://github.com/pytorch/pytorch/issues/22024,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: internals,scan when time allows,2020-10-13T19:20:02Z
issue,1750,Add support for colors (and maybe other attributes) to NVTX API,https://github.com/pytorch/pytorch/issues/1750,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2020-10-12T21:32:50Z
issue,46184,redesign of Dropout,https://github.com/pytorch/pytorch/issues/46184,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2020-10-12T19:52:57Z
issue,46176,BucketSampler for easy variable-length input batching,https://github.com/pytorch/pytorch/issues/46176,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2020-10-12T19:36:56Z
issue,46149,seeding does not work when I initialize a linear model even if I do not use it in the code,https://github.com/pytorch/pytorch/issues/46149,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2020-10-12T16:59:41Z
issue,46139,"Policy CMP0012 is not set, CMake, Building For C++ With PyTorch CUDA",https://github.com/pytorch/pytorch/issues/46139,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2020-10-12T16:53:16Z
issue,46093,"Spawn wrapper should catch BaseException, not Exception",https://github.com/pytorch/pytorch/issues/46093,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: multiprocessing,scan when time allows,2020-10-12T14:37:47Z
issue,46066,TestXNNPACKConv1dTransformPass.test_conv1d_with_relu_fc takes 2+ min to finsh,https://github.com/pytorch/pytorch/issues/46066,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ci, module: tests",scan when time allows,2020-10-12T14:26:02Z
issue,46061,Support undispatched ops in codegen,https://github.com/pytorch/pytorch/issues/46061,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: codegen,scan when time allows,2020-10-12T14:22:05Z
issue,43467,Data loader struct pack issue(overflow)?,https://github.com/pytorch/pytorch/issues/43467,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dataloader, module: multiprocessing",scan when time allows,2020-10-10T12:00:24Z
issue,37682,LSTMCell consumes x1.5 more memory on CUDA on pytorch >=1.3 comparing to pytorch 1.2,https://github.com/pytorch/pytorch/issues/37682,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: memory usage, module: rnn",scan when time allows,2020-10-10T11:28:21Z
issue,10172,"[Feature request] Batch eig/symeig functions (for small matrices, with CUDA)",https://github.com/pytorch/pytorch/issues/10172,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: batching, module: linear algebra",scan when time allows,2020-10-10T10:19:35Z
issue,19685,Bad overload order for zeros_like,https://github.com/pytorch/pytorch/issues/19685,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: pybind, module: tensor creation",scan when time allows,2020-10-10T10:19:03Z
issue,23382,Tensor from mmaped storage loads the entire file into memory,https://github.com/pytorch/pytorch/issues/23382,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: memory usage,scan when time allows,2020-10-10T10:18:48Z
issue,26428,Remove TensorOptions logic from generated code,https://github.com/pytorch/pytorch/issues/26428,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: internals,scan when time allows,2020-10-10T10:18:36Z
issue,33040,[debatable] Better infer dtype in torch.as_tensor,https://github.com/pytorch/pytorch/issues/33040,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tensor creation,scan when time allows,2020-10-10T10:17:48Z
issue,42597,Logsigmoid as chaining operation,https://github.com/pytorch/pytorch/issues/42597,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-10-10T10:16:20Z
issue,2180,Discrepancy in BCEWithLogitsLoss and ClassNLLLoss,https://github.com/pytorch/pytorch/issues/2180,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: cuda, module: loss",scan when time allows,2020-10-10T10:11:52Z
issue,1747,In-place bernoulli_ has more functionality than torch.bernoulli with output parameter,https://github.com/pytorch/pytorch/issues/1747,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distributions,scan when time allows,2020-10-10T10:11:40Z
issue,1328,Unhelpful CrossEntropyLoss dimension error message,https://github.com/pytorch/pytorch/issues/1328,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: error checking, module: loss",scan when time allows,2020-10-10T10:11:19Z
issue,20935,torch.full_like missing documentation for out input variable,https://github.com/pytorch/pytorch/issues/20935,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2020-10-10T10:08:04Z
issue,16599,Feature Request: Earth Mover's Distance Loss,https://github.com/pytorch/pytorch/issues/16599,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: loss,scan when time allows,2020-10-10T10:04:39Z
issue,20323,Support size to `torch.normal`,https://github.com/pytorch/pytorch/issues/20323,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: numpy,scan when time allows,2020-10-10T10:03:41Z
issue,21457,downsampling with grid_sample doesn't match interpolate,https://github.com/pytorch/pytorch/issues/21457,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: interpolation, module: vision",scan when time allows,2020-10-10T09:59:15Z
issue,21688,Batched Conv2d for sequence data,https://github.com/pytorch/pytorch/issues/21688,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: batching, module: convolution",scan when time allows,2020-10-10T09:58:47Z
issue,21856,Batch Normalization axis,https://github.com/pytorch/pytorch/issues/21856,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: batching, module: norms and normalization",scan when time allows,2020-10-10T09:56:19Z
issue,22550,Eigen Tensor library for convolutions on CPU,https://github.com/pytorch/pytorch/issues/22550,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: arm, module: convolution, module: cpu, module: performance",scan when time allows,2020-10-10T09:51:44Z
issue,22707,torch.fill_() exists and modifies the input tensor: Expected or bug?,https://github.com/pytorch/pytorch/issues/22707,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bc-breaking, module: deprecation, module: ux",scan when time allows,2020-10-10T09:48:36Z
issue,22330,Training CNNs with deconvolution,https://github.com/pytorch/pytorch/issues/22330,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: vision",scan when time allows,2020-10-10T09:47:52Z
issue,22487,"""Floating point exception"" after trying the method from the issue #22382",https://github.com/pytorch/pytorch/issues/22487,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-10-10T09:47:11Z
issue,22513,No assertion when using scatter_ on a non-contiguous tensor,https://github.com/pytorch/pytorch/issues/22513,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: error checking, module: partial aliasing, module: scatter & gather ops",scan when time allows,2020-10-10T09:46:19Z
issue,22803,Port `fmod` operator from the TH code to Aten,https://github.com/pytorch/pytorch/issues/22803,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: cuda, module: porting",scan when time allows,2020-10-10T09:45:17Z
issue,22919,Unify tensor shape formatting in shape checks,https://github.com/pytorch/pytorch/issues/22919,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: error checking",scan when time allows,2020-10-10T09:44:48Z
issue,23159,"Can't `torch.sum(tensor, dim)` where `dim >= 64`",https://github.com/pytorch/pytorch/issues/23159,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: error checking, module: reductions, module: tensoriterator",scan when time allows,2020-10-10T09:44:03Z
issue,23305,make torch.utils._download_url_from_file public and add a docstring,https://github.com/pytorch/pytorch/issues/23305,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: hub, module: ux",scan when time allows,2020-10-10T09:43:51Z
issue,24031,fractional_max_pool2d_with_indices silently ignores output_ratio if output_size is provided,https://github.com/pytorch/pytorch/issues/24031,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: error checking, module: pooling",scan when time allows,2020-10-10T09:38:34Z
issue,24041,tensor.var_mean variant for existing torch.var_mean (and same for std_mean),https://github.com/pytorch/pytorch/issues/24041,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: reductions,scan when time allows,2020-10-10T09:38:10Z
issue,24155,"""To compact weights again call flatten_parameters()"" is printed every step for every GPU",https://github.com/pytorch/pytorch/issues/24155,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: data parallel, module: nn, module: rnn",scan when time allows,2020-10-10T09:37:30Z
issue,24261,"""PyTorch core"" thread local flag",https://github.com/pytorch/pytorch/issues/24261,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: internals,scan when time allows,2020-10-10T09:33:50Z
issue,24303,Port `masked_fill` operator from the TH code to Aten,https://github.com/pytorch/pytorch/issues/24303,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: cuda, module: porting",scan when time allows,2020-10-10T09:31:53Z
issue,24313,Value_select to perform region-wise selection,https://github.com/pytorch/pytorch/issues/24313,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-10-10T09:31:46Z
issue,24168,torch.unique is inconsistent with NumPy's unique,https://github.com/pytorch/pytorch/issues/24168,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: numpy",scan when time allows,2020-10-10T09:23:17Z
issue,25039,Add new interpolation modes to `grid_sample`,https://github.com/pytorch/pytorch/issues/25039,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: interpolation, module: nn",scan when time allows,2020-10-10T09:11:10Z
issue,25156,euclidean distance between two tensors,https://github.com/pytorch/pytorch/issues/25156,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distance functions,scan when time allows,2020-10-10T09:10:08Z
issue,25978,Provide a way to select SVD algorithm in PyTorch?,https://github.com/pytorch/pytorch/issues/25978,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: linear algebra,scan when time allows,2020-10-10T09:08:21Z
issue,28214,reflective padding for 5D tensor,https://github.com/pytorch/pytorch/issues/28214,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: padding,scan when time allows,2020-10-10T09:01:57Z
issue,28223,Convert manually bound `cuda` `cpu` `byte` `float` operators to native_functions,https://github.com/pytorch/pytorch/issues/28223,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ux,scan when time allows,2020-10-10T09:01:39Z
issue,27147,Stop binding in-place methods to `torch.*`,https://github.com/pytorch/pytorch/issues/27147,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ux,scan when time allows,2020-10-10T09:00:40Z
issue,27936,Specifying `pos_weight` in F.binary_cross_entropy_with_logits leads to RuntimeError: class size not match,https://github.com/pytorch/pytorch/issues/27936,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: error checking, module: nn",scan when time allows,2020-10-10T08:59:38Z
issue,29510,torch.stack: bad shape error message,https://github.com/pytorch/pytorch/issues/29510,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: error checking",scan when time allows,2020-10-10T08:57:56Z
issue,30263,pairwise_dist eps argument is confusing,https://github.com/pytorch/pytorch/issues/30263,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bc-breaking, module: docs, module: numerical-stability",scan when time allows,2020-10-10T08:53:25Z
issue,30565,`F.interpolate` returns unexpected result when dealing with output size `1`,https://github.com/pytorch/pytorch/issues/30565,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: interpolation, module: nn, module: vision",scan when time allows,2020-10-10T08:53:06Z
issue,31261,pinned memory requires DeviceGuard in multi-process envs,https://github.com/pytorch/pytorch/issues/31261,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: docs, module: multiprocessing",scan when time allows,2020-10-10T08:47:51Z
issue,31458,Integrate `torch.xxx` and `Tensor.xxx`,https://github.com/pytorch/pytorch/issues/31458,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ux,scan when time allows,2020-10-10T08:46:34Z
issue,31474,Slow clip_grad_norm_ because of .item() calls when run on device,https://github.com/pytorch/pytorch/issues/31474,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: norms and normalization, module: performance",scan when time allows,2020-10-10T08:46:05Z
issue,31655,pytorch out of memory when calculation squared difference of unfold,https://github.com/pytorch/pytorch/issues/31655,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: memory usage,scan when time allows,2020-10-10T08:44:29Z
issue,31679,broadcast randperm with dim specification,https://github.com/pytorch/pytorch/issues/31679,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-10-10T08:44:17Z
issue,31776,Documentation for `scatter` incorrectly states that index values must be unique,https://github.com/pytorch/pytorch/issues/31776,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bc-breaking, module: docs, module: scatter & gather ops",scan when time allows,2020-10-10T08:43:33Z
issue,32549,"""Tried to register multiple operators with the same name and the same overload name"" error is confusing",https://github.com/pytorch/pytorch/issues/32549,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dispatch,scan when time allows,2020-10-10T08:29:47Z
issue,32614,Logical AND and OR for Tensors in C++ API.,https://github.com/pytorch/pytorch/issues/32614,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp,scan when time allows,2020-10-10T08:02:25Z
issue,32872,Confusing error messages of tensor.scatter_ on both CPU and CUDA,https://github.com/pytorch/pytorch/issues/32872,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: error checking, module: scatter & gather ops",scan when time allows,2020-10-10T07:58:10Z
issue,33035,Support batch linear transformation,https://github.com/pytorch/pytorch/issues/33035,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: batching, module: nn",scan when time allows,2020-10-10T07:56:44Z
issue,33041,bytearray(tensor) behaves very differently from bytearray(tensor.numpy()),https://github.com/pytorch/pytorch/issues/33041,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: numpy, module: printing",scan when time allows,2020-10-10T07:56:04Z
issue,33373,torch.clamp_ not inplace during backward,https://github.com/pytorch/pytorch/issues/33373,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: memory usage",scan when time allows,2020-10-10T07:42:29Z
issue,33571,empty_sparse shouldn't be called with memory layout but is,https://github.com/pytorch/pytorch/issues/33571,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: memory format, module: sparse, module: tensor creation",scan when time allows,2020-10-10T07:39:12Z
issue,34544,Strange behaviour of F.interpolate with bicubic mode.,https://github.com/pytorch/pytorch/issues/34544,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2020-10-10T07:34:24Z
issue,34576,Significant speed difference between P100 and V100,https://github.com/pytorch/pytorch/issues/34576,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cublas, module: cuda, module: performance",scan when time allows,2020-10-10T07:33:44Z
issue,34675,[feature request]Support dilation parameter for unfold2d_* function (slow cpu maxpool2d #28733),https://github.com/pytorch/pytorch/issues/34675,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: mkldnn, module: nn, module: performance, module: pooling",scan when time allows,2020-10-10T07:30:42Z
issue,34874,Extend nn.functional.softmax for arbitrary dimensions,https://github.com/pytorch/pytorch/issues/34874,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2020-10-10T07:28:46Z
issue,35553,caffe2 `DEPTHWISE3x3.Conv` test is broken,https://github.com/pytorch/pytorch/issues/35553,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tests,scan when time allows,2020-10-10T07:25:21Z
issue,37863,Deprecate type() and type_as() call,https://github.com/pytorch/pytorch/issues/37863,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: deprecation, module: ux",scan when time allows,2020-10-10T07:15:53Z
issue,38394,4D tensor support for torch.nn.functionnal.fold() (col2im),https://github.com/pytorch/pytorch/issues/38394,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2020-10-10T07:15:36Z
issue,35775,Unexpected behaviour for affine_grid and grid_sample with 3D inputs,https://github.com/pytorch/pytorch/issues/35775,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: numerical-stability",scan when time allows,2020-10-10T07:10:45Z
issue,46035,"Error installing on source: ""src.cxx:1:10: fatal error: glog/stl_logging.h: No such file or directory""",https://github.com/pytorch/pytorch/issues/46035,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2020-10-10T04:41:13Z
issue,41916,ASAN build broken when using USE_ASAN=1,https://github.com/pytorch/pytorch/issues/41916,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: power",scan when time allows,2020-10-09T19:49:14Z
issue,41918,ASAN build breaks when using third_party/protobuf,https://github.com/pytorch/pytorch/issues/41918,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: power, module: protobuf",scan when time allows,2020-10-09T19:49:10Z
issue,40095,Bump up NCCL to 2.7.3,https://github.com/pytorch/pytorch/issues/40095,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nccl, oncall: distributed",scan when time allows,2020-10-09T00:20:04Z
issue,31913,`index_add_` with multidimensional index,https://github.com/pytorch/pytorch/issues/31913,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: advanced indexing,scan when time allows,2020-10-08T21:19:07Z
issue,45690,Complex and real results do not agree when computing reciprocal or pow(-1) of 0,https://github.com/pytorch/pytorch/issues/45690,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: complex,scan when time allows,2020-10-08T20:18:01Z
issue,45851,The same function can have different signatures in the torch and torch.nn.functional namespaces,https://github.com/pytorch/pytorch/issues/45851,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2020-10-08T16:55:50Z
issue,26354,torch.tensor / torch.as_tensor not working with list of tensors,https://github.com/pytorch/pytorch/issues/26354,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: error checking, module: numpy, module: tensor creation",scan when time allows,2020-10-08T08:05:17Z
issue,26527,Multilinear map,https://github.com/pytorch/pytorch/issues/26527,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2020-10-08T08:03:08Z
issue,30658,"Dedicated inverse AdaptiveMaxPool1d operation (e.g., AdaptiveMaxUnpool1d)",https://github.com/pytorch/pytorch/issues/30658,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: pooling, module: vision",scan when time allows,2020-10-08T07:54:17Z
issue,30934,Spurious negative output in convolution of positive tensors,https://github.com/pytorch/pytorch/issues/30934,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: docs",scan when time allows,2020-10-08T07:50:52Z
issue,39632,F.affine_grid dispatch async issue,https://github.com/pytorch/pytorch/issues/39632,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: nn, module: performance",scan when time allows,2020-10-08T07:48:31Z
issue,40419,Add a `like` argument to creation ops,https://github.com/pytorch/pytorch/issues/40419,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tensor creation,scan when time allows,2020-10-08T07:47:35Z
issue,41793,General reduction mode selection for in-place and out-variants for wider range (hopefully all) of ops,https://github.com/pytorch/pytorch/issues/41793,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: reductions,scan when time allows,2020-10-08T07:45:55Z
issue,42902,Avoid dynamic isCustomClassRegistered() checks in kernel call paths,https://github.com/pytorch/pytorch/issues/42902,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dispatch,scan when time allows,2020-10-08T07:43:48Z
issue,27276,Add in-place view (view_),https://github.com/pytorch/pytorch/issues/27276,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: viewing and reshaping,scan when time allows,2020-10-08T07:42:10Z
issue,13218,Support for integer interpolation (torch.nn.functional.interpolate),https://github.com/pytorch/pytorch/issues/13218,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2020-10-08T07:37:15Z
issue,18987,Performance issue of unique on CPU,https://github.com/pytorch/pytorch/issues/18987,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: performance, module: sorting and selection",scan when time allows,2020-10-08T07:29:32Z
issue,19149,Make operators like logsumexp and cumsum operate over dimension 0 by default (or at least for 1D arrays),https://github.com/pytorch/pytorch/issues/19149,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: numpy, module: reductions",scan when time allows,2020-10-08T07:28:28Z
issue,19150,Improve unit test coverage of torch.unique,https://github.com/pytorch/pytorch/issues/19150,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: sorting and selection, module: tests",scan when time allows,2020-10-08T07:27:42Z
issue,28090,[discussion] Smarter version of torch.reshape (can avoid realloc in some cases),https://github.com/pytorch/pytorch/issues/28090,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: viewing and reshaping,scan when time allows,2020-10-08T07:16:16Z
issue,29116,torch.masked_fill missing out argument,https://github.com/pytorch/pytorch/issues/29116,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ux,scan when time allows,2020-10-08T07:07:03Z
issue,10735,Have all C++ modules expose a __file__ attribute,https://github.com/pytorch/pytorch/issues/10735,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2020-10-08T06:53:20Z
issue,36036,torch.multinomial is misnamed.,https://github.com/pytorch/pytorch/issues/36036,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: distributions, module: ux",scan when time allows,2020-10-08T06:42:17Z
issue,36309,"torch.ceil, torch.floor should accept a dtype argument",https://github.com/pytorch/pytorch/issues/36309,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: numpy, module: ux",scan when time allows,2020-10-08T06:40:42Z
issue,36314,[discussion] Refactor spectral_norm to use the newly merged lowrank solvers and proposal for Linear Algebra Cookbook page,https://github.com/pytorch/pytorch/issues/36314,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: linear algebra,scan when time allows,2020-10-08T06:36:53Z
issue,37160,Enhanced operator context when reporting errors,https://github.com/pytorch/pytorch/issues/37160,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: shape checking,scan when time allows,2020-10-08T06:34:20Z
issue,17169,Global Second Order Pooling,https://github.com/pytorch/pytorch/issues/17169,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: pooling,scan when time allows,2020-10-08T03:51:49Z
issue,9570,[feature request] More options for Fractional Max Pooling,https://github.com/pytorch/pytorch/issues/9570,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: pooling,scan when time allows,2020-10-08T03:51:37Z
issue,3396,CUDA topk is slow for some input sizes,https://github.com/pytorch/pytorch/issues/3396,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: performance, module: sorting and selection",scan when time allows,2020-10-08T03:51:17Z
issue,27542,Make topk sort stable,https://github.com/pytorch/pytorch/issues/27542,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: determinism, module: sorting and selection",scan when time allows,2020-10-08T03:50:41Z
issue,15822,AvgPool2d doesn't test if kernel is smaller than input size,https://github.com/pytorch/pytorch/issues/15822,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, module: error checking, module: pooling",scan when time allows,2020-10-08T03:38:21Z
issue,18728,MaxPool with n-dimensional tensors,https://github.com/pytorch/pytorch/issues/18728,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: pooling",scan when time allows,2020-10-08T03:34:23Z
issue,6138,[feature request] SIGNUM an optimizer that takes the sign of gradient or momentum.,https://github.com/pytorch/pytorch/issues/6138,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2020-10-07T20:51:41Z
issue,6165,Note about unusual stride situations in dev docs / make it easier to test for this in the library,https://github.com/pytorch/pytorch/issues/6165,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: memory format",scan when time allows,2020-10-07T20:50:27Z
issue,6328,[feature request] Include libomp support (macOS),https://github.com/pytorch/pytorch/issues/6328,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: macos",scan when time allows,2020-10-07T20:46:53Z
issue,6350,worker assignments in torch.utils.dataloader.py,https://github.com/pytorch/pytorch/issues/6350,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2020-10-07T20:46:12Z
issue,41926,torch.exp() cannot be modified by an inplace operation,https://github.com/pytorch/pytorch/issues/41926,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: docs",scan when time allows,2020-10-07T17:23:08Z
issue,31598,ATen not compiled with MKL support,https://github.com/pytorch/pytorch/issues/31598,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: mkl",scan when time allows,2020-10-07T17:10:14Z
issue,38550,Returning a tensor instead of a list in split and chunk,https://github.com/pytorch/pytorch/issues/38550,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-10-07T14:34:36Z
issue,45840,Error with DataParallel and dataclass,https://github.com/pytorch/pytorch/issues/45840,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: data parallel,scan when time allows,2020-10-07T10:01:17Z
issue,42805,Unbuffered operation,https://github.com/pytorch/pytorch/issues/42805,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: numpy,scan when time allows,2020-10-07T09:47:36Z
issue,31710,AttributeError: module 'torch.distributed' has no attribute 'init_process_group' on torch 1.3 aarch64,https://github.com/pytorch/pytorch/issues/31710,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2020-10-07T09:06:30Z
issue,18315,[FR] support default size (scalar) in torch.randint,https://github.com/pytorch/pytorch/issues/18315,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: numpy, module: tensor creation",scan when time allows,2020-10-07T05:24:16Z
issue,18634,Speed-up torch.cat on CPU,https://github.com/pytorch/pytorch/issues/18634,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: performance",scan when time allows,2020-10-07T05:22:47Z
issue,42246,"`torch.tensor([1, 2], dtype=torch.int).fmod(torch.tensor(0, dtype=torch.float))` leads to ""RuntimeError: result type Float can't be cast to the desired output type Int""",https://github.com/pytorch/pytorch/issues/42246,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: type promotion,scan when time allows,2020-10-07T05:16:47Z
issue,42529,renorm dim argument is extremely confusing,https://github.com/pytorch/pytorch/issues/42529,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: ux",scan when time allows,2020-10-07T05:15:16Z
issue,42625,F.grid_sample produces weird results on single-pixel images,https://github.com/pytorch/pytorch/issues/42625,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2020-10-07T05:11:38Z
issue,38470,[Feature] Option to have zeros/ones/full output tensor with zero strides,https://github.com/pytorch/pytorch/issues/38470,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tensor creation,scan when time allows,2020-10-07T05:05:24Z
issue,39260,Naming inconsistency: padding_mode vs pad_mode + F.conv* docs ops miss padding_mode arg at all,https://github.com/pytorch/pytorch/issues/39260,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ux,scan when time allows,2020-10-07T04:59:05Z
issue,39224,Negative stride values in `as_strided`,https://github.com/pytorch/pytorch/issues/39224,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: error checking, module: internals, module: memory format",scan when time allows,2020-10-07T04:57:57Z
issue,39134,Crossentropy inconsistent results depending on tensor order,https://github.com/pytorch/pytorch/issues/39134,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: determinism, module: nn",scan when time allows,2020-10-07T04:57:34Z
issue,38955,fold variation,https://github.com/pytorch/pytorch/issues/38955,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2020-10-07T04:53:46Z
issue,39959,Different max_pool2d cpp signatures due to indices.,https://github.com/pytorch/pytorch/issues/39959,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: pooling,scan when time allows,2020-10-07T04:45:45Z
issue,42858,"prelu_backward, hardshrink_backward shouldn't be a method",https://github.com/pytorch/pytorch/issues/42858,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dispatch,scan when time allows,2020-10-07T03:51:05Z
issue,41528,"torch.nn.functional.grid_sample()is doing bilinear interpolation when the input is 5D, i think the mode should add 'trilinear'",https://github.com/pytorch/pytorch/issues/41528,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2020-10-07T03:47:43Z
issue,45901,Add support for reading the whole file in from_file,https://github.com/pytorch/pytorch/issues/45901,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: numpy, module: serialization",scan when time allows,2020-10-06T21:25:51Z
issue,32210,"dist.send/recv ""IndexError: map::at"" error when bool tensors are used with mpi backend",https://github.com/pytorch/pytorch/issues/32210,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: boolean tensor,scan when time allows,2020-10-06T20:04:57Z
issue,45044,torch rpc cannot handle UnsupportedNodeError exception,https://github.com/pytorch/pytorch/issues/45044,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: rpc, oncall: distributed",scan when time allows,2020-10-06T19:41:59Z
issue,5272,Unsafe out= keyword argument with tensors sharing storage,https://github.com/pytorch/pytorch/issues/5272,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: numpy, module: safe resize",scan when time allows,2020-10-06T19:05:35Z
issue,30459,[TensorBoard] Graph with objects other than torch.nn.Module can not be visualized.,https://github.com/pytorch/pytorch/issues/30459,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tensorboard,scan when time allows,2020-10-06T14:04:49Z
issue,2129,[Feature request] truncated normal initializer(sampler),https://github.com/pytorch/pytorch/issues/2129,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: initialization,scan when time allows,2020-10-06T10:05:28Z
issue,44917,Incorrect documentation of SGD momentum,https://github.com/pytorch/pytorch/issues/44917,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: optimizer",scan when time allows,2020-10-06T09:49:18Z
issue,44968,[feature request] dtype argument for torch.sign,https://github.com/pytorch/pytorch/issues/44968,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: numpy,scan when time allows,2020-10-06T09:49:08Z
issue,44969,Tensordot does not support Bool,https://github.com/pytorch/pytorch/issues/44969,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: boolean tensor, module: numpy",scan when time allows,2020-10-06T09:48:11Z
issue,45444,Using operator[] in GenericPackedTensorAccessor impossible in cpu code in .cu files (tensor on cpu),https://github.com/pytorch/pytorch/issues/45444,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2020-10-06T09:46:41Z
issue,22447,Weak Symbols Resolution Causes Segmentation Fault in External Libraries,https://github.com/pytorch/pytorch/issues/22447,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-10-06T09:38:32Z
issue,38197,Why no `torch.randperm_like`?,https://github.com/pytorch/pytorch/issues/38197,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tensor creation,scan when time allows,2020-10-06T09:37:13Z
issue,2400,Add CRF Layer,https://github.com/pytorch/pytorch/issues/2400,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2020-10-06T09:35:45Z
issue,2629,[feature request] add pairwise ranking loss,https://github.com/pytorch/pytorch/issues/2629,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: loss,scan when time allows,2020-10-06T09:35:32Z
issue,43544,AMP much worse performance with groupped Conv2d than fp32,https://github.com/pytorch/pytorch/issues/43544,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision), module: cuda, module: dependency bug, module: performance",scan when time allows,2020-10-06T07:33:02Z
issue,38034,Support Slicing of ModuleList during JIT model tracing/scripting,https://github.com/pytorch/pytorch/issues/38034,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, oncall: jit",scan when time allows,2020-10-06T06:46:22Z
issue,44295,Request: Add PyTorch version to state dicts,https://github.com/pytorch/pytorch/issues/44295,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: serialization,scan when time allows,2020-10-06T04:01:15Z
issue,45767,`defineMethodsInModule` no longer exists,https://github.com/pytorch/pytorch/issues/45767,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2020-10-05T20:52:55Z
issue,14996,as_tensor does not use the device of the default tensor type,https://github.com/pytorch/pytorch/issues/14996,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: docs",scan when time allows,2020-10-05T17:04:56Z
issue,41622,New Feature : A very fast algorithm for computing matrix rank,https://github.com/pytorch/pytorch/issues/41622,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: linear algebra, module: numpy",scan when time allows,2020-10-05T06:46:35Z
issue,18094,[JIT Script] Need support on distributions.Categorical.,https://github.com/pytorch/pytorch/issues/18094,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-10-04T12:13:52Z
issue,45717,Better documentation of vec256 API,https://github.com/pytorch/pytorch/issues/45717,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: vectorization",scan when time allows,2020-10-02T15:27:35Z
issue,13865,[c10d] Coordinated file truncation for FileStore,https://github.com/pytorch/pytorch/issues/13865,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2020-10-01T22:43:20Z
issue,16402,error with nccl when distributed training on caffe2,https://github.com/pytorch/pytorch/issues/16402,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-10-01T22:16:52Z
issue,16318,Pytorch with CUDA aware OpenMPI for Infiniband not working with HCOLL and MXM,https://github.com/pytorch/pytorch/issues/16318,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: c10d, oncall: distributed",scan when time allows,2020-10-01T22:16:10Z
issue,14372,[c10d] Configurable timeout per operation for MPI backend,https://github.com/pytorch/pytorch/issues/14372,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2020-10-01T22:07:37Z
issue,10800,[distributed] Synchronization on CUDA side with MPI backend,https://github.com/pytorch/pytorch/issues/10800,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: c10d, module: docs, oncall: distributed",scan when time allows,2020-10-01T22:01:25Z
issue,45573,Build Failed Ubuntu Cuda. CUDAHooks.cpp:97:15: error: ‘struct cudaPointerAttributes’ has no member named ‘type’ return attr.type == cudaMemoryTypeHost;,https://github.com/pytorch/pytorch/issues/45573,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2020-10-01T07:21:40Z
issue,41329,Reference the randomness issue in DataLoader & Dataset documentation.,https://github.com/pytorch/pytorch/issues/41329,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dataloader, module: docs",scan when time allows,2020-09-30T19:45:13Z
issue,38541,Doc update regarding predictability of experiments using Seeds and Workers,https://github.com/pytorch/pytorch/issues/38541,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: determinism, module: docs",scan when time allows,2020-09-30T19:44:52Z
issue,23061,The speed of `torch.einsum` and `torch.matmul` when using `fp16` is slow,https://github.com/pytorch/pytorch/issues/23061,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: linear algebra, module: performance",scan when time allows,2020-09-30T19:15:14Z
issue,45569,Distutils Error in torch.hub Load(),https://github.com/pytorch/pytorch/issues/45569,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: hub,scan when time allows,2020-09-30T15:58:30Z
issue,45540,Why does NLLLoss and CrossEntropyLoss require type long?,https://github.com/pytorch/pytorch/issues/45540,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-09-30T15:48:02Z
issue,45560,Android: allow preallocation of output buffers,https://github.com/pytorch/pytorch/issues/45560,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: android, oncall: mobile",scan when time allows,2020-09-30T15:45:55Z
issue,40916,Vectorized torch.eig(),https://github.com/pytorch/pytorch/issues/40916,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: performance, module: vectorization",scan when time allows,2020-09-30T14:41:43Z
issue,45481,Combining add_scalar with add_hparams with different frequencey,https://github.com/pytorch/pytorch/issues/45481,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tensorboard,scan when time allows,2020-09-29T20:28:27Z
issue,45483,Extra arguments included in the doc where they are not actually presented in the source code,https://github.com/pytorch/pytorch/issues/45483,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: optimizer",scan when time allows,2020-09-29T20:27:47Z
issue,43347,backward for dense+sparse does not work,https://github.com/pytorch/pytorch/issues/43347,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: sparse",scan when time allows,2020-09-28T20:29:54Z
issue,45399,Load ExtraFilesMap with Java library,https://github.com/pytorch/pytorch/issues/45399,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: java,scan when time allows,2020-09-28T16:18:37Z
issue,6071,Repeated 'python setup.py install' with clang leads to -lcpuinfo not found,https://github.com/pytorch/pytorch/issues/6071,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2020-09-28T15:29:23Z
issue,31570,Can't pin storage memory,https://github.com/pytorch/pytorch/issues/31570,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: docs",scan when time allows,2020-09-26T08:58:05Z
issue,45308,Blocked version of Cholesky backward,https://github.com/pytorch/pytorch/issues/45308,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: linear algebra",scan when time allows,2020-09-25T14:41:31Z
issue,44943,In pytorch 1.6。Run model with input no contiguous tensor will become very slow.,https://github.com/pytorch/pytorch/issues/44943,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: performance",scan when time allows,2020-09-24T17:49:47Z
issue,45053,about benchmark issue,https://github.com/pytorch/pytorch/issues/45053,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cudnn,scan when time allows,2020-09-24T06:58:57Z
issue,44724,16-bit input + AMP breaks AdaptiveLogSoftmaxWithLoss,https://github.com/pytorch/pytorch/issues/44724,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision),scan when time allows,2020-09-23T21:40:51Z
issue,5924,Install doesn't work with spaces in directory,https://github.com/pytorch/pytorch/issues/5924,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2020-09-23T19:49:54Z
issue,44504,FP16 inference latency after sleeping,https://github.com/pytorch/pytorch/issues/44504,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: performance,scan when time allows,2020-09-23T01:26:54Z
issue,43646,c10::string_view pybind11 custom type caster.,https://github.com/pytorch/pytorch/issues/43646,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: pybind,scan when time allows,2020-09-22T22:17:18Z
issue,45129,Inconsistent wheel name in https://download.pytorch.org/whl/torch_stable.html,https://github.com/pytorch/pytorch/issues/45129,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2020-09-22T16:25:52Z
issue,23730,Failed to build pytorch ...,https://github.com/pytorch/pytorch/issues/23730,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2020-09-22T10:15:16Z
issue,44948,Clarify use of GLog/GFlags,https://github.com/pytorch/pytorch/issues/44948,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: build warnings, module: dependency bug",scan when time allows,2020-09-22T07:08:27Z
issue,43502,Useless Exception traces when DataSet timing out,https://github.com/pytorch/pytorch/issues/43502,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, module: dataloader",scan when time allows,2020-09-21T18:56:52Z
issue,36333,Illegal memory access / CUDNN_STATUS_MAPPING_ERROR on Quadro 8000,https://github.com/pytorch/pytorch/issues/36333,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cudnn,scan when time allows,2020-09-21T06:59:38Z
issue,43815,Buffers are moved to another device by copy and not in-place,https://github.com/pytorch/pytorch/issues/43815,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-09-18T21:54:14Z
issue,44631,Multi-process Dataloader and multi-parameter exceptions,https://github.com/pytorch/pytorch/issues/44631,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2020-09-18T14:58:36Z
issue,44898,Don't query current device on stream construction,https://github.com/pytorch/pytorch/issues/44898,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: bootcamp,scan when time allows,2020-09-18T14:09:32Z
issue,43510,Make busy waiting time in DDP (kSynchronizeBusyWaitMillis) a configurable environment variable,https://github.com/pytorch/pytorch/issues/43510,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2020-09-17T22:58:43Z
issue,13447,Support gathering nested lists in DataParallel,https://github.com/pytorch/pytorch/issues/13447,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: data parallel,scan when time allows,2020-09-17T07:16:34Z
issue,44817,ComplexHelper.h contains non-inline functions,https://github.com/pytorch/pytorch/issues/44817,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: complex,scan when time allows,2020-09-17T03:54:48Z
issue,44771,Timed out RRef can still be used in subsequent RPCs,https://github.com/pytorch/pytorch/issues/44771,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rpc,scan when time allows,2020-09-17T00:28:48Z
issue,42969,Enable torch.testing._internal typechecks during CI,https://github.com/pytorch/pytorch/issues/42969,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: typing,scan when time allows,2020-09-16T22:56:55Z
issue,44415,Batchnorm2d in PT1.6 couldn't turn off track_running_stats,https://github.com/pytorch/pytorch/issues/44415,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: nn",scan when time allows,2020-09-16T22:12:11Z
issue,30402,There is no support for `weight_decay`/`momentum` in SGD for sparse tensors.,https://github.com/pytorch/pytorch/issues/30402,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2020-09-16T14:39:16Z
issue,44741,cblas_gemv is not being used for gemv on complex on CPU,https://github.com/pytorch/pytorch/issues/44741,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: complex,scan when time allows,2020-09-15T22:27:09Z
issue,44547,floating point exception (core dumped) in training 10^4 steps,https://github.com/pytorch/pytorch/issues/44547,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2020-09-15T20:03:25Z
issue,44704,Error message concerning dtype GRU not identical between devices,https://github.com/pytorch/pytorch/issues/44704,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: error checking,scan when time allows,2020-09-15T17:44:01Z
issue,44719,404 Page not found - inference_api.md & management_api.md,https://github.com/pytorch/pytorch/issues/44719,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: doc infra,scan when time allows,2020-09-15T17:39:13Z
issue,44721,Get rid of copy_from,https://github.com/pytorch/pytorch/issues/44721,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: internals,scan when time allows,2020-09-15T17:23:52Z
issue,31856,Fix softplus clampling issues using logsigmoid,https://github.com/pytorch/pytorch/issues/31856,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2020-09-15T12:52:14Z
issue,42428,KeyError: xxxxxxxxxx when calling optimizer.state_dict(),https://github.com/pytorch/pytorch/issues/42428,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2020-09-15T05:30:41Z
issue,40400,New ProcessGroups created with dist.new_group may leak memory,https://github.com/pytorch/pytorch/issues/40400,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: memory usage, oncall: distributed",scan when time allows,2020-09-14T23:44:14Z
issue,44657,Use macro to define `DispatchKey::toString` method,https://github.com/pytorch/pytorch/issues/44657,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dispatch,scan when time allows,2020-09-14T21:07:49Z
issue,27839,RuntimeError: invalid argument 0: Sizes of tensors must match except in dimension 1. Got 24 and 195 in dimension 0 at /opt/conda/conda-bld/pytorch_1565272271120/work/aten/src/TH/generic/THTensor.cpp:689,https://github.com/pytorch/pytorch/issues/27839,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, oncall: jit",scan when time allows,2020-09-14T20:04:10Z
issue,44611,MacOS CPU torch.tan and torch.tanh do not compute some values properly,https://github.com/pytorch/pytorch/issues/44611,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: complex, module: cpu, module: macos",scan when time allows,2020-09-14T19:37:34Z
issue,44624,CUDA RuntimeError unrecoverably bricks session,https://github.com/pytorch/pytorch/issues/44624,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2020-09-14T17:24:32Z
issue,43638,[FR] hub uses default github branch,https://github.com/pytorch/pytorch/issues/43638,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: hub,scan when time allows,2020-09-14T17:24:02Z
issue,29010,torch.stack does not yet support named tensors,https://github.com/pytorch/pytorch/issues/29010,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: named tensor,scan when time allows,2020-09-14T17:03:51Z
issue,44552,torch.norm does not broadcast for dim > 2,https://github.com/pytorch/pytorch/issues/44552,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2020-09-14T16:45:37Z
issue,44528,DataLoader consumes extremely large shared memory (shm) in its initialization.,https://github.com/pytorch/pytorch/issues/44528,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2020-09-13T18:53:02Z
issue,19160,Suggest model.eval() in torch.no_grad (and vice versa),https://github.com/pytorch/pytorch/issues/19160,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2020-09-13T16:14:00Z
issue,44545,too large data in Queue cause dead lock in Multiprocessing,https://github.com/pytorch/pytorch/issues/44545,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: multiprocessing,scan when time allows,2020-09-11T21:23:04Z
issue,32988,[jit] Dict set item type mismatch error doesn't say the type was inferred,https://github.com/pytorch/pytorch/issues/32988,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-09-10T10:36:32Z
issue,44403,Investigate SyncBatchNorm cleanup with NCCL Async Error Handling,https://github.com/pytorch/pytorch/issues/44403,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2020-09-10T01:49:40Z
issue,39358,SyncBatchNorm doesn't scale well across multiple nodes on large data sizes,https://github.com/pytorch/pytorch/issues/39358,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: performance, oncall: distributed",scan when time allows,2020-09-09T22:10:50Z
issue,44421,[tools.codegen] Rename api.legacy_dispatcher to api.native,https://github.com/pytorch/pytorch/issues/44421,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: internals,scan when time allows,2020-09-09T21:09:24Z
issue,44084,[RFC] Manage CUDA Stream in TensorPipe RPC Agent,https://github.com/pytorch/pytorch/issues/44084,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: rpc, module: tensorpipe",scan when time allows,2020-09-09T17:41:34Z
issue,40959,Will the model run slower when deployed using libtorch ?,https://github.com/pytorch/pytorch/issues/40959,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: performance",scan when time allows,2020-09-09T16:59:52Z
issue,44169,Deadlock with RPC and dist.barrier() for TensorPipeAgent and NCCL.,https://github.com/pytorch/pytorch/issues/44169,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nccl, module: rpc, module: tensorpipe",scan when time allows,2020-09-09T16:47:50Z
issue,44179,REINFORCEMENT LEARNING (DQN) TUTORIAL is not working anymore due to matplotlib.pyplot.imshow(),https://github.com/pytorch/pytorch/issues/44179,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2020-09-09T16:46:40Z
issue,44185,caffe2: Python 3 deprecation warnings about inspect.getargspec,https://github.com/pytorch/pytorch/issues/44185,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-09-09T16:45:35Z
issue,44313,`Undefined symbol: torch::kNearest` when building App iOS,https://github.com/pytorch/pytorch/issues/44313,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ios, oncall: mobile",scan when time allows,2020-09-09T16:26:33Z
issue,44341,Consider cache effects in Timer,https://github.com/pytorch/pytorch/issues/44341,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: bootcamp,scan when time allows,2020-09-09T16:21:11Z
issue,44362,Libtorch CMakeList.txt config help,https://github.com/pytorch/pytorch/issues/44362,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2020-09-09T16:12:06Z
issue,43012,Implement map-style caching DataSet as PyTorch build-in DataSet.,https://github.com/pytorch/pytorch/issues/43012,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, module: dataloader",scan when time allows,2020-09-08T18:07:15Z
issue,17313,Implement Adaptive Input Representations for Neural Language Modeling,https://github.com/pytorch/pytorch/issues/17313,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2020-09-08T14:52:33Z
issue,44274,Slower speeds when using half().,https://github.com/pytorch/pytorch/issues/44274,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: performance,scan when time allows,2020-09-08T01:14:04Z
issue,42405,slow data loading in VisionDataset - need to allow batch loading.,https://github.com/pytorch/pytorch/issues/42405,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dataloader, module: performance, module: vision",scan when time allows,2020-09-08T00:03:59Z
issue,44103,Item seems to affect the backward process of DataParallel,https://github.com/pytorch/pytorch/issues/44103,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: data parallel",scan when time allows,2020-09-05T02:58:50Z
issue,23849,Make MultiProcessTestCase pickable,https://github.com/pytorch/pytorch/issues/23849,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: tests, oncall: distributed",scan when time allows,2020-09-04T17:52:32Z
issue,28929,torch::tensor(scalar) behaves differently from at::tensor(scalar),https://github.com/pytorch/pytorch/issues/28929,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: pybind",scan when time allows,2020-09-03T16:42:28Z
issue,38622,Clarification for usage of negative loss with optim.lr_scheduler.ReduceLROnPlateau,https://github.com/pytorch/pytorch/issues/38622,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2020-09-03T16:00:55Z
issue,5435,Gaussian Sampling,https://github.com/pytorch/pytorch/issues/5435,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2020-09-03T14:35:31Z
issue,5528,TestNN.test_data_parallel takes 10G of memory,https://github.com/pytorch/pytorch/issues/5528,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: memory usage, module: tests",scan when time allows,2020-09-03T14:32:30Z
issue,37334,Why do we not use TorchScript to build graph for tensorboard,https://github.com/pytorch/pytorch/issues/37334,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: tensorboard, oncall: visualization",scan when time allows,2020-09-03T13:22:53Z
issue,43947,torch.cuda.synchronize Influence distributed training,https://github.com/pytorch/pytorch/issues/43947,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: cudnn, module: data parallel, module: performance",scan when time allows,2020-09-03T11:38:24Z
issue,39522,Valgrind leak checking flags losses in libtorch,https://github.com/pytorch/pytorch/issues/39522,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp,scan when time allows,2020-09-03T07:54:07Z
issue,43964,Make it so that leading underscore operators are truly private and can be changed without worry for BC,https://github.com/pytorch/pytorch/issues/43964,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-09-02T21:28:13Z
issue,5524,Redo torch.nn.functional docstring strategy,https://github.com/pytorch/pytorch/issues/5524,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2020-09-01T17:40:58Z
issue,43816,Move randperm() to DistributionTemplates,https://github.com/pytorch/pytorch/issues/43816,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: random,scan when time allows,2020-08-31T22:54:25Z
issue,14200,Failed to run 'bash ../tools/build_pytorch_libs.sh --use-nnpack --use-mkldnn --use-qnnpack caffe2',https://github.com/pytorch/pytorch/issues/14200,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2020-08-31T22:50:57Z
issue,43904,Documentation and `torch.sparse` alias for `torch.bmm` sparse-dense,https://github.com/pytorch/pytorch/issues/43904,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2020-08-31T22:44:17Z
issue,42912,"Add typing information into variable declarations for JIT script "".code"" output",https://github.com/pytorch/pytorch/issues/42912,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-08-31T22:09:50Z
issue,41006,Simplify Adam Optimizer,https://github.com/pytorch/pytorch/issues/41006,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2020-08-30T11:54:01Z
issue,42311,Inconsistent results when trying to enable Tensor Cores on NVIDIA T4,https://github.com/pytorch/pytorch/issues/42311,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision), module: cudnn, module: memory format",scan when time allows,2020-08-29T15:35:04Z
issue,40790,Encoding dimension argument for F.one_hot,https://github.com/pytorch/pytorch/issues/40790,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, module: memory format",scan when time allows,2020-08-29T10:06:52Z
issue,5096,[feature request]Add an env variable to cover different pathes when testing code with openmp,https://github.com/pytorch/pytorch/issues/5096,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tests,scan when time allows,2020-08-28T23:20:04Z
issue,43758,OSError: /lib64/libc.so.6: version `GLIBC_2.14' not found,https://github.com/pytorch/pytorch/issues/43758,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: binaries,scan when time allows,2020-08-28T18:21:59Z
issue,33855,Unclear output for Pytorch Profiler,https://github.com/pytorch/pytorch/issues/33855,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: docs",scan when time allows,2020-08-28T13:04:07Z
issue,43604,ImportError: libtorch_cpu.so: cannot open shared object file: No such file or directory,https://github.com/pytorch/pytorch/issues/43604,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2020-08-27T23:29:55Z
issue,41478,Implement backend fallback for Tracer,https://github.com/pytorch/pytorch/issues/41478,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: internals,scan when time allows,2020-08-27T21:13:46Z
issue,43706,checkpoint restore of optimizers changes dtype of Floating-point state,https://github.com/pytorch/pytorch/issues/43706,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2020-08-27T15:49:51Z
issue,42177,record_function no longer appears in the docs,https://github.com/pytorch/pytorch/issues/42177,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2020-08-27T15:20:07Z
issue,43692,named tensor INTERNAL ASSERT FAILED when indexing with a list.,https://github.com/pytorch/pytorch/issues/43692,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: named tensor,scan when time allows,2020-08-27T13:54:47Z
issue,11937,[feature request] Kumaraswamy distribution,https://github.com/pytorch/pytorch/issues/11937,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distributions,scan when time allows,2020-08-27T13:28:23Z
issue,43486,scatter not place inputs into every device,https://github.com/pytorch/pytorch/issues/43486,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-08-27T06:55:02Z
issue,43685,Syncbatchnorm and DDP,https://github.com/pytorch/pytorch/issues/43685,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, oncall: distributed",scan when time allows,2020-08-27T05:34:32Z
issue,43677,"Some largeCUDATensorTest fails with OOM when running with the entire test suit, but not when running standalone",https://github.com/pytorch/pytorch/issues/43677,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tests,scan when time allows,2020-08-27T00:03:04Z
issue,43663,test_doc_template is not working correctly,https://github.com/pytorch/pytorch/issues/43663,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tests,scan when time allows,2020-08-26T23:23:14Z
issue,43195,one_hot tensors are channels_last but marked as contiguous,https://github.com/pytorch/pytorch/issues/43195,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: memory format,scan when time allows,2020-08-26T21:02:31Z
issue,43617,Type hints from _VariableFunctions and elsewhere clash,https://github.com/pytorch/pytorch/issues/43617,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: typing,scan when time allows,2020-08-26T20:32:55Z
issue,43561,[RFC] RPC BENCHMARK,https://github.com/pytorch/pytorch/issues/43561,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rpc,scan when time allows,2020-08-26T04:07:47Z
issue,5326,TestMultiprocessing.test_fd_sharing hangs with ASAN,https://github.com/pytorch/pytorch/issues/5326,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tests,scan when time allows,2020-08-26T03:01:01Z
issue,5385,Handle python_arg_parser dtype constants better,https://github.com/pytorch/pytorch/issues/5385,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-08-26T02:59:18Z
issue,5212,Weird error message in torch.split_size_or_sections,https://github.com/pytorch/pytorch/issues/5212,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: numpy,scan when time allows,2020-08-24T21:53:24Z
issue,42950,TorchBind C++ Enum Class,https://github.com/pytorch/pytorch/issues/42950,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: torchbind,scan when time allows,2020-08-24T20:34:03Z
issue,43490,Implement a set_printoptions method in libtorch,https://github.com/pytorch/pytorch/issues/43490,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp,scan when time allows,2020-08-24T20:19:00Z
issue,43503,torch.quantization.quantize_dynamic document refers `module` as a parameter,https://github.com/pytorch/pytorch/issues/43503,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2020-08-24T19:56:09Z
issue,43453,CUDAGuard might not create CUcontext,https://github.com/pytorch/pytorch/issues/43453,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2020-08-24T19:21:21Z
issue,5231,[feature request] Stratified splits in random_split function,https://github.com/pytorch/pytorch/issues/5231,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2020-08-24T16:25:26Z
issue,5266,[feature request] warnings for functions with unspecified dim arguments,https://github.com/pytorch/pytorch/issues/5266,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-08-24T16:07:47Z
issue,43459,Trying to build PyTorch from source with LLD 8 fails,https://github.com/pytorch/pytorch/issues/43459,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: static linking",scan when time allows,2020-08-24T13:36:07Z
issue,43411,Add support of random state generator objects to nn.init module,https://github.com/pytorch/pytorch/issues/43411,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: initialization, module: nn",scan when time allows,2020-08-22T18:36:26Z
issue,40736,torch::autograd::Function should set AutoNonVariableTypeMode when running forward,https://github.com/pytorch/pytorch/issues/40736,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2020-08-21T16:20:19Z
issue,43385,OffsetCalculator.cuh(and THCIntegerDivider.cuh) should be available with PyTorch cpu-only binaries,https://github.com/pytorch/pytorch/issues/43385,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: internals,scan when time allows,2020-08-21T13:21:54Z
issue,43402,ModuleDict does not preserve order of initializing dictionary,https://github.com/pytorch/pytorch/issues/43402,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2020-08-21T13:17:36Z
issue,38853,Weight decay in AdamW,https://github.com/pytorch/pytorch/issues/38853,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2020-08-21T04:37:11Z
issue,43325,CUDA error: an illegal memory access was encountered (laugh_kernel at ...../cuda/CUDALoops.cuh:112),https://github.com/pytorch/pytorch/issues/43325,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cublas, module: cuda",scan when time allows,2020-08-21T04:03:34Z
issue,43113,Missing caffe2_pybind11_state in pip install after cmake.,https://github.com/pytorch/pytorch/issues/43113,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: pybind",scan when time allows,2020-08-21T00:53:41Z
issue,43245,mp.spawn 'args' are not clear,https://github.com/pytorch/pytorch/issues/43245,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: multiprocessing",scan when time allows,2020-08-20T19:04:29Z
issue,43281,Cannot find CUDA devices when the machine stays idle for a while,https://github.com/pytorch/pytorch/issues/43281,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2020-08-20T17:25:57Z
issue,43253,Reduce number of stack frames used up by dispatcher,https://github.com/pytorch/pytorch/issues/43253,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, module: internals",scan when time allows,2020-08-20T17:11:49Z
issue,43109,[docs] LR schedulers' get_last_lr missing in docs online,https://github.com/pytorch/pytorch/issues/43109,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2020-08-20T16:47:47Z
issue,43123,Add CutMix transform to torchvision.transforms,https://github.com/pytorch/pytorch/issues/43123,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: vision,scan when time allows,2020-08-20T15:36:59Z
issue,43238,Expose counters for torch.distributed to report communication overhead within a block.,https://github.com/pytorch/pytorch/issues/43238,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2020-08-20T07:59:13Z
issue,24984,DataLoader slow down when `pin_memory=False`,https://github.com/pytorch/pytorch/issues/24984,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dataloader, module: multiprocessing, module: performance",scan when time allows,2020-08-19T10:06:59Z
issue,43116,Accessing tensor by element is super slow,https://github.com/pytorch/pytorch/issues/43116,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: performance,scan when time allows,2020-08-18T15:52:18Z
issue,43140,"Pytorch ""Import torch"" lead to ""core dump""",https://github.com/pytorch/pytorch/issues/43140,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: protobuf",scan when time allows,2020-08-18T15:41:34Z
issue,43115,torch.multinomial with replacement=True produces inaccurate results for large number of categories,https://github.com/pytorch/pytorch/issues/43115,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: distributions, module: numerical-stability",scan when time allows,2020-08-18T14:10:08Z
issue,43091,Enable torch.optim typechecks during CI,https://github.com/pytorch/pytorch/issues/43091,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: typing,scan when time allows,2020-08-18T13:46:05Z
issue,37048,copy_ slowness,https://github.com/pytorch/pytorch/issues/37048,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: performance",scan when time allows,2020-08-18T02:11:32Z
issue,43072,LSTM::permute_hidden breaks Liskov substitution principle,https://github.com/pytorch/pytorch/issues/43072,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: rnn, module: typing",scan when time allows,2020-08-17T17:17:26Z
issue,37585,Unable to use torch.det() inside nn.DataParallel with multiple gpus,https://github.com/pytorch/pytorch/issues/37585,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: data parallel, module: linear algebra",scan when time allows,2020-08-15T18:46:35Z
issue,43064,torch.dot throws an error for input tensors of different dtypes,https://github.com/pytorch/pytorch/issues/43064,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: numpy, module: type promotion",scan when time allows,2020-08-14T18:07:22Z
issue,43040,undefined symbol: FLAGS_caffe2_keep_on_shrink,https://github.com/pytorch/pytorch/issues/43040,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-08-14T17:47:09Z
issue,43030,BUILD_PYTHON switched to OFF at second run of CMake,https://github.com/pytorch/pytorch/issues/43030,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2020-08-14T16:51:32Z
issue,43033,No vectorization for int8 and uint8,https://github.com/pytorch/pytorch/issues/43033,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: vectorization,scan when time allows,2020-08-14T16:50:16Z
issue,42879,Tensorpipe requires setting both GLOO_SOCKET_IFNAME and TP_SOCKET_IFNAME,https://github.com/pytorch/pytorch/issues/42879,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2020-08-14T09:01:00Z
issue,42916,Build problems caffe2 -- pytorch from source & CUDA 11.0,https://github.com/pytorch/pytorch/issues/42916,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cuda",scan when time allows,2020-08-14T01:29:41Z
issue,11324,Why does DistributedDataSampler not use default RNG?,https://github.com/pytorch/pytorch/issues/11324,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2020-08-13T18:20:10Z
issue,42949,Verify TorchBind works with nested class type,https://github.com/pytorch/pytorch/issues/42949,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: torchbind,scan when time allows,2020-08-13T18:19:33Z
issue,42863,TensorOptions extensibility has rusted shut,https://github.com/pytorch/pytorch/issues/42863,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: codegen,scan when time allows,2020-08-13T18:05:47Z
issue,42971,Enable torch.quantization.fuse_modules typechecks during CI,https://github.com/pytorch/pytorch/issues/42971,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: typing,scan when time allows,2020-08-13T04:56:34Z
issue,42807,build error redefinition of ‘struct c10::complex<T>’ with release v1.6.0,https://github.com/pytorch/pytorch/issues/42807,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2020-08-12T19:50:53Z
issue,41400,test_jit.py fails on Power,https://github.com/pytorch/pytorch/issues/41400,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-08-12T15:09:22Z
issue,42729,Mismatch in docs and behavior of align_corners for nn.functional.interpolate,https://github.com/pytorch/pytorch/issues/42729,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: nn",scan when time allows,2020-08-12T10:15:38Z
issue,32047,Sobol point implementation,https://github.com/pytorch/pytorch/issues/32047,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: random,scan when time allows,2020-08-11T19:25:35Z
issue,5070,NVIDIA_DRIVER_CAPABILITIES env variable is missing in pytorch docker images,https://github.com/pytorch/pytorch/issues/5070,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docker,scan when time allows,2020-08-11T14:35:38Z
issue,5106,Saving model with runtime code changes,https://github.com/pytorch/pytorch/issues/5106,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: serialization,scan when time allows,2020-08-11T14:33:23Z
issue,5159,TakeBackward taking a significant portion of backward time,https://github.com/pytorch/pytorch/issues/5159,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: performance",scan when time allows,2020-08-11T14:16:49Z
issue,5161,ASAN detected leaks on python -c 'import torch',https://github.com/pytorch/pytorch/issues/5161,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: memory usage,scan when time allows,2020-08-11T14:15:51Z
issue,42848,Add support for user defined types in serialization in libtorch,https://github.com/pytorch/pytorch/issues/42848,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: serialization",scan when time allows,2020-08-11T14:08:25Z
issue,42663,Why Conv3D is slower than Conv2D when its flops is smaller than Conv2D,https://github.com/pytorch/pytorch/issues/42663,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: cudnn, module: performance",scan when time allows,2020-08-11T01:51:50Z
issue,42800,RuntimeError: cublas runtime error : the GPU program failed to execute at /pytorch/aten/src/THC/THCBlas.cu:27 for spectral norm,https://github.com/pytorch/pytorch/issues/42800,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cublas,scan when time allows,2020-08-11T01:36:34Z
issue,40679,[JIT] Tracing BCE loss throws error when using weight,https://github.com/pytorch/pytorch/issues/40679,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-08-10T22:26:59Z
issue,42774,Missing API reference for SWALR and AveragedModel,https://github.com/pytorch/pytorch/issues/42774,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2020-08-10T22:12:45Z
issue,13234,torch.utils.checkpoint is not compatible with nn.DataParallel,https://github.com/pytorch/pytorch/issues/13234,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: activation checkpointing, module: data parallel",scan when time allows,2020-08-10T22:00:25Z
issue,42829,torch/library.h doc rendering issue,https://github.com/pytorch/pytorch/issues/42829,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2020-08-10T21:50:01Z
issue,42830,Undocumented parameters in the Variables section of the recurrent bidirectional layers regarding the backward pass,https://github.com/pytorch/pytorch/issues/42830,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2020-08-10T21:49:00Z
issue,42818,Can't registered boxed kernel for operator that doesn't support boxing,https://github.com/pytorch/pytorch/issues/42818,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dispatch,scan when time allows,2020-08-10T21:47:59Z
issue,16542,Error with setting tensors to use cpu in packed_padded_sequence when CUDA tensor is set as default,https://github.com/pytorch/pytorch/issues/16542,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: rnn",scan when time allows,2020-08-08T00:12:55Z
issue,40882,JIT fuser result of dropout doesn't fully match eager mode,https://github.com/pytorch/pytorch/issues/40882,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-08-07T23:11:30Z
issue,39022,Cap DDP total number of buckets,https://github.com/pytorch/pytorch/issues/39022,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2020-08-07T22:21:46Z
issue,42705,torch.distributed.rpc package not work well with generator and lambda,https://github.com/pytorch/pytorch/issues/42705,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rpc,scan when time allows,2020-08-07T21:07:57Z
issue,41639,Error with minimal hogwild test (multiprocessing shared memory),https://github.com/pytorch/pytorch/issues/41639,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: multiprocessing,scan when time allows,2020-08-07T17:07:01Z
issue,42671,[discussion] Support __round__ magic,https://github.com/pytorch/pytorch/issues/42671,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: numpy,scan when time allows,2020-08-07T08:42:01Z
issue,42350,An error occurred deploying CRNN using libtorch,https://github.com/pytorch/pytorch/issues/42350,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: rnn, oncall: jit",scan when time allows,2020-08-07T02:50:54Z
issue,42498,pytorch tests failed,https://github.com/pytorch/pytorch/issues/42498,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tests,scan when time allows,2020-08-07T01:45:04Z
issue,42526,Python3 segfaults or doesn't load torch after installation,https://github.com/pytorch/pytorch/issues/42526,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2020-08-06T11:13:54Z
issue,40027,[Android Pytorch] TorchScript traced model returns inconsistent output tensors on each run,https://github.com/pytorch/pytorch/issues/40027,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: oncall: jit, oncall: mobile",scan when time allows,2020-08-06T10:10:08Z
issue,42655,"When one distributed test fails in CI, the next one can fail spuriously",https://github.com/pytorch/pytorch/issues/42655,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ci, module: tests, oncall: distributed",scan when time allows,2020-08-06T04:06:32Z
issue,42183,Skip all reduce globally unused parameters in DDP,https://github.com/pytorch/pytorch/issues/42183,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: data parallel, oncall: distributed",scan when time allows,2020-08-05T23:34:43Z
issue,42544,torch rpc lose device information between rpc calls,https://github.com/pytorch/pytorch/issues/42544,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rpc,scan when time allows,2020-08-05T22:01:11Z
issue,16322,We're binding a bunch of crap to 'torch' namespace which shouldn't be there,https://github.com/pytorch/pytorch/issues/16322,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-08-05T21:06:11Z
issue,42439,How to use libtorch on Jetson TX2,https://github.com/pytorch/pytorch/issues/42439,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: arm, module: binaries",scan when time allows,2020-08-05T21:03:37Z
issue,42545,CuDNN RNN bindings are basically all deprecated in cudnn 8,https://github.com/pytorch/pytorch/issues/42545,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cudnn, module: deprecation, module: rnn",scan when time allows,2020-08-05T17:03:51Z
issue,42366,Stop implementing JIT stack as a std::vector,https://github.com/pytorch/pytorch/issues/42366,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-08-05T14:35:10Z
issue,13746,[caffe2] test depthwise3x3_conv_op_test fails to run,https://github.com/pytorch/pytorch/issues/13746,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-08-05T14:25:31Z
issue,15294,manylinux2014 compatible wheels,https://github.com/pytorch/pytorch/issues/15294,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2020-08-04T18:11:36Z
issue,29554,[feature request] Print some measure of fragmentation at CUDA out-of-memory,https://github.com/pytorch/pytorch/issues/29554,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2020-08-04T18:07:15Z
issue,42539,Docs for Tensor.copy_ non_blocking should mention pinned memory,https://github.com/pytorch/pytorch/issues/42539,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2020-08-04T17:55:33Z
issue,42478,Using hooks with `DataParallel` gets `autograd` error,https://github.com/pytorch/pytorch/issues/42478,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: data parallel,scan when time allows,2020-08-04T14:40:27Z
issue,42188,Eliminate warning when cloning a tensor using `torch.tensor(x)`,https://github.com/pytorch/pytorch/issues/42188,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2020-08-03T17:25:20Z
issue,42447,OneCycleLR argument `pct_start` used as a proportion not percentage,https://github.com/pytorch/pytorch/issues/42447,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: optimizer",scan when time allows,2020-08-03T17:18:15Z
issue,42440,[feature request].Time Series Algorithms,https://github.com/pytorch/pytorch/issues/42440,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-08-03T17:12:50Z
issue,31565,Force libtorch to use CUDA context,https://github.com/pytorch/pytorch/issues/31565,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-08-03T12:44:56Z
issue,18101,Android NNAPI support of Caffe2,https://github.com/pytorch/pytorch/issues/18101,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-08-03T07:34:14Z
issue,41816,install path can set prefix install dir,https://github.com/pytorch/pytorch/issues/41816,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2020-08-02T20:01:55Z
issue,4952,[Feature request] Optimize autograd/ATen when a gradient is clearly zero,https://github.com/pytorch/pytorch/issues/4952,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2020-08-01T15:29:51Z
issue,42362,libpytorch_jni is not provided in conda builds,https://github.com/pytorch/pytorch/issues/42362,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2020-08-01T08:16:16Z
issue,4927,[Feature Request]would PackedSequence support unsorted sequences?,https://github.com/pytorch/pytorch/issues/4927,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2020-08-01T08:00:21Z
issue,4829,[Feature Request] clip_grad_norm for sparse gradients,https://github.com/pytorch/pytorch/issues/4829,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2020-07-31T21:40:59Z
issue,42345,Generate unique `worker_id` for each node in the RPC framework for the store.,https://github.com/pytorch/pytorch/issues/42345,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rpc,scan when time allows,2020-07-31T01:02:36Z
issue,42301,Is it possible that one-version pytorch supports two gpu (GTX TITAN X and TITAN) using at the same time?,https://github.com/pytorch/pytorch/issues/42301,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2020-07-30T21:04:56Z
issue,42295,KeyError: 'track_running_stats' in batchnorm.extra_repr,https://github.com/pytorch/pytorch/issues/42295,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: printing, module: serialization",scan when time allows,2020-07-30T20:56:36Z
issue,42316,Proper testing of `nn.Module` loading backward compatibility,https://github.com/pytorch/pytorch/issues/42316,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: serialization, module: tests",scan when time allows,2020-07-30T20:42:28Z
issue,42109,Slow `index_add_` on `torch.long` tensors,https://github.com/pytorch/pytorch/issues/42109,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: performance",scan when time allows,2020-07-30T16:06:01Z
issue,36989,pytorch latest update(1.4) broke CosineAnnealingWarmRestarts: T_cur is not define,https://github.com/pytorch/pytorch/issues/36989,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2020-07-30T11:43:26Z
issue,42228,[docs] Bitwise ops miss arguments,https://github.com/pytorch/pytorch/issues/42228,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2020-07-30T01:51:19Z
issue,42094,Setting threads number to the number of default by torch.set_num_threads is faster than not setting it,https://github.com/pytorch/pytorch/issues/42094,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: multithreading,scan when time allows,2020-07-29T02:08:13Z
issue,29516,RuntimeError: CUDA error: invalid device ordinal,https://github.com/pytorch/pytorch/issues/29516,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2020-07-28T20:23:19Z
issue,42159,[docs] Add example of interop with DLPack created in plain C,https://github.com/pytorch/pytorch/issues/42159,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2020-07-28T17:22:00Z
issue,41322,SGD documentatiuon detail on g_{t+1},https://github.com/pytorch/pytorch/issues/41322,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2020-07-28T13:53:36Z
issue,42156,"RuntimeError: test_autograd failed! Build success, test failed on IBM POWER9",https://github.com/pytorch/pytorch/issues/42156,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: power",scan when time allows,2020-07-28T13:18:16Z
issue,12773,"Build the docker image from source, but torch.cuda.is_available()==false",https://github.com/pytorch/pytorch/issues/12773,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docker,scan when time allows,2020-07-28T11:46:10Z
issue,42074,Memory usage of torch.nn.functional.interpolate increased with v1.5.0 when run on numpy input,https://github.com/pytorch/pytorch/issues/42074,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: memory usage, module: nn, module: numpy",scan when time allows,2020-07-27T18:44:47Z
issue,41797,[C++] Libtorch error in Release build only,https://github.com/pytorch/pytorch/issues/41797,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: abi,scan when time allows,2020-07-27T17:26:11Z
issue,42112,Function names in title should not be capitalized in documentation,https://github.com/pytorch/pytorch/issues/42112,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2020-07-27T16:56:43Z
issue,42012,EMNIST looks different to MNIST,https://github.com/pytorch/pytorch/issues/42012,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: vision,scan when time allows,2020-07-26T19:14:31Z
issue,42049,"Code hangs when using `set_start_method('spawn', force=True)` in `torch.multiprocessing.pool`",https://github.com/pytorch/pytorch/issues/42049,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: multiprocessing,scan when time allows,2020-07-26T18:58:55Z
issue,7807,LMDB read error for Mnist,https://github.com/pytorch/pytorch/issues/7807,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-07-26T12:49:21Z
issue,41904,Inconsistencies on ModelLayer,https://github.com/pytorch/pytorch/issues/41904,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-07-25T14:15:17Z
issue,4758,Clang color diagnostics don't work with ninja,https://github.com/pytorch/pytorch/issues/4758,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2020-07-24T14:56:31Z
issue,41999,PyBind11 submodule required even when pybind11_PREFER_third_party=OFF,https://github.com/pytorch/pytorch/issues/41999,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2020-07-24T14:48:53Z
issue,32351,"Problem with multiprocessing, custom __getstate__ with Tensors and forkserver",https://github.com/pytorch/pytorch/issues/32351,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: multiprocessing, module: serialization",scan when time allows,2020-07-24T08:25:42Z
issue,41823,bitwise_or / bitwise_and /... reductions across dim (+ multidim / keepdim for consistency),https://github.com/pytorch/pytorch/issues/41823,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-07-24T08:18:16Z
issue,38457,Named Tensor and Indexing,https://github.com/pytorch/pytorch/issues/38457,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: named tensor,scan when time allows,2020-07-24T07:24:20Z
issue,41955,can not use nn::Functional(torch::softmax(-1)) in Sequential,https://github.com/pytorch/pytorch/issues/41955,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: nn",scan when time allows,2020-07-24T03:57:42Z
issue,41938,Installing pytorch from source on Power9 (PPC64LE) + CUDA 10.2 + RHLE7,https://github.com/pytorch/pytorch/issues/41938,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: power",scan when time allows,2020-07-23T23:04:33Z
issue,41950,[docs] Clarify behavior of torch.cuda.device_count() when torch compiled without CUDA (cpu-only) or when CUDA is not available,https://github.com/pytorch/pytorch/issues/41950,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: docs",scan when time allows,2020-07-23T22:41:55Z
issue,41844,test_bottleneck_cuda fails on Power,https://github.com/pytorch/pytorch/issues/41844,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: power, module: tests",scan when time allows,2020-07-23T22:36:50Z
issue,41903,[C++] adding type checking or type casting to torch::PackedTensorAccessor indexing,https://github.com/pytorch/pytorch/issues/41903,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp,scan when time allows,2020-07-23T18:21:16Z
issue,41915,Automating the various Installation processes for Linux,https://github.com/pytorch/pytorch/issues/41915,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2020-07-23T18:17:55Z
issue,41531,Test failure in test_shared_allgather_nccl: NCCL error in: ../torch/lib/c10d/ProcessGroupNCCL.cpp:537,https://github.com/pytorch/pytorch/issues/41531,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: power, oncall: distributed",scan when time allows,2020-07-23T18:14:29Z
issue,41920,Compilation on Power fails with clang due to vec_xl,https://github.com/pytorch/pytorch/issues/41920,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: power",scan when time allows,2020-07-23T18:05:18Z
issue,41921,Failure to compile Eigen on Power with clang,https://github.com/pytorch/pytorch/issues/41921,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: third_party",scan when time allows,2020-07-23T18:04:17Z
issue,41673,"../caffe2/perfkernels/common_avx2.cc:17:2: error: #error ( ""You found a build system error: __AVX2__ is defined (via e.g. -mavx2) "" ""but CAFFE2_PERF_WITH_AVX2 is not defined."");",https://github.com/pytorch/pytorch/issues/41673,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2020-07-23T03:12:33Z
issue,40971,The values calculated according to the document isn't equal to the values calculated by framework,https://github.com/pytorch/pytorch/issues/40971,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: optimizer",scan when time allows,2020-07-22T22:14:28Z
issue,41623,Can we have a way to reset a scheduler back to epoch -1,https://github.com/pytorch/pytorch/issues/41623,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2020-07-22T22:06:02Z
issue,41369,[RFC] Device Placement API for RPC,https://github.com/pytorch/pytorch/issues/41369,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: rpc, module: tensorpipe",scan when time allows,2020-07-22T14:33:56Z
issue,29710,Reliable way to identify RuntimeErrors (CUDA),https://github.com/pytorch/pytorch/issues/29710,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2020-07-22T00:19:07Z
issue,41801,Caffe2 GPU Inference not working on pytorch nightly version,https://github.com/pytorch/pytorch/issues/41801,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-07-21T18:41:46Z
issue,37354,Efficient handling special gradient values in the autograd,https://github.com/pytorch/pytorch/issues/37354,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: performance",scan when time allows,2020-07-21T09:43:03Z
issue,41694,Torch.multiprocessing.spawn can deadlock,https://github.com/pytorch/pytorch/issues/41694,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: multiprocessing,scan when time allows,2020-07-20T23:26:43Z
issue,41660,Potential bug in net_printer,https://github.com/pytorch/pytorch/issues/41660,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-07-20T23:24:13Z
issue,4716,Compilation issue: problem with GPU capability check,https://github.com/pytorch/pytorch/issues/4716,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cuda",scan when time allows,2020-07-20T23:06:20Z
issue,4703,Very slow on CPU,https://github.com/pytorch/pytorch/issues/4703,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: performance, module: rnn",scan when time allows,2020-07-20T23:05:37Z
issue,41392,"About the description of the mathematical formula of nn.RNN, I think it is wrong",https://github.com/pytorch/pytorch/issues/41392,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2020-07-20T00:56:03Z
issue,41525,c++ indexing vs python,https://github.com/pytorch/pytorch/issues/41525,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp,scan when time allows,2020-07-19T05:29:08Z
issue,41561,[RPC] Should we support users _not_ calling rpc.shutdown()?,https://github.com/pytorch/pytorch/issues/41561,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rpc,scan when time allows,2020-07-17T20:18:43Z
issue,41530,Parallel computation of the diagonal of a Jacobian,https://github.com/pytorch/pytorch/issues/41530,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2020-07-17T16:31:35Z
issue,41243,[discussion] Unite nn modules nn.Something*d to just nn.Something with appropriate options,https://github.com/pytorch/pytorch/issues/41243,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2020-07-17T16:26:48Z
issue,41428,Add a done() API to torch.futures.Future and ProcessGroup::Work,https://github.com/pytorch/pytorch/issues/41428,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2020-07-17T03:19:13Z
issue,39426,Format issue in `torch.quantization.add_quant_dequant` documentation parameter section,https://github.com/pytorch/pytorch/issues/39426,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2020-07-16T20:23:13Z
issue,40936,[RFC] [RPC] Automatic retries of all requests in TensorPipe agent,https://github.com/pytorch/pytorch/issues/40936,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: rpc, module: tensorpipe",scan when time allows,2020-07-16T09:29:56Z
issue,41399,Bottleneck when publishing the model using flask about 3 times slower.,https://github.com/pytorch/pytorch/issues/41399,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: performance",scan when time allows,2020-07-16T03:01:10Z
issue,4622,Protect user from No module named _C import error,https://github.com/pytorch/pytorch/issues/4622,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: error checking, module: pybind",scan when time allows,2020-07-16T02:25:59Z
issue,4564,Mixed Tensor/TensorList arguments in ATen functions with explicit derivatives,https://github.com/pytorch/pytorch/issues/4564,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2020-07-15T15:10:53Z
issue,41383,Large overhead (7 microseconds) for PyTorch operation,https://github.com/pytorch/pytorch/issues/41383,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: performance,scan when time allows,2020-07-15T14:24:39Z
issue,4636,Better header hygiene in ATen,https://github.com/pytorch/pytorch/issues/4636,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: internals,scan when time allows,2020-07-15T14:19:52Z
issue,40215,Move all torch.Tensor methods to codegen,https://github.com/pytorch/pytorch/issues/40215,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: internals, module: pybind",scan when time allows,2020-07-15T12:19:20Z
issue,4574,`from` keyword in `random_` gives error,https://github.com/pytorch/pytorch/issues/4574,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distributions,scan when time allows,2020-07-14T23:14:07Z
issue,4551,Met 'cudnnDestroyDropoutDescriptor' while run multiply gpu-based models in multiply processes,https://github.com/pytorch/pytorch/issues/4551,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cudnn, module: multi-gpu",scan when time allows,2020-07-14T23:13:36Z
issue,4530,descriptor 'add' of 'torch._C._VariableBase' object needs an argument,https://github.com/pytorch/pytorch/issues/4530,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2020-07-14T23:13:22Z
issue,4542,Carefully audit contiguity requirements of code,https://github.com/pytorch/pytorch/issues/4542,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cudnn,scan when time allows,2020-07-14T23:11:12Z
issue,41284,add a reparameterized version of inverse Gaussian distribution,https://github.com/pytorch/pytorch/issues/41284,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distributions,scan when time allows,2020-07-14T07:21:18Z
issue,41187,test_upsampling_not_recompute_scale_factor fails with Eigen/OpenBLAS,https://github.com/pytorch/pytorch/issues/41187,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tests,scan when time allows,2020-07-14T06:33:47Z
issue,38788,Convergence issues when using pytorch's native AMP,https://github.com/pytorch/pytorch/issues/38788,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision),scan when time allows,2020-07-13T18:32:01Z
issue,41311,ConvTranspose1d layer behaviour under different channel numbers,https://github.com/pytorch/pytorch/issues/41311,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: docs",scan when time allows,2020-07-13T16:24:40Z
issue,41282,[docs] Strange arg names of torch.bmm/mm and tensor.bmm/mm,https://github.com/pytorch/pytorch/issues/41282,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2020-07-13T12:27:16Z
issue,34126,torch::jit::script::Object::attr should throw AttributeError instead of RuntimeError,https://github.com/pytorch/pytorch/issues/34126,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-07-13T08:46:06Z
issue,1641,[Feature request] In-place 'max' method for Tensor,https://github.com/pytorch/pytorch/issues/1641,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-07-12T12:17:42Z
issue,40969,Regarding graphs page on site,https://github.com/pytorch/pytorch/issues/40969,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2020-07-11T17:32:05Z
issue,24236,Tensorboard: Add disable flag for debugging,https://github.com/pytorch/pytorch/issues/24236,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tensorboard,scan when time allows,2020-07-11T03:19:27Z
issue,13023,Python dataloader Improvements,https://github.com/pytorch/pytorch/issues/13023,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2020-07-10T22:36:54Z
issue,41101,DDP should provide an option for not touching grad of locally unused params,https://github.com/pytorch/pytorch/issues/41101,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2020-07-10T16:35:50Z
issue,41236,"""out of scope"" variable `cmd` used",https://github.com/pytorch/pytorch/issues/41236,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2020-07-10T15:29:14Z
issue,41246,torch.abs(complex) is divergent from NumPy on vectorized NaN values,https://github.com/pytorch/pytorch/issues/41246,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: complex, module: nans and infs",scan when time allows,2020-07-10T15:27:40Z
issue,41244,torch.tan(complex) on CUDA doesn't handle nonfinite values properly,https://github.com/pytorch/pytorch/issues/41244,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: complex, module: cuda",scan when time allows,2020-07-10T15:27:15Z
issue,40729,Custom c++ extension build process doesn't preserve color from compiler,https://github.com/pytorch/pytorch/issues/40729,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: cpp-extensions",scan when time allows,2020-07-10T15:25:05Z
issue,41178,Failure to build on Power9 due to FXDIV_SOURCE_DIR not being set,https://github.com/pytorch/pytorch/issues/41178,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: power",scan when time allows,2020-07-10T08:11:13Z
issue,40470,jit's default dtype is different in sandcastle and test_jit.py,https://github.com/pytorch/pytorch/issues/40470,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: tests, oncall: jit",scan when time allows,2020-07-10T01:51:47Z
issue,32130,Idempotency Keys for RPC Retry,https://github.com/pytorch/pytorch/issues/32130,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rpc,scan when time allows,2020-07-09T22:41:51Z
issue,41148,Source code for some /_modules/.../.../ not displaying on https://pytorch.org/docs/master/_modules/,https://github.com/pytorch/pytorch/issues/41148,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2020-07-09T15:12:35Z
issue,41135,libtorch cmake should PUBLIC ally advertise it as requiring C++14,https://github.com/pytorch/pytorch/issues/41135,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2020-07-09T15:11:05Z
issue,37142,Propagation of channels-last layout leads to massive slowdowns in 1.5 compared to 1.4,https://github.com/pytorch/pytorch/issues/37142,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: internals, module: memory format, module: performance",scan when time allows,2020-07-08T22:34:21Z
issue,4501,Bind in Python _backward ATen functions,https://github.com/pytorch/pytorch/issues/4501,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2020-07-08T20:47:02Z
issue,4400,Installation Optimise For Chinese Users Who Behind the Wall,https://github.com/pytorch/pytorch/issues/4400,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2020-07-08T05:25:08Z
issue,4406,DistributedDataParallel doesn't converge well when using MPI,https://github.com/pytorch/pytorch/issues/4406,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2020-07-08T05:24:24Z
issue,17901,should disable AVX on 32bit x86 / refine AVX availability tests,https://github.com/pytorch/pytorch/issues/17901,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2020-07-08T03:49:20Z
issue,40563,"Compiling PyTorch with 11.0, V11.0.167",https://github.com/pytorch/pytorch/issues/40563,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2020-07-08T02:42:25Z
issue,41106,Missing Code for Audio Classification Tutorial,https://github.com/pytorch/pytorch/issues/41106,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2020-07-08T00:12:32Z
issue,14959,get/set device in c++,https://github.com/pytorch/pytorch/issues/14959,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp,scan when time allows,2020-07-06T19:58:18Z
issue,40997,Is this a bug? The values calculated according to the document isn't equal to the values calculated by framework,https://github.com/pytorch/pytorch/issues/40997,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2020-07-06T16:31:27Z
issue,31883,"The dependency target ""nccl_external"" of target ""gloo_cuda"" does not exist.",https://github.com/pytorch/pytorch/issues/31883,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, oncall: distributed",scan when time allows,2020-07-06T12:15:59Z
issue,40886,Improve error messaging when using dictionaries as inputs to a trace,https://github.com/pytorch/pytorch/issues/40886,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-07-06T04:22:29Z
issue,40978,[FR] NCCL and bool type,https://github.com/pytorch/pytorch/issues/40978,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: data parallel, module: nccl",scan when time allows,2020-07-04T20:27:13Z
issue,32937,Operation Registration Error,https://github.com/pytorch/pytorch/issues/32937,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dispatch,scan when time allows,2020-07-04T04:39:46Z
issue,40974,Add a launching script for RPC,https://github.com/pytorch/pytorch/issues/40974,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rpc,scan when time allows,2020-07-04T01:13:55Z
issue,40891,[FR] torch.floor/ceil should support output int dtype,https://github.com/pytorch/pytorch/issues/40891,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: type promotion,scan when time allows,2020-07-03T21:24:45Z
issue,40854,Build pytorch with cuda11,https://github.com/pytorch/pytorch/issues/40854,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cuda",scan when time allows,2020-07-02T22:36:28Z
issue,40880,Allow TLS to keep distributed autograd context alive,https://github.com/pytorch/pytorch/issues/40880,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rpc,scan when time allows,2020-07-01T19:21:17Z
issue,31759,Allow using digits in names of named tensors,https://github.com/pytorch/pytorch/issues/31759,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: named tensor,scan when time allows,2020-07-01T04:12:01Z
issue,40456,Print values (but not strings) when STRIP_ERROR_MESSAGES is defined,https://github.com/pytorch/pytorch/issues/40456,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, module: internals",scan when time allows,2020-06-30T22:24:55Z
issue,40590,Inconsistent behavior between numpy.exp and torch.exp on CPU for complex numbers,https://github.com/pytorch/pytorch/issues/40590,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: complex, module: numpy",scan when time allows,2020-06-30T21:29:55Z
issue,35446,[JIT] dropout fails on legacy mode,https://github.com/pytorch/pytorch/issues/35446,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-06-30T17:19:24Z
issue,40441,"[feature request] batch_apply, a general-purpose device-agnostic batch iterator",https://github.com/pytorch/pytorch/issues/40441,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: batching,scan when time allows,2020-06-30T15:06:59Z
issue,40763,Enhance supported types of functional.pad,https://github.com/pytorch/pytorch/issues/40763,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: numpy",scan when time allows,2020-06-30T14:19:04Z
issue,40761,Sparse Convolutional support,https://github.com/pytorch/pytorch/issues/40761,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2020-06-30T14:16:40Z
issue,40756,RuntimeError: rois.device().is_cpu() ASSERT FAILED at /vision/torchvision/csrc/cpu/ROIAlign_cpu.cpp:386,https://github.com/pytorch/pytorch/issues/40756,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: vision,scan when time allows,2020-06-30T14:15:09Z
issue,38057,DNNL's backward pass much slower when using nn.grad.conv2d_input and nn.grad.conv2d_weight,https://github.com/pytorch/pytorch/issues/38057,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: mkldnn, module: nn",scan when time allows,2020-06-30T07:14:53Z
issue,40633,RuntimeError: broken pipe from NCCL,https://github.com/pytorch/pytorch/issues/40633,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nccl, oncall: distributed",scan when time allows,2020-06-29T23:44:42Z
issue,14461,The speed of scatter is influenced by the data size while using nn.DataParallel,https://github.com/pytorch/pytorch/issues/14461,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: data parallel, module: dataloader, module: performance",scan when time allows,2020-06-29T20:37:58Z
issue,40687,libtorch: macros in logging_is_not_google_glog.h have very common names like CHECK or LOG,https://github.com/pytorch/pytorch/issues/40687,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: logging,scan when time allows,2020-06-29T20:32:32Z
issue,40667,Segmentation fault in forward pass using DataParallel and multiple GPUs,https://github.com/pytorch/pytorch/issues/40667,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: data parallel",scan when time allows,2020-06-29T15:25:27Z
issue,36386,Docs of distributed module do not include the full documentation for torch.distributed.launch,https://github.com/pytorch/pytorch/issues/36386,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2020-06-28T06:32:39Z
issue,40568,Converting NumPy dtype to Torch dtype when using `as_tensor`,https://github.com/pytorch/pytorch/issues/40568,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: numpy,scan when time allows,2020-06-26T15:26:29Z
issue,40269,caffe2 error during compilation of PyTorch with ROCm on Archlinux,https://github.com/pytorch/pytorch/issues/40269,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-06-26T09:00:07Z
issue,40492,Learning rate change is not applied at designated iteration with a scheduler,https://github.com/pytorch/pytorch/issues/40492,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2020-06-26T03:08:25Z
issue,40564,Stop registering kernels that use DispatchStub as catch all,https://github.com/pytorch/pytorch/issues/40564,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: internals,scan when time allows,2020-06-25T21:49:51Z
issue,40566,"If all parameters are unused by forward pass in a process, backward will not work with DDP.",https://github.com/pytorch/pytorch/issues/40566,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2020-06-25T21:49:21Z
issue,40548,"Embedding with DataParallel can return ""incomplete"" results",https://github.com/pytorch/pytorch/issues/40548,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: data parallel",scan when time allows,2020-06-25T15:42:31Z
issue,40355,Tracking output dimensions of the convolutional layers,https://github.com/pytorch/pytorch/issues/40355,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: internals,scan when time allows,2020-06-25T14:35:47Z
issue,38911,[FR] cuda.comm.broadcast/reduce_add support `out=`,https://github.com/pytorch/pytorch/issues/38911,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, oncall: distributed",scan when time allows,2020-06-24T20:13:33Z
issue,39444,Torch hub: object has no attribute nms,https://github.com/pytorch/pytorch/issues/39444,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docker,scan when time allows,2020-06-24T19:08:40Z
issue,40114,ppc64le: test cpp_extensions/rng_extension.cpp failure (without altivec override),https://github.com/pytorch/pytorch/issues/40114,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: power, module: random",scan when time allows,2020-06-24T19:08:33Z
issue,31232,Default shuffle behavior of DistributedSampler,https://github.com/pytorch/pytorch/issues/31232,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dataloader, oncall: distributed",scan when time allows,2020-06-24T17:16:08Z
issue,13786,PyTorch streams are not cuda-memcheck clean,https://github.com/pytorch/pytorch/issues/13786,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2020-06-23T19:55:41Z
issue,40417,Missing explanation in torch.utils.tensorboard.add_histogram(),https://github.com/pytorch/pytorch/issues/40417,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: tensorboard",scan when time allows,2020-06-23T19:38:08Z
issue,37556,"torch.cartesian_prod(*tensors) error when you have tensors with [x,y]",https://github.com/pytorch/pytorch/issues/37556,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: linear algebra,scan when time allows,2020-06-22T17:04:59Z
issue,39990,libtorch 1.5 macos crash when loading on some mac,https://github.com/pytorch/pytorch/issues/39990,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cpp, module: mkldnn",scan when time allows,2020-06-22T14:58:21Z
issue,4123,Use the int64 version of MKL calls,https://github.com/pytorch/pytorch/issues/4123,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: internals, module: mkl",scan when time allows,2020-06-19T21:38:04Z
issue,33085,not able to import * from fastai.vision in Google collab,https://github.com/pytorch/pytorch/issues/33085,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-06-19T20:20:26Z
issue,4159,Variable outputs of stochastic functions should never require grad,https://github.com/pytorch/pytorch/issues/4159,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distributions,scan when time allows,2020-06-19T18:53:02Z
issue,4186,Feature Request: CPU performance optimization with MKL-DNN,https://github.com/pytorch/pytorch/issues/4186,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: mkldnn,scan when time allows,2020-06-19T18:25:37Z
issue,4181,Fused RNN refactor plan,https://github.com/pytorch/pytorch/issues/4181,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cudnn,scan when time allows,2020-06-19T18:25:12Z
issue,4148,ATen explicitly differentiated native function resolution hazard (call is ambiguous),https://github.com/pytorch/pytorch/issues/4148,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: internals,scan when time allows,2020-06-19T18:23:40Z
issue,40266,CUDA error: out of memory when running tensorpipe test_cuda,https://github.com/pytorch/pytorch/issues/40266,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: rpc, module: tensorpipe",scan when time allows,2020-06-19T18:18:42Z
issue,40250,Include expanded TensorOptions version of op in at:: namespace,https://github.com/pytorch/pytorch/issues/40250,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: codegen,scan when time allows,2020-06-19T18:15:20Z
issue,36125,build Pytorch-1.5.0-rc1 from source fail,https://github.com/pytorch/pytorch/issues/36125,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2020-06-19T13:08:09Z
issue,37079,Result parameters from Single-Process Multi-GPU DDP training on RNN do not match local training,https://github.com/pytorch/pytorch/issues/37079,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2020-06-19T06:20:11Z
issue,40237,Move torch cpp Errors to c10::Error,https://github.com/pytorch/pytorch/issues/40237,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: error checking, module: internals",scan when time allows,2020-06-18T20:23:58Z
issue,40098,Updating learning rate with Libtorch 1.5 and optimiser options,https://github.com/pytorch/pytorch/issues/40098,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: optimizer",scan when time allows,2020-06-18T16:11:45Z
issue,40086,TestListwiseL2rOps::test_lambda_rank_loss fails,https://github.com/pytorch/pytorch/issues/40086,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tests,scan when time allows,2020-06-18T14:16:16Z
issue,40030,matchTensor tests. Add tests to make sure that `isSubtypeOf` and `matchTensor` return equivalent results,https://github.com/pytorch/pytorch/issues/40030,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-06-16T22:49:59Z
issue,39356,Torch script to support dictionary with keys of type tuple,https://github.com/pytorch/pytorch/issues/39356,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-06-16T17:39:21Z
issue,39852,Shared file-system initialization in pytorch distributed is slow,https://github.com/pytorch/pytorch/issues/39852,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: performance, oncall: distributed",scan when time allows,2020-06-16T07:32:49Z
issue,40020,Caffe2 operation switches current CUDA stream,https://github.com/pytorch/pytorch/issues/40020,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2020-06-16T03:27:08Z
issue,40071,[JIT][to-backend] Finalize the PyTorchBackendInterface API/naming for external sharing,https://github.com/pytorch/pytorch/issues/40071,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-06-16T01:23:44Z
issue,40070,[JIT][to-backend] Support type refinement for container types in the generated code,https://github.com/pytorch/pytorch/issues/40070,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-06-16T01:19:48Z
issue,39745,[JIT] exported dunder methods are ignored,https://github.com/pytorch/pytorch/issues/39745,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-06-15T20:54:40Z
issue,39806,Simplify layers of optionals in `VaryingShape` and `Stride`,https://github.com/pytorch/pytorch/issues/39806,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-06-15T20:53:37Z
issue,39719,User Objects aren't recursively scripted as nn.Module attributes,https://github.com/pytorch/pytorch/issues/39719,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-06-15T17:19:25Z
issue,39993,How do I derive weights for CrossEntropy Loss on my custom dataset?,https://github.com/pytorch/pytorch/issues/39993,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: vision,scan when time allows,2020-06-15T14:37:49Z
issue,39972,Port old registration API to new one,https://github.com/pytorch/pytorch/issues/39972,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-06-15T13:49:00Z
issue,39947,[Caffe2] compilation in c++ undefined reference to `caffe2,https://github.com/pytorch/pytorch/issues/39947,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2020-06-12T17:36:22Z
issue,4085,GridSampler behaviours,https://github.com/pytorch/pytorch/issues/4085,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cudnn,scan when time allows,2020-06-12T04:34:00Z
issue,4102,Make the generator tools data model more explicit,https://github.com/pytorch/pytorch/issues/4102,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: codegen,scan when time allows,2020-06-12T04:32:02Z
issue,32078,Reduce degrees of freedom in strides so that they unambiguously specify layout permutation,https://github.com/pytorch/pytorch/issues/32078,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-06-12T02:12:23Z
issue,35609,Randomly error reports,https://github.com/pytorch/pytorch/issues/35609,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-06-11T21:49:35Z
issue,37010,Make it an error to def() an operator multiple times,https://github.com/pytorch/pytorch/issues/37010,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: bootcamp,scan when time allows,2020-06-11T19:46:47Z
issue,39836,Feedback to split certain doc pages into sub-topics,https://github.com/pytorch/pytorch/issues/39836,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2020-06-11T17:24:23Z
issue,39864,Parallelize arguments serde for RPC with TorchScript functions.,https://github.com/pytorch/pytorch/issues/39864,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: performance, module: rpc, oncall: jit",scan when time allows,2020-06-11T16:13:55Z
issue,36572,[1.4.1] cmake3 not found,https://github.com/pytorch/pytorch/issues/36572,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2020-06-11T12:33:32Z
issue,32497,Don't unnecesarily send cleanup dist autograd context RPCs to other nodes,https://github.com/pytorch/pytorch/issues/32497,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, module: rpc",scan when time allows,2020-06-11T05:54:10Z
issue,39064,Add cpack support to CMakeLists.txt,https://github.com/pytorch/pytorch/issues/39064,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cpp",scan when time allows,2020-06-11T01:43:07Z
issue,39799,Tensor.size() API not correctly documented?,https://github.com/pytorch/pytorch/issues/39799,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2020-06-10T21:25:01Z
issue,36570,Libtorch build error when setting both `USE_GLOO` and `USE_SYSTEM_NCCL` to `ON`,https://github.com/pytorch/pytorch/issues/36570,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: docs, module: nccl, oncall: distributed",scan when time allows,2020-06-10T12:48:03Z
issue,39463,JIT test suite has dependencies across tests,https://github.com/pytorch/pytorch/issues/39463,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-06-09T18:52:59Z
issue,30373,Refactor (a bit) `torch.hub(.load)`,https://github.com/pytorch/pytorch/issues/30373,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: hub,scan when time allows,2020-06-09T18:31:31Z
issue,39706,[distributed] calling nccl reduce with inconsistent dst hangs,https://github.com/pytorch/pytorch/issues/39706,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: deadlock, module: nccl, oncall: distributed",scan when time allows,2020-06-09T16:12:17Z
issue,13716,depthwise convolution are slow on cpu,https://github.com/pytorch/pytorch/issues/13716,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: cpu, module: performance",scan when time allows,2020-06-09T16:06:05Z
issue,39705,[doc] [distributed] example of specifying url etc in url,https://github.com/pytorch/pytorch/issues/39705,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, oncall: distributed",scan when time allows,2020-06-09T15:46:55Z
issue,39682,Run `clang-tidy` on the `aten` folder?,https://github.com/pytorch/pytorch/issues/39682,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build warnings, module: lint",scan when time allows,2020-06-09T01:22:51Z
issue,39665,Better err msg for tensor ctor from sequence,https://github.com/pytorch/pytorch/issues/39665,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-06-09T00:28:55Z
issue,39603,friend constexpr in templated struct loses constexpr-ness in nvcc,https://github.com/pytorch/pytorch/issues/39603,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-06-08T17:07:54Z
issue,39226,Try Address Sanitizer in MSVC builds,https://github.com/pytorch/pytorch/issues/39226,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: windows,scan when time allows,2020-06-08T15:58:20Z
issue,39435,BCEWithLogitsLoss() not equal to BCELoss() with sigmoid(),https://github.com/pytorch/pytorch/issues/39435,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: loss,scan when time allows,2020-06-06T22:27:24Z
issue,39625,All keys matched successfully missing when loading state dict on optimizers,https://github.com/pytorch/pytorch/issues/39625,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2020-06-06T21:45:56Z
issue,39619,Switching from CPU build to CUDA build in conda environments is a bit tricky,https://github.com/pytorch/pytorch/issues/39619,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2020-06-06T20:37:37Z
issue,29026,RuntimeError: !t.is_cuda() INTERNAL ASSERT FAILED at /pytorch/aten/src/ATen/native/sparse/SparseTensorMath.cpp:591,https://github.com/pytorch/pytorch/issues/29026,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: assert failure, module: sparse",scan when time allows,2020-06-06T07:57:59Z
issue,39362,"The conversion of Inplace aten::append to outplace is incomplete, creating incorrect subgraphs",https://github.com/pytorch/pytorch/issues/39362,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-06-06T07:05:30Z
issue,39537,Segfault = docker + tensorboard + pytorch,https://github.com/pytorch/pytorch/issues/39537,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tensorboard,scan when time allows,2020-06-06T00:08:12Z
issue,39123,The MacOS compiler is generating illegal instruction for the division of c10::complex,https://github.com/pytorch/pytorch/issues/39123,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: complex,scan when time allows,2020-06-05T14:40:04Z
issue,39524,test_nn_module_tests should run less tests,https://github.com/pytorch/pytorch/issues/39524,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: tests",scan when time allows,2020-06-04T19:42:40Z
issue,27840,caffe2 install VS2019 CUDA 10.1 lib\torch.lib : fatal error LNK1248: image size (10028FA9F) exceeds maximum allowable size (FFFFFFFF),https://github.com/pytorch/pytorch/issues/27840,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-06-04T09:34:39Z
issue,38974,ATen registrable operator list,https://github.com/pytorch/pytorch/issues/38974,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: internals,scan when time allows,2020-06-03T18:15:09Z
issue,39317,send a Tensor to Cuda very slow,https://github.com/pytorch/pytorch/issues/39317,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: performance",scan when time allows,2020-06-03T15:19:00Z
issue,38643,torch while loading weighs found runtime error on storage has wrong size: expected 4254413747647032608 got 1024,https://github.com/pytorch/pytorch/issues/38643,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: serialization,scan when time allows,2020-06-03T00:39:12Z
issue,39399,Update RPC doc to recommend async user functions for non-blocking server execution,https://github.com/pytorch/pytorch/issues/39399,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: rpc",scan when time allows,2020-06-02T15:05:07Z
issue,38973,ATen operator API versioning,https://github.com/pytorch/pytorch/issues/38973,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: internals,scan when time allows,2020-06-02T00:55:33Z
issue,39355,Clean up GIL that used to guard deleted RRef destructions,https://github.com/pytorch/pytorch/issues/39355,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rpc,scan when time allows,2020-06-01T21:37:10Z
issue,32162,Using Tensor.to(device) after distributed all_reduce intermittently causes deadlock with NCCL,https://github.com/pytorch/pytorch/issues/32162,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nccl, oncall: distributed",scan when time allows,2020-06-01T19:46:44Z
issue,38233,When TorchScripted module has bad type annotation you get bad error message,https://github.com/pytorch/pytorch/issues/38233,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, oncall: jit",scan when time allows,2020-06-01T19:04:35Z
issue,38677,TestCase.assertEqual does not distinguish Python builtin types and single-element Tensor,https://github.com/pytorch/pytorch/issues/38677,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tests,scan when time allows,2020-06-01T17:25:03Z
issue,39298,Caffe 2 android app not working with proguard and minifyEnabled,https://github.com/pytorch/pytorch/issues/39298,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-06-01T06:55:33Z
issue,39124,[TensorPipe] Errors in pipeWrite should clear out the future in pendingResponseMessage,https://github.com/pytorch/pytorch/issues/39124,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: rpc, module: tensorpipe",scan when time allows,2020-06-01T01:43:55Z
issue,39280,Let future.wait() take in an optional timeout,https://github.com/pytorch/pytorch/issues/39280,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rpc,scan when time allows,2020-06-01T01:40:47Z
issue,39305,Compilation errors,https://github.com/pytorch/pytorch/issues/39305,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: mkldnn",scan when time allows,2020-06-01T01:40:37Z
issue,39270,Current implementation of c10::complex does not support being used in shared memory,https://github.com/pytorch/pytorch/issues/39270,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: complex,scan when time allows,2020-06-01T01:32:20Z
issue,39130,`torch.backends` undocumented,https://github.com/pytorch/pytorch/issues/39130,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2020-06-01T01:24:25Z
issue,39082,[TensorPipe] Avoid wrapping the future message in order to do atomic test-and-set,https://github.com/pytorch/pytorch/issues/39082,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tensorpipe,scan when time allows,2020-06-01T01:23:23Z
issue,39101,Build issue when installing pytorch with USE_FFMPEG=1 USE_OPENCV=1,https://github.com/pytorch/pytorch/issues/39101,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2020-06-01T01:22:58Z
issue,38051,Div by zero error not triggered and inf not returned when dividing by 0 for some dtypes,https://github.com/pytorch/pytorch/issues/38051,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: numerical-stability, module: type promotion",scan when time allows,2020-05-31T17:59:52Z
issue,30458,add method to make tensor constant for debug purposes,https://github.com/pytorch/pytorch/issues/30458,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2020-05-29T09:26:41Z
issue,39128,"Problems implementing complex support for acosh, asinh, tanh",https://github.com/pytorch/pytorch/issues/39128,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: complex,scan when time allows,2020-05-28T14:56:17Z
issue,38239,Failed to link torch_library using cmake,https://github.com/pytorch/pytorch/issues/38239,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cpp",scan when time allows,2020-05-28T09:05:27Z
issue,38975,Use general ATen dispatch mechanism,https://github.com/pytorch/pytorch/issues/38975,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: internals,scan when time allows,2020-05-28T00:19:13Z
issue,38996,[FR] API consistency for cuda.comm and distributed,https://github.com/pytorch/pytorch/issues/38996,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: docs, oncall: distributed",scan when time allows,2020-05-26T19:34:05Z
issue,38997,[doc] document cuda.nccl,https://github.com/pytorch/pytorch/issues/38997,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: docs, module: nccl",scan when time allows,2020-05-26T19:33:57Z
issue,38980,runtime error while using default arguments in add_graph() function,https://github.com/pytorch/pytorch/issues/38980,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: tensorboard, oncall: visualization",scan when time allows,2020-05-26T18:55:28Z
issue,38863,[JIT] named_(parameters | buffers) do not support recursive iteration.,https://github.com/pytorch/pytorch/issues/38863,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-05-26T17:47:13Z
issue,38782,SPMG in DDP does not have gradient computation and communication overlap?,https://github.com/pytorch/pytorch/issues/38782,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2020-05-26T17:32:58Z
issue,38940,[FR] Support SyncBatchNorm in DataParallel,https://github.com/pytorch/pytorch/issues/38940,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: data parallel, module: nccl",scan when time allows,2020-05-26T17:32:21Z
issue,33668,"/usr/bin/x86_64-linux-gnu-ld: warning: libcusparse.so.10.0, needed by /pytorch_master/build/lib/libtorch_cuda.so, not found (try using -rpath or -rpath-link)",https://github.com/pytorch/pytorch/issues/33668,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cuda",scan when time allows,2020-05-26T16:26:04Z
issue,3791,Make pytest stop printing docstrings in its default diagnostic output,https://github.com/pytorch/pytorch/issues/3791,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tests,scan when time allows,2020-05-26T16:14:48Z
issue,3697,Deprecate inplace argument in torch.nn.functional,https://github.com/pytorch/pytorch/issues/3697,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bc-breaking, module: deprecation, module: nn",scan when time allows,2020-05-26T16:08:17Z
issue,38876,Override `__call__` instead of `forward`,https://github.com/pytorch/pytorch/issues/38876,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: typing,scan when time allows,2020-05-26T13:52:37Z
issue,34879,Addition of Siamese loss or contrastive loss function,https://github.com/pytorch/pytorch/issues/34879,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: loss,scan when time allows,2020-05-25T16:47:47Z
issue,10119,"[Caffe2] MNIST Tutorial LMDB Error ""Cannot open db""",https://github.com/pytorch/pytorch/issues/10119,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-05-24T09:37:02Z
issue,34651,Support passing memoryview to torch.as_tensor,https://github.com/pytorch/pytorch/issues/34651,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-05-23T22:17:44Z
issue,38737,equivalent of tensorflow's embedding_lookup_sparse,https://github.com/pytorch/pytorch/issues/38737,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-05-23T20:21:48Z
issue,37837,[DRAFT] Channels Last + AMP support plan for 1.6 release,https://github.com/pytorch/pytorch/issues/37837,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: memory format,scan when time allows,2020-05-22T23:38:11Z
issue,38936,[FR] DataParallel arg rename device_ids->devices,https://github.com/pytorch/pytorch/issues/38936,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: data parallel,scan when time allows,2020-05-22T19:34:01Z
issue,14699,[Feature Request] linux distribution friendly build system,https://github.com/pytorch/pytorch/issues/14699,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2020-05-22T16:11:13Z
issue,36575,One confusion about the CompilationUnit destructuring process in torch/jit/__init__.py,https://github.com/pytorch/pytorch/issues/36575,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-05-22T03:48:34Z
issue,36378,CMake targets wrongly forward unknown options to NVCC (v1.5+),https://github.com/pytorch/pytorch/issues/36378,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2020-05-21T22:53:32Z
issue,38855,Can not compile GridSamplerKernel.cpp with gcc-9.3,https://github.com/pytorch/pytorch/issues/38855,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2020-05-21T03:56:54Z
issue,38854,Can't compile QuantizedOpKernels.cpp using gcc-9.3,https://github.com/pytorch/pytorch/issues/38854,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2020-05-21T03:56:27Z
issue,38719,Building NVCC (Device) failed when building from source,https://github.com/pytorch/pytorch/issues/38719,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2020-05-20T02:52:23Z
issue,38437,[JIT] torch.tensor needs a Tensor overload,https://github.com/pytorch/pytorch/issues/38437,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, oncall: jit",scan when time allows,2020-05-19T18:24:30Z
issue,38640,Maxunpool seems to give a weird error message,https://github.com/pytorch/pytorch/issues/38640,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-05-19T16:13:13Z
issue,38662,Unify CPU/CUDA exponential transformation formula,https://github.com/pytorch/pytorch/issues/38662,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: random,scan when time allows,2020-05-19T15:40:35Z
issue,38720,DOC: add documentation for undocumented classes and methods,https://github.com/pytorch/pytorch/issues/38720,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2020-05-19T15:40:06Z
issue,38611,Investigate using -cospi(u) / sinpi(u) instead of tan(pi * (u - 0.5)) in transformation::cauchy,https://github.com/pytorch/pytorch/issues/38611,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: distributions, module: random",scan when time allows,2020-05-19T15:21:27Z
issue,38613,Investigate using log1p instead of log in transformation functions(TransformationHelper.h),https://github.com/pytorch/pytorch/issues/38613,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: distributions, module: random",scan when time allows,2020-05-19T15:21:11Z
issue,38619,"Can you add NMS,RoIAlign,RoIPool for libtorch?",https://github.com/pytorch/pytorch/issues/38619,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: vision,scan when time allows,2020-05-19T14:49:52Z
issue,38709,NaN Loss for FasterRCNN on Multiclass Object Detection on Custom Dataset COCO,https://github.com/pytorch/pytorch/issues/38709,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: vision,scan when time allows,2020-05-19T14:43:40Z
issue,38612,Investigate exponential distribution improvements,https://github.com/pytorch/pytorch/issues/38612,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: distributions, module: random",scan when time allows,2020-05-19T14:16:54Z
issue,38595,torch.norm p/ord parameter documentation is wrong,https://github.com/pytorch/pytorch/issues/38595,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2020-05-19T14:05:45Z
issue,37136,[Design][RFC] RemoteModule API Design,https://github.com/pytorch/pytorch/issues/37136,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: rpc, oncall: jit",scan when time allows,2020-05-19T06:31:14Z
issue,38703,torch.as_tensor(np_array) is sometimes much faster than torch.tensor(np_array),https://github.com/pytorch/pytorch/issues/38703,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: numpy, module: performance",scan when time allows,2020-05-19T04:20:56Z
issue,37373,Move bernoulli_() to DistributionTemplates,https://github.com/pytorch/pytorch/issues/37373,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: random,scan when time allows,2020-05-19T01:30:04Z
issue,21794,Linker errors when building project with OpenCV,https://github.com/pytorch/pytorch/issues/21794,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2020-05-17T21:39:26Z
issue,38553,"When I use cuda(), wg = th.matmul(extra_obs, extra_obs.transpose(-2, -1)) take a mistake",https://github.com/pytorch/pytorch/issues/38553,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cublas,scan when time allows,2020-05-17T06:01:51Z
issue,38472,[torch.jit.trace] torch.jit.trace fixed batch size CNN,https://github.com/pytorch/pytorch/issues/38472,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-05-17T04:59:03Z
issue,38474,Scale parameter downcasted and rounded down in pytorch.distributions.Normal,https://github.com/pytorch/pytorch/issues/38474,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distributions,scan when time allows,2020-05-15T23:08:31Z
issue,38564,tensorboard projector mode with custom metadata_header with only one label name,https://github.com/pytorch/pytorch/issues/38564,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tensorboard,scan when time allows,2020-05-15T22:20:17Z
issue,38543,Make rebuildBucket() to be async in c10d reducer,https://github.com/pytorch/pytorch/issues/38543,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: data parallel, oncall: distributed",scan when time allows,2020-05-15T21:49:51Z
issue,35937,MKLDNN_conv2d 2X slower than the native TH implementation,https://github.com/pytorch/pytorch/issues/35937,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: mkldnn, module: performance",scan when time allows,2020-05-15T15:56:24Z
issue,38548,Add Distributed LR Scheduler to RPC,https://github.com/pytorch/pytorch/issues/38548,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, module: rpc",scan when time allows,2020-05-15T14:34:37Z
issue,29816,AdamSparse fails to run,https://github.com/pytorch/pytorch/issues/29816,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2020-05-15T05:42:43Z
issue,38510,[DISCUSSION] RPC server-side ThreadLocalState,https://github.com/pytorch/pytorch/issues/38510,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: multithreading, module: rpc, oncall: jit",scan when time allows,2020-05-14T22:07:06Z
issue,38473,Python builtin function next() is currently not supported in Torchscript,https://github.com/pytorch/pytorch/issues/38473,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-05-14T17:33:22Z
issue,33672,[dev] `RecursiveScriptModule` does not expose `jit.ignore`d methods,https://github.com/pytorch/pytorch/issues/33672,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-05-14T11:53:28Z
issue,38204,missing dependency `protobuf` for caffe2,https://github.com/pytorch/pytorch/issues/38204,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2020-05-14T04:07:45Z
issue,38354,"Can I add more ""Project Documentation"" on the PYPI webpage?",https://github.com/pytorch/pytorch/issues/38354,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: docs",scan when time allows,2020-05-14T01:48:23Z
issue,38319,c10/macros/cmake_macros.h not exists,https://github.com/pytorch/pytorch/issues/38319,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: mobile,scan when time allows,2020-05-13T07:38:24Z
issue,3600,making .cuda() falls back to an identity function when gpu is not available,https://github.com/pytorch/pytorch/issues/3600,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2020-05-12T21:05:26Z
issue,9046,[feature request] freeze() for nn.Module,https://github.com/pytorch/pytorch/issues/9046,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2020-05-12T19:22:36Z
issue,38241,SIGXCPU at test_cholesky_solve with AMD EPYC 7742 64-Core Processor,https://github.com/pytorch/pytorch/issues/38241,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tests,scan when time allows,2020-05-12T16:38:15Z
issue,38208,nn.Module Abstract Class attribute overrides Child.,https://github.com/pytorch/pytorch/issues/38208,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-05-12T16:36:38Z
issue,38212,"Please help, building failing with MAGMA support",https://github.com/pytorch/pytorch/issues/38212,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2020-05-12T16:34:06Z
issue,38170,"OneCycleLR raises ""Tried to step { step_num + 1 } times"" after the value is more than expected.",https://github.com/pytorch/pytorch/issues/38170,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-05-12T16:24:11Z
issue,38323,Support alternate casting rules,https://github.com/pytorch/pytorch/issues/38323,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: numpy, module: type promotion",scan when time allows,2020-05-12T15:56:32Z
issue,38328,Provide an issubdtype API,https://github.com/pytorch/pytorch/issues/38328,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: numpy, module: type promotion",scan when time allows,2020-05-12T15:56:06Z
issue,38272,"PyTorch's rot90 returns a new tensor, inconsistent with NumPy's returning a view",https://github.com/pytorch/pytorch/issues/38272,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: numpy,scan when time allows,2020-05-12T04:26:53Z
issue,33089,test_baddbmm_cpu_float32 fails locally for me when built with DEBUG=1,https://github.com/pytorch/pytorch/issues/33089,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tests,scan when time allows,2020-05-12T02:59:36Z
issue,32868,__cuda_array_interface__ conversion does not support readonly arrays,https://github.com/pytorch/pytorch/issues/32868,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: internals, module: numba",scan when time allows,2020-05-11T23:40:10Z
issue,38245,Make torch.rpc accept store as optional parameter,https://github.com/pytorch/pytorch/issues/38245,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rpc,scan when time allows,2020-05-11T20:10:07Z
issue,18283,Doubly freed pointer in torch::cat error handling when called via pybind11.,https://github.com/pytorch/pytorch/issues/18283,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: error checking",scan when time allows,2020-05-11T17:42:42Z
issue,18776,Value of torch.backends.cudnn.benchmark Baked into JIT-Traced Modules ( 150x slowdown on ConvTranspose2d() ) [jit] [libtorch] [cudnn],https://github.com/pytorch/pytorch/issues/18776,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-05-11T09:54:55Z
issue,38009,Resnet Model always predicting same label,https://github.com/pytorch/pytorch/issues/38009,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: vision,scan when time allows,2020-05-11T09:16:11Z
issue,38202,IndexError reports the wrong dimension when fancy indexing,https://github.com/pytorch/pytorch/issues/38202,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: error checking,scan when time allows,2020-05-10T15:02:52Z
issue,38177,valgrind says libtorch has memory leak,https://github.com/pytorch/pytorch/issues/38177,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: memory usage,scan when time allows,2020-05-10T14:22:47Z
issue,21779,"Use of word ""elements"" in `torch.utils.data` samplers",https://github.com/pytorch/pytorch/issues/21779,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2020-05-10T09:27:23Z
issue,35414,"MacOS Error: subprocess.CalledProcessError: Command '['cmake', '--build', '.', '--target', 'install', '--config', 'Release', '--', '-j', '12']' returned non-zero exit status 1.",https://github.com/pytorch/pytorch/issues/35414,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cuda, module: macos",scan when time allows,2020-05-09T15:25:03Z
issue,38190,test_ReplicationPad3d (test_nn.TestNN) takes too long to run,https://github.com/pytorch/pytorch/issues/38190,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tests,scan when time allows,2020-05-09T07:45:21Z
issue,38189,test_LocalResponseNorm_3d_custom_params (test_nn.TestNN) takes too long to run,https://github.com/pytorch/pytorch/issues/38189,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tests,scan when time allows,2020-05-09T07:44:46Z
issue,38188,test_interpolate_nearest_scale_3d in test_nn takes too long to run,https://github.com/pytorch/pytorch/issues/38188,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tests,scan when time allows,2020-05-09T07:43:57Z
issue,18351,Port SpatialConvolutionMM and VolumetricConvolutionMM to ATen,https://github.com/pytorch/pytorch/issues/18351,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: porting,scan when time allows,2020-05-09T07:16:18Z
issue,7890,[feature request] batch_first of RNN hidden weight for Multi GPU training,https://github.com/pytorch/pytorch/issues/7890,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: data parallel, module: nn",scan when time allows,2020-05-09T06:35:39Z
issue,38184,Feature Request: Log uniform candidate sampler,https://github.com/pytorch/pytorch/issues/38184,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2020-05-09T06:18:29Z
issue,2691,Feature Request: Support grad of grad in fused RNNs,https://github.com/pytorch/pytorch/issues/2691,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: double backwards, module: nn",scan when time allows,2020-05-09T06:09:52Z
issue,1932,Feature Request: ReLU on LSTMs and GRUs,https://github.com/pytorch/pytorch/issues/1932,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: rnn",scan when time allows,2020-05-09T05:58:48Z
issue,1686,Feature Request: load_state_dict should take filenames,https://github.com/pytorch/pytorch/issues/1686,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2020-05-09T05:53:37Z
issue,4987,MultiGPU hangs Titan Xp in multiprocessing/queue.py,https://github.com/pytorch/pytorch/issues/4987,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: macos, module: multiprocessing",scan when time allows,2020-05-08T22:27:00Z
issue,38142,[JIT] Aliased Python references to script::Object escape IValue reference counting,https://github.com/pytorch/pytorch/issues/38142,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-05-08T22:02:08Z
issue,37985,torch.save incompatible with lzma file,https://github.com/pytorch/pytorch/issues/37985,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: serialization,scan when time allows,2020-05-08T16:06:53Z
issue,38118,Add helpers to save/load RPC-based models,https://github.com/pytorch/pytorch/issues/38118,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rpc,scan when time allows,2020-05-08T14:35:39Z
issue,38010,Split more pages into sub-topics,https://github.com/pytorch/pytorch/issues/38010,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: doc infra, module: docs",scan when time allows,2020-05-08T14:12:43Z
issue,27958,how to use libtorch library in cuda file with nvcc compiler(c++)?,https://github.com/pytorch/pytorch/issues/27958,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp,scan when time allows,2020-05-08T08:30:40Z
issue,24463,Crashes on torch.cuda.memory_allocated(device),https://github.com/pytorch/pytorch/issues/24463,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: error checking,scan when time allows,2020-05-08T05:42:47Z
issue,38012,Formatting of topic and sub-topic pages,https://github.com/pytorch/pytorch/issues/38012,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: doc infra, module: docs",scan when time allows,2020-05-08T03:35:44Z
issue,33867,cuDNN batchnorm with non-contiguous running mean silently discards updates,https://github.com/pytorch/pytorch/issues/33867,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cudnn,scan when time allows,2020-05-07T18:12:31Z
issue,38000,"New padding size format for F.pad, allowing named tensors and more clear syntax overall",https://github.com/pytorch/pytorch/issues/38000,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2020-05-07T17:36:07Z
issue,7786,[feature request] Simple and Efficient way to get gradients of each element of a sum,https://github.com/pytorch/pytorch/issues/7786,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2020-05-07T14:41:06Z
issue,37988,JIT compilation of NamedTuple Containing a NamedTuple fails,https://github.com/pytorch/pytorch/issues/37988,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-05-07T03:43:38Z
issue,37987,Special methods on torchscript custom class,https://github.com/pytorch/pytorch/issues/37987,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-05-07T03:40:12Z
issue,37860,autograd engine callbacks don't respect non-default cuda streams,https://github.com/pytorch/pytorch/issues/37860,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: cuda",scan when time allows,2020-05-06T21:11:40Z
issue,37880,"[RFC] Don't install CI dependencies in build scripts, install them in underlying docker images",https://github.com/pytorch/pytorch/issues/37880,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2020-05-06T20:54:53Z
issue,37895,torch.addmv can't take as input tensors with different dtypes,https://github.com/pytorch/pytorch/issues/37895,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: type promotion,scan when time allows,2020-05-06T18:14:35Z
issue,37814,DistributedDataParallel does not support Modules that take no inputs.,https://github.com/pytorch/pytorch/issues/37814,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: data parallel,scan when time allows,2020-05-06T02:10:05Z
issue,3281,"DataLoader ""casting"" non statndard objects to lists",https://github.com/pytorch/pytorch/issues/3281,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2020-05-05T23:27:37Z
issue,3364,type of torch.bernoulli and torch.multinomial inconsistent,https://github.com/pytorch/pytorch/issues/3364,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distributions,scan when time allows,2020-05-05T23:15:28Z
issue,3390,High CPU use by clock_gettime syscall,https://github.com/pytorch/pytorch/issues/3390,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: performance",scan when time allows,2020-05-05T23:14:36Z
issue,37847,llvmlite version issue when upgrading CI docker image to python 3.8,https://github.com/pytorch/pytorch/issues/37847,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2020-05-05T22:58:29Z
issue,37588,"test_qnnpack_sigmoid failing on old CPU, built pytorch 1.5.0",https://github.com/pytorch/pytorch/issues/37588,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: build",scan when time allows,2020-05-05T21:24:54Z
issue,27752,Version number is still duplicated in a bunch of places,https://github.com/pytorch/pytorch/issues/27752,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: binaries,scan when time allows,2020-05-05T20:37:33Z
issue,34606,Multiple CMake target errors ever since commit 0e52627358,https://github.com/pytorch/pytorch/issues/34606,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2020-05-05T19:52:26Z
issue,37512,Improve visibility in test suite timings,https://github.com/pytorch/pytorch/issues/37512,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ci, module: tests",scan when time allows,2020-05-05T05:37:35Z
issue,36804,c++: error: unrecognized command line option '-Wthread-safety',https://github.com/pytorch/pytorch/issues/36804,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2020-05-05T05:05:29Z
issue,37629,logging_is_not_google_glog.h:24:11: error: 'const int ERROR' redeclared as different kind of symbol (v1.3.0),https://github.com/pytorch/pytorch/issues/37629,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2020-05-05T03:04:02Z
issue,7250,feature request: support for new/future hardware accelerators,https://github.com/pytorch/pytorch/issues/7250,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-05-05T00:53:38Z
issue,37684,"Add Compound key, make custom ops default to it (but keep internal users using CatchAll)",https://github.com/pytorch/pytorch/issues/37684,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-05-05T00:31:39Z
issue,37731,Variational Dropout In RNN,https://github.com/pytorch/pytorch/issues/37731,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rnn,scan when time allows,2020-05-04T22:04:18Z
issue,37782,Guard Gloo and TensorPipe related code in RPC with #ifdef,https://github.com/pytorch/pytorch/issues/37782,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, module: rpc, module: tensorpipe",scan when time allows,2020-05-04T19:19:51Z
issue,37457,[RuntimeError] Tensor creation using storage fails,https://github.com/pytorch/pytorch/issues/37457,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: error checking,scan when time allows,2020-05-02T23:24:31Z
issue,37529,Implement generic function scheduler in c10/util,https://github.com/pytorch/pytorch/issues/37529,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rpc,scan when time allows,2020-05-02T01:38:47Z
issue,37444,Multi-Process Single-GPU is bad,https://github.com/pytorch/pytorch/issues/37444,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: data parallel, oncall: distributed",scan when time allows,2020-05-02T00:29:40Z
issue,37434,rsub incorrectly exposed in torch,https://github.com/pytorch/pytorch/issues/37434,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-05-01T23:03:42Z
issue,37549,Are there any differences in kernel memory between RX2080's and Quadro RTX4000?,https://github.com/pytorch/pytorch/issues/37549,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: memory usage",scan when time allows,2020-05-01T21:50:34Z
issue,37510,"Cannot build pytorch with linker arguments in C{,XX}FLAGS",https://github.com/pytorch/pytorch/issues/37510,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2020-05-01T21:02:19Z
issue,30929,How to set not to build libtorch_cpu.so and libmkl_*.so dependencies?,https://github.com/pytorch/pytorch/issues/30929,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: mkl",scan when time allows,2020-05-01T18:47:25Z
issue,37598,NCCL fails to find cuda include dir,https://github.com/pytorch/pytorch/issues/37598,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nccl, oncall: distributed",scan when time allows,2020-05-01T13:10:43Z
issue,37441,Tensor.as_strided_ is not documented,https://github.com/pytorch/pytorch/issues/37441,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2020-05-01T04:05:53Z
issue,37440,Tensor.is_same_size not documented,https://github.com/pytorch/pytorch/issues/37440,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2020-05-01T04:04:15Z
issue,3076,BN slows down double-backprop enormously,https://github.com/pytorch/pytorch/issues/3076,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: performance,scan when time allows,2020-04-30T22:07:49Z
issue,32444,DataParallel does not work with sparse parameters,https://github.com/pytorch/pytorch/issues/32444,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: data parallel,scan when time allows,2020-04-30T20:56:19Z
issue,32919,"Torchscript used to work, but now it fails with VariableTensorId error",https://github.com/pytorch/pytorch/issues/32919,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: android, oncall: jit, oncall: mobile",scan when time allows,2020-04-30T17:07:25Z
issue,31579,Option to apply weights to gradients when using DistributedDataParallel,https://github.com/pytorch/pytorch/issues/31579,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: data parallel, oncall: distributed",scan when time allows,2020-04-30T02:21:52Z
issue,37484,No speedup from channels_last with DataParallel,https://github.com/pytorch/pytorch/issues/37484,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: data parallel, module: memory format, module: performance",scan when time allows,2020-04-29T23:33:07Z
issue,37282,python setup.py install error,https://github.com/pytorch/pytorch/issues/37282,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2020-04-29T23:21:40Z
issue,36058,Half type promotion with Numpy arrays is incorrect,https://github.com/pytorch/pytorch/issues/36058,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: numpy,scan when time allows,2020-04-29T19:19:15Z
issue,33327,Creating Torch tensors slows OpenCV video reading a lot,https://github.com/pytorch/pytorch/issues/33327,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: multithreading, module: performance",scan when time allows,2020-04-29T18:48:41Z
issue,36307,[docs] Matrix transpose missing in formula in nn.functional.bilinear docs,https://github.com/pytorch/pytorch/issues/36307,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2020-04-28T17:05:49Z
issue,37416,[docs] Unclear return type of torch.randint and extra comma in arg spec,https://github.com/pytorch/pytorch/issues/37416,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2020-04-28T15:55:04Z
issue,33055,[docs] Strange order of items in docs contents in left pane,https://github.com/pytorch/pytorch/issues/33055,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2020-04-28T15:32:15Z
issue,37153,CPU out of bound memory access in CUDA reduction kernel config,https://github.com/pytorch/pytorch/issues/37153,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: tensoriterator",scan when time allows,2020-04-28T14:38:13Z
issue,31615,'torch.load' report 'bad pickle data',https://github.com/pytorch/pytorch/issues/31615,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: serialization,scan when time allows,2020-04-28T09:09:27Z
issue,34998,Better examples in functional autograd functions,https://github.com/pytorch/pytorch/issues/34998,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: docs",scan when time allows,2020-04-27T18:23:27Z
issue,37204,Make it harder to make SIOF bugs for torchbind classes referenced by schemas,https://github.com/pytorch/pytorch/issues/37204,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-04-26T20:30:05Z
issue,37300,Data caching module a la `Sampler`,https://github.com/pytorch/pytorch/issues/37300,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-04-26T20:18:14Z
issue,37314,equal_nan keyword not implemented for complex torch.isclose,https://github.com/pytorch/pytorch/issues/37314,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: complex, module: numpy",scan when time allows,2020-04-26T09:22:37Z
issue,20613,groupby function,https://github.com/pytorch/pytorch/issues/20613,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-04-25T14:31:57Z
issue,36004,C++ tensor print doesn't show requires_grad and grad_fn like Python tensor print,https://github.com/pytorch/pytorch/issues/36004,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: printing",scan when time allows,2020-04-25T01:06:23Z
issue,37103,Legacy fuser doesn't do remainder consistently with `aten::remainder`,https://github.com/pytorch/pytorch/issues/37103,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-04-24T01:26:11Z
issue,37149,Cuda profiler + DataParallel + manual profiling start = strange profiling overhead pattern,https://github.com/pytorch/pytorch/issues/37149,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2020-04-23T22:54:00Z
issue,36713,variable name N_ conflicts with an internationalization macro in glib,https://github.com/pytorch/pytorch/issues/36713,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, oncall: jit",scan when time allows,2020-04-23T20:38:19Z
issue,8550,caffe2: Does caffe support the model running on different device (CPU and GPU) same time ?,https://github.com/pytorch/pytorch/issues/8550,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-04-23T07:58:47Z
issue,36917,"jit trace failed due to ""failed to differentiate `prim::ListConstruct`""",https://github.com/pytorch/pytorch/issues/36917,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-04-22T19:04:11Z
issue,36931,Pull hacked twins out of prim ops,https://github.com/pytorch/pytorch/issues/36931,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-04-22T19:00:16Z
issue,37065,Deprecated mask fill mask type can causes pages and pages of repeated messages,https://github.com/pytorch/pytorch/issues/37065,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-04-22T18:10:02Z
issue,3152,Sparse tensor .new(size) can be confusing,https://github.com/pytorch/pytorch/issues/3152,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2020-04-22T17:50:48Z
issue,3176,Add safety checks in `index_add`/`scatter_add`,https://github.com/pytorch/pytorch/issues/3176,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-04-22T17:49:29Z
issue,36754,Clearer guidance on when to define an operator as a method on torchbind'ed class versus standalone function,https://github.com/pytorch/pytorch/issues/36754,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-04-22T14:49:47Z
issue,28863,Command for downloading torch 1.12.0 CUDA 10 linked to cu92 version,https://github.com/pytorch/pytorch/issues/28863,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: binaries,scan when time allows,2020-04-22T12:18:49Z
issue,37004,"RuntimeError: NCCL error in ProcessGroupNCCL.cpp:290, unhandled system error",https://github.com/pytorch/pytorch/issues/37004,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dependency bug, module: nccl, oncall: distributed",scan when time allows,2020-04-22T01:52:04Z
issue,36739,Custom class type name is very wordy,https://github.com/pytorch/pytorch/issues/36739,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-04-21T20:35:37Z
issue,36874,Memory leak upon exception for interactive consoles,https://github.com/pytorch/pytorch/issues/36874,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: memory usage,scan when time allows,2020-04-21T16:55:39Z
issue,36999,Add a CI configuration to test USE_DISTRIBUTED=0,https://github.com/pytorch/pytorch/issues/36999,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ci, module: tests, oncall: distributed",scan when time allows,2020-04-21T14:33:23Z
issue,32018,DistributedDataParallel non-floating point dtype parameter with requires_grad=False,https://github.com/pytorch/pytorch/issues/32018,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2020-04-20T19:12:24Z
issue,36891,pytorch and c++ inference disagree,https://github.com/pytorch/pytorch/issues/36891,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp,scan when time allows,2020-04-20T16:42:27Z
issue,36792,[1.4.1] Intel MPI libs not found.,https://github.com/pytorch/pytorch/issues/36792,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: mkldnn",scan when time allows,2020-04-20T07:02:53Z
issue,17774,RuntimeError: cuDNN error: CUDNN_STATUS_EXECUTION_FAILED when batch size is too large,https://github.com/pytorch/pytorch/issues/17774,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cudnn,scan when time allows,2020-04-18T17:46:16Z
issue,36638,Poor elmenetwise_kernel performance becomes critical on small mini-batch sizes,https://github.com/pytorch/pytorch/issues/36638,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: performance, module: tensoriterator",scan when time allows,2020-04-18T05:27:18Z
issue,36753,When you try to register a kernel that make boxed from unboxed functor doesn't support you get a horrible error message,https://github.com/pytorch/pytorch/issues/36753,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-04-18T04:48:12Z
issue,36733,Custom Generators don't work in JIT,https://github.com/pytorch/pytorch/issues/36733,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-04-17T21:15:43Z
issue,36517,XNNPACK operators are not actually registered under xnnpack namespace,https://github.com/pytorch/pytorch/issues/36517,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: xnnpack,scan when time allows,2020-04-16T22:33:52Z
issue,36407,Unable to build wheel from RC3,https://github.com/pytorch/pytorch/issues/36407,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2020-04-16T04:27:06Z
issue,34754,General Purpose Faulty RPC Agent,https://github.com/pytorch/pytorch/issues/34754,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: rpc, oncall: distributed",scan when time allows,2020-04-14T19:08:43Z
issue,36444,Comparison ops for Complex Tensors,https://github.com/pytorch/pytorch/issues/36444,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: complex, module: numpy",scan when time allows,2020-04-14T16:15:06Z
issue,36553,Typecasting issue in MSELoss,https://github.com/pytorch/pytorch/issues/36553,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: loss, module: type promotion",scan when time allows,2020-04-14T14:20:53Z
issue,36519,Restructure test_c10d.py and test_distributed.py,https://github.com/pytorch/pytorch/issues/36519,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2020-04-13T21:33:11Z
issue,36516,DDP should divide bucket contents by the number of global replicas instead of world size,https://github.com/pytorch/pytorch/issues/36516,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2020-04-13T21:30:44Z
issue,19911,"BatchNorm1d does not support batchsize>65535 in eval mode with 3 dimension (NxCxL), raise CUDNN_STATUS_NOT_SUPPORTED",https://github.com/pytorch/pytorch/issues/19911,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cudnn, module: dependency bug",scan when time allows,2020-04-13T15:06:19Z
issue,31264,Failed to config caffe2_rocksdb in cmake,https://github.com/pytorch/pytorch/issues/31264,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2020-04-13T05:18:58Z
issue,36436,"After `create_graph=True`, calculating `backward()` on sparse Tensor fails",https://github.com/pytorch/pytorch/issues/36436,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-04-13T04:51:04Z
issue,36469,Wrong results for multiplication of non-finite complex numbers with real numbers,https://github.com/pytorch/pytorch/issues/36469,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: complex, module: numpy",scan when time allows,2020-04-13T04:28:25Z
issue,36356,Memory leak issue still exists in CI,https://github.com/pytorch/pytorch/issues/36356,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: ci",scan when time allows,2020-04-13T04:18:25Z
issue,36370,Jit doesn't match schema like eager mode does,https://github.com/pytorch/pytorch/issues/36370,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-04-10T20:59:45Z
issue,36317,Updating a ModelDict instance with another ModelDict instance generates error.,https://github.com/pytorch/pytorch/issues/36317,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2020-04-10T13:50:11Z
issue,32916,from torch._C import * (ImportError: DLL load failed),https://github.com/pytorch/pytorch/issues/32916,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-04-10T11:40:08Z
issue,32370,Backward `Functional.conv3d` is slow when cuDNN is enabled,https://github.com/pytorch/pytorch/issues/32370,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: cudnn",scan when time allows,2020-04-09T16:05:28Z
issue,32287,Python 3.8 Windows JIT test failure,https://github.com/pytorch/pytorch/issues/32287,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-04-09T15:40:10Z
issue,34715,Maybe gpu_kernel shouldn't ASSERT_HOST_DEVICE_LAMBDA,https://github.com/pytorch/pytorch/issues/34715,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-04-09T10:25:02Z
issue,23110,[RFC] RPC Based Distributed Model Parallel,https://github.com/pytorch/pytorch/issues/23110,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rpc,scan when time allows,2020-04-09T04:38:01Z
issue,35942,Build PyTorch-1.4.0 from source failed,https://github.com/pytorch/pytorch/issues/35942,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2020-04-09T01:00:04Z
issue,36160,Decouple DDP from CUDA,https://github.com/pytorch/pytorch/issues/36160,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, oncall: distributed",scan when time allows,2020-04-08T08:20:37Z
issue,14219,C++ model load error,https://github.com/pytorch/pytorch/issues/14219,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-04-07T18:35:59Z
issue,36140,Inconsistent ProcessGroupMPI work data structure for send/recv and collectives,https://github.com/pytorch/pytorch/issues/36140,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: performance, oncall: distributed",scan when time allows,2020-04-07T14:28:18Z
issue,36099,"In AutogradContext, get_saved_variables() should be renamed to get_saved_tensors()",https://github.com/pytorch/pytorch/issues/36099,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp,scan when time allows,2020-04-06T20:59:03Z
issue,36062,RuntimeError CUDA error despite CUDA available and GPU supported,https://github.com/pytorch/pytorch/issues/36062,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: cuda, module: error checking",scan when time allows,2020-04-06T20:23:02Z
issue,35944,"Copy activations from one parts to another part in tensor, but report error",https://github.com/pytorch/pytorch/issues/35944,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-04-06T17:25:31Z
issue,35832,Linking error after marking an op `manual_kernel_registration: True`,https://github.com/pytorch/pytorch/issues/35832,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2020-04-06T17:17:02Z
issue,35964,TensorOptions shouldn't provide default values,https://github.com/pytorch/pytorch/issues/35964,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-04-06T16:41:17Z
issue,36028,About tensorboard in pytorch record graph in different state,https://github.com/pytorch/pytorch/issues/36028,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: visualization,scan when time allows,2020-04-06T14:59:10Z
issue,36059,Some tutorials cannot be found from both side panel and tutorial welcome page,https://github.com/pytorch/pytorch/issues/36059,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2020-04-06T14:50:46Z
issue,30528,Warn about driver CUDA mismatch in torch.cuda.is_available(),https://github.com/pytorch/pytorch/issues/30528,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2020-04-05T12:52:11Z
issue,29093,error: 'SO_REUSEPORT' was not declared in this scope,https://github.com/pytorch/pytorch/issues/29093,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: third_party",scan when time allows,2020-04-04T22:41:03Z
issue,35788,[jit] Structural typing shouldn't work,https://github.com/pytorch/pytorch/issues/35788,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-04-03T21:07:56Z
issue,31916,I'm not able to build pytorch with tensorrt (current master),https://github.com/pytorch/pytorch/issues/31916,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2020-04-01T19:24:29Z
issue,35820,[JIT] Tracer bakes List[Tensor] attribute as constant,https://github.com/pytorch/pytorch/issues/35820,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-04-01T19:08:42Z
issue,26139,c10 List API hard to use,https://github.com/pytorch/pytorch/issues/26139,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: internals,scan when time allows,2020-04-01T17:15:12Z
issue,35766,Clarify tensor storage communication behavior in RPC,https://github.com/pytorch/pytorch/issues/35766,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: rpc",scan when time allows,2020-04-01T16:49:01Z
issue,35655,Wrong conv2d output on GPU when kernel has many zeros,https://github.com/pytorch/pytorch/issues/35655,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: cuda, module: dependency bug, module: numerical-stability",scan when time allows,2020-04-01T13:32:19Z
issue,35770,BatchNormFuncOptions object cant be printed in C++,https://github.com/pytorch/pytorch/issues/35770,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: nn",scan when time allows,2020-04-01T13:25:23Z
issue,35749,Cannot JIT functions with custom backwards (e.g. swish),https://github.com/pytorch/pytorch/issues/35749,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, oncall: jit",scan when time allows,2020-04-01T13:24:28Z
issue,35759,how to do 3d data augmentation in parallel on the gpu?,https://github.com/pytorch/pytorch/issues/35759,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: dataloader",scan when time allows,2020-04-01T13:23:34Z
issue,35755,Expected object of scalar type Double but got scalar type Float for argument #2 'mat2' in call to _th_mm,https://github.com/pytorch/pytorch/issues/35755,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-03-31T20:38:48Z
issue,35758,UserRRef should store error if it sees any and prevent subsequent usage,https://github.com/pytorch/pytorch/issues/35758,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rpc,scan when time allows,2020-03-31T14:58:52Z
issue,35681,TensorBoard add_scalars throws error when dict has keys of type int,https://github.com/pytorch/pytorch/issues/35681,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-03-31T08:27:00Z
issue,34939,Enable OpaqueTensor to possess Storage then allow it to view from CPUTensor,https://github.com/pytorch/pytorch/issues/34939,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: internals, module: mkldnn",scan when time allows,2020-03-31T06:45:20Z
issue,33227,[v1.5] Python/C++ API parity master tracking task,https://github.com/pytorch/pytorch/issues/33227,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp,scan when time allows,2020-03-31T01:58:00Z
issue,12764,[feature request] Operator Overloading,https://github.com/pytorch/pytorch/issues/12764,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-03-30T23:00:30Z
issue,35648,Caffe2 ReshapeOpGPUTest crashes on Windows,https://github.com/pytorch/pytorch/issues/35648,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tests,scan when time allows,2020-03-30T22:47:48Z
issue,35651,Caffe2 utility_ops_gpu_test fails on Windows,https://github.com/pytorch/pytorch/issues/35651,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tests,scan when time allows,2020-03-30T22:47:38Z
issue,35640,How do you change Adam learning rate since the latest commits ?,https://github.com/pytorch/pytorch/issues/35640,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp,scan when time allows,2020-03-30T22:38:52Z
issue,2854,Add the new lr_scheduler which called poly,https://github.com/pytorch/pytorch/issues/2854,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2020-03-30T22:25:27Z
issue,2849,Learning rate scheduler have different APIs,https://github.com/pytorch/pytorch/issues/2849,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2020-03-30T22:25:17Z
issue,35678,USE_AVX/USE_AVX2 does not affect __AVX2__ macro defition,https://github.com/pytorch/pytorch/issues/35678,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build warnings,scan when time allows,2020-03-30T22:21:17Z
issue,35671,allgather_coalesced for tensors of different types seems to be broken,https://github.com/pytorch/pytorch/issues/35671,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2020-03-30T22:14:06Z
issue,35408,torch.jit.script works for x*sigmoid(x) but not for x*sin(x),https://github.com/pytorch/pytorch/issues/35408,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-03-30T03:39:58Z
issue,35118,Customize batch size based on gpu id,https://github.com/pytorch/pytorch/issues/35118,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: data parallel,scan when time allows,2020-03-30T00:38:30Z
issue,31822,Libtorch's files conflict with glog's file?,https://github.com/pytorch/pytorch/issues/31822,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2020-03-28T17:43:10Z
issue,28746,super().__init__() not called in torch.nn.Module.__init__,https://github.com/pytorch/pytorch/issues/28746,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2020-03-27T12:39:08Z
issue,9604,WARNING:root:This caffe2 python run does not have GPU support. Will run in CPU only mode. CRITICAL:root:Cannot load caffe2.python. Error: libcaffe2.so: cannot open shared object file: No such file or directory,https://github.com/pytorch/pytorch/issues/9604,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-03-27T08:00:11Z
issue,34367,Better testing of the autograd engine,https://github.com/pytorch/pytorch/issues/34367,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2020-03-27T02:02:57Z
issue,2739,Proposal: simplify overloaded Tensor function signatures,https://github.com/pytorch/pytorch/issues/2739,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-03-27T01:31:15Z
issue,35160,Documentation doesn't cover MWE using launch.py script,https://github.com/pytorch/pytorch/issues/35160,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2020-03-25T21:43:23Z
issue,34300,Inconsistent semantics of converting inf/-inf to long,https://github.com/pytorch/pytorch/issues/34300,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2020-03-25T16:28:07Z
issue,35234,Raspberry Pi Zero build fails,https://github.com/pytorch/pytorch/issues/35234,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2020-03-25T14:16:17Z
issue,35035,How install old version pytorch 1.2.0 from source?,https://github.com/pytorch/pytorch/issues/35035,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2020-03-24T19:58:57Z
issue,35208,The BatchNorm error in `DataParallel`,https://github.com/pytorch/pytorch/issues/35208,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: data parallel, module: nn",scan when time allows,2020-03-24T19:45:29Z
issue,35078,Generator C++ API should match Python API,https://github.com/pytorch/pytorch/issues/35078,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: random",scan when time allows,2020-03-24T17:38:10Z
issue,35097,undefined symbols when using libtorch and ITK,https://github.com/pytorch/pytorch/issues/35097,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2020-03-23T21:12:43Z
issue,22961,performance much worse on 2080ti than 1080ti,https://github.com/pytorch/pytorch/issues/22961,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: performance",scan when time allows,2020-03-23T19:12:34Z
issue,35040,Deprecate and remove RegisterOperators,https://github.com/pytorch/pytorch/issues/35040,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dispatch,scan when time allows,2020-03-23T13:51:03Z
issue,35038,Rename Dispatcher::findSchema to Dispatcher::findOperator,https://github.com/pytorch/pytorch/issues/35038,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dispatch,scan when time allows,2020-03-23T13:50:20Z
issue,35128,Autograd Engine leaks reentrant threadpool threads on deletion,https://github.com/pytorch/pytorch/issues/35128,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2020-03-20T23:26:03Z
issue,34878,Set the number of CUDAStream,https://github.com/pytorch/pytorch/issues/34878,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2020-03-20T16:32:34Z
issue,2703,Have ppc64le docker images?,https://github.com/pytorch/pytorch/issues/2703,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: power,scan when time allows,2020-03-20T05:21:34Z
issue,2724,DataLoader gets stuck after model initialization,https://github.com/pytorch/pytorch/issues/2724,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2020-03-20T05:20:33Z
issue,34483,JIT: Tracing faster than scripting,https://github.com/pytorch/pytorch/issues/34483,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-03-19T14:00:50Z
issue,32641,changed format of trace graph in torch 1.4.0,https://github.com/pytorch/pytorch/issues/32641,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-03-19T10:13:34Z
issue,34941,Issues with DataParallel on Multiple GPUs,https://github.com/pytorch/pytorch/issues/34941,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: data parallel,scan when time allows,2020-03-19T08:16:38Z
issue,29190,How to run two different jit models in two GPUs respectively in one scrip?,https://github.com/pytorch/pytorch/issues/29190,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-03-19T06:14:43Z
issue,27975,"Sending CUDA tensors via queue between processes, memory of Consumer process grows infinitely",https://github.com/pytorch/pytorch/issues/27975,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: memory usage, module: multiprocessing",scan when time allows,2020-03-19T02:47:41Z
issue,31478,DataParallel has different tensor copy behavior if batch size = 1,https://github.com/pytorch/pytorch/issues/31478,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: data parallel,scan when time allows,2020-03-19T00:23:29Z
issue,34949,Equality operator for torch.distribution.*,https://github.com/pytorch/pytorch/issues/34949,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distributions,scan when time allows,2020-03-18T23:32:24Z
issue,34812,"Semantic differences between forward and backward of upsampling2d/3d in channels last (NHWC, NDHWC) format.",https://github.com/pytorch/pytorch/issues/34812,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: memory format,scan when time allows,2020-03-18T18:57:03Z
issue,34412,error in pytorch Docs: https://pytorch.org/docs/stable/distributions.html,https://github.com/pytorch/pytorch/issues/34412,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2020-03-18T18:42:21Z
issue,34210,[docs] Unclear arg spec for torch.full,https://github.com/pytorch/pytorch/issues/34210,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: numpy",scan when time allows,2020-03-18T16:37:09Z
issue,34833,NO_CUDA and NO_DISTRIBUTED referenced in docs,https://github.com/pytorch/pytorch/issues/34833,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2020-03-18T15:05:29Z
issue,34765,Add ability to return a copy to Module's state_dict member function.,https://github.com/pytorch/pytorch/issues/34765,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2020-03-18T14:42:02Z
issue,34881,libtorch memory leak,https://github.com/pytorch/pytorch/issues/34881,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: memory usage",scan when time allows,2020-03-18T03:17:01Z
issue,31467,Distributed hangs on process termination with world_size=1,https://github.com/pytorch/pytorch/issues/31467,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, oncall: distributed",scan when time allows,2020-03-18T02:39:06Z
issue,25416,Add GPU support to c10d allreduce_coalesced,https://github.com/pytorch/pytorch/issues/25416,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2020-03-17T22:55:52Z
issue,29108,a retrained and saved jit module could not be reload.,https://github.com/pytorch/pytorch/issues/29108,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-03-16T23:17:22Z
issue,31545,IndexExpressions (or slice) for jit.script functions,https://github.com/pytorch/pytorch/issues/31545,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-03-16T21:04:53Z
issue,34728,CPU softmax performance poor when dim is not the last dimension,https://github.com/pytorch/pytorch/issues/34728,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: performance",scan when time allows,2020-03-16T20:21:06Z
issue,25267,JIT leaks memory when I change the max sequence length,https://github.com/pytorch/pytorch/issues/25267,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-03-16T19:13:57Z
issue,34814,[JIT] Tensor.add_ incorrect schema matching on overloads,https://github.com/pytorch/pytorch/issues/34814,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-03-16T17:02:09Z
issue,32214,Pytorch compilation error,https://github.com/pytorch/pytorch/issues/32214,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2020-03-16T12:38:38Z
issue,32851,Silent failing of batch_sampler when the data points are lists of tensors.,https://github.com/pytorch/pytorch/issues/32851,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dataloader, module: nestedtensor",scan when time allows,2020-03-13T21:59:44Z
issue,34649,Confusing error message of tensor constructor when passing a storage,https://github.com/pytorch/pytorch/issues/34649,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-03-12T22:42:00Z
issue,2628,BatchNorm{1-2-3}d are redundant,https://github.com/pytorch/pytorch/issues/2628,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2020-03-12T22:34:15Z
issue,34323,PyTorch GPU memory allocation,https://github.com/pytorch/pytorch/issues/34323,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2020-03-12T22:28:40Z
issue,6760,[Feature request] LayerNormLSTMCell and LayerNormLSTM,https://github.com/pytorch/pytorch/issues/6760,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-03-12T15:24:17Z
issue,33129,Python package using CMake,https://github.com/pytorch/pytorch/issues/33129,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2020-03-12T00:46:20Z
issue,34621,A better way to show users all build options,https://github.com/pytorch/pytorch/issues/34621,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2020-03-11T23:43:21Z
issue,33166,TensorIterator does not work with different input/output types,https://github.com/pytorch/pytorch/issues/33166,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: tensoriterator, module: vectorization",scan when time allows,2020-03-11T16:29:39Z
issue,34409,PyTorch build with Ideep support?,https://github.com/pytorch/pytorch/issues/34409,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: mkldnn",scan when time allows,2020-03-11T08:22:47Z
issue,34361,[JIT legacy executor] device propagation regression,https://github.com/pytorch/pytorch/issues/34361,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-03-10T20:48:53Z
issue,34516,Expose chunk_sizes for DataParallel,https://github.com/pytorch/pytorch/issues/34516,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: data parallel, module: nn",scan when time allows,2020-03-10T20:29:39Z
issue,25641,[distributed] all_gather on a List of Tensors directly,https://github.com/pytorch/pytorch/issues/25641,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, oncall: distributed",scan when time allows,2020-03-09T23:07:23Z
issue,33876,Torchscript incompatible with torch.cat for tensor lists,https://github.com/pytorch/pytorch/issues/33876,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-03-09T17:10:18Z
issue,34207,Support vars/__dict__ on torch.return_types tuples e.g. topk,https://github.com/pytorch/pytorch/issues/34207,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-03-07T16:28:11Z
issue,33787,MSE Loss Implementation,https://github.com/pytorch/pytorch/issues/33787,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-03-06T19:43:56Z
issue,33799,Add `start_process_in_context` to `torch.multiprocessing`,https://github.com/pytorch/pytorch/issues/33799,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: multiprocessing,scan when time allows,2020-03-06T02:13:33Z
issue,34063,[docs] Unclear description of indices arg in torch.index_put_,https://github.com/pytorch/pytorch/issues/34063,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2020-03-06T00:19:29Z
issue,34052,Model loaded in C++ runtime is not thread safe,https://github.com/pytorch/pytorch/issues/34052,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: sanitizers",scan when time allows,2020-03-05T22:18:37Z
issue,34082,Autograd deep copy avoidance optimization unsound in the presence of views,https://github.com/pytorch/pytorch/issues/34082,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2020-03-04T22:56:38Z
issue,33898,Docs for uniform_ don't make any sense,https://github.com/pytorch/pytorch/issues/33898,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distributions,scan when time allows,2020-03-04T21:33:46Z
issue,34172,END_HANDLE_TH_ERRORS_PYBIND prevents pybind11 Exception translation,https://github.com/pytorch/pytorch/issues/34172,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: pybind,scan when time allows,2020-03-04T04:17:49Z
issue,20963,official libtorch static build zip file error,https://github.com/pytorch/pytorch/issues/20963,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: build",scan when time allows,2020-03-03T23:24:49Z
issue,18906,Provide option to use alias method in Categorical.sample(),https://github.com/pytorch/pytorch/issues/18906,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distributions,scan when time allows,2020-03-03T21:38:41Z
issue,34086,Enable profiler tracing tests on windows,https://github.com/pytorch/pytorch/issues/34086,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-03-03T16:11:43Z
issue,34068,[C++ API Parity] Incorrect documentation for optim initialization in serialization docs,https://github.com/pytorch/pytorch/issues/34068,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: docs, module: optimizer",scan when time allows,2020-03-02T20:44:19Z
issue,34080,RPC API Changes for TensorPipes,https://github.com/pytorch/pytorch/issues/34080,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rpc,scan when time allows,2020-03-02T20:21:39Z
issue,34001,_broadcast_coalesced_reshape doesn't respect zero-size tensor,https://github.com/pytorch/pytorch/issues/34001,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: data parallel,scan when time allows,2020-03-02T19:39:21Z
issue,33991,TensorPipes RPC Agent Default Args/Result Device Mapping,https://github.com/pytorch/pytorch/issues/33991,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rpc,scan when time allows,2020-03-02T19:36:59Z
issue,33990,TensorPipes RPC Agent Multiple Placement Retries,https://github.com/pytorch/pytorch/issues/33990,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rpc,scan when time allows,2020-03-02T19:04:41Z
issue,33989,TensorPipes RPC Agent Message Acknowledgements,https://github.com/pytorch/pytorch/issues/33989,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rpc,scan when time allows,2020-03-02T19:04:25Z
issue,33983,TensorPipes RPC Agent Listener shortcut,https://github.com/pytorch/pytorch/issues/33983,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rpc,scan when time allows,2020-03-02T19:04:05Z
issue,33982,Initialize TensorPipe RPC Agent Transport,https://github.com/pytorch/pytorch/issues/33982,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rpc,scan when time allows,2020-03-02T19:02:20Z
issue,33979,TensorPipes RPC Agent CUDA Support,https://github.com/pytorch/pytorch/issues/33979,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: rpc",scan when time allows,2020-03-02T19:00:44Z
issue,17658,C++ API: Crash in cudnnDestroy() when deconstructing,https://github.com/pytorch/pytorch/issues/17658,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: abi, module: cpp, module: cudnn",scan when time allows,2020-03-02T17:15:45Z
issue,17703,Training hangs when using DistributedDataParallel in two pod on two nodes,https://github.com/pytorch/pytorch/issues/17703,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2020-03-01T07:16:52Z
issue,2545,Counter-intuitive Patience & Cooldown of ReduceLROnPlateau,https://github.com/pytorch/pytorch/issues/2545,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2020-02-28T23:15:08Z
issue,24915,Shared Dataset Functionality,https://github.com/pytorch/pytorch/issues/24915,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2020-02-28T21:28:17Z
issue,30635,Cannot use torch.jit.script with nn.DataParallel,https://github.com/pytorch/pytorch/issues/30635,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: data parallel, oncall: jit",scan when time allows,2020-02-28T20:25:04Z
issue,33887,[jit] Python arg parser / TorchScript incompatibilities,https://github.com/pytorch/pytorch/issues/33887,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-02-28T18:57:00Z
issue,33844,[jit] Returning different types with `Any` segfaults,https://github.com/pytorch/pytorch/issues/33844,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-02-28T18:57:00Z
issue,33782,[JIT] BroadcastingList annotations don't work with ignore'd functions,https://github.com/pytorch/pytorch/issues/33782,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-02-28T18:56:59Z
issue,33501,"When a Node fails to resolve to an Operator, print out the types of arguments, and all ""close matches"" in known operators",https://github.com/pytorch/pytorch/issues/33501,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-02-28T18:56:59Z
issue,32544,[jit] Use `typing.get_type_hints` instead of parsing types manually,https://github.com/pytorch/pytorch/issues/32544,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-02-28T18:56:58Z
issue,30788,Master Task: JIT/C++ Parity With Pytorch Python API,https://github.com/pytorch/pytorch/issues/30788,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-02-28T18:56:58Z
issue,33353,[jit] `Node` callstack is incorrect,https://github.com/pytorch/pytorch/issues/33353,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-02-28T18:54:57Z
issue,32751,PyTorch freezes on second call to scripted densenet model from torchvision,https://github.com/pytorch/pytorch/issues/32751,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: vision, oncall: jit",scan when time allows,2020-02-28T18:54:55Z
issue,32690,How to customize build torchscript model to be used in end devices codebase,https://github.com/pytorch/pytorch/issues/32690,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: oncall: jit, oncall: mobile",scan when time allows,2020-02-28T18:54:55Z
issue,32565,Reduce RPC branches for Python/Built-inOp/TorchScript,https://github.com/pytorch/pytorch/issues/32565,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: rpc, oncall: jit",scan when time allows,2020-02-28T18:54:54Z
issue,32463,[JIT] Make `torch.jit.script` work on all objects which we can represent as IValues,https://github.com/pytorch/pytorch/issues/32463,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-02-28T18:54:54Z
issue,32337,Save LocalResponceNorm as a single aten node when convert a pytorch model to jit model,https://github.com/pytorch/pytorch/issues/32337,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-02-28T18:54:54Z
issue,32253,[TorchScript] Device comparison implemented in eq but not ne,https://github.com/pytorch/pytorch/issues/32253,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-02-28T18:54:53Z
issue,30138,Unable to register custom JIT Operator with AliasAnalysisKind::CONSERVATIVE,https://github.com/pytorch/pytorch/issues/30138,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-02-28T18:54:52Z
issue,28308,jit script fails with `AttributeError: 'str' object has no attribute 'lineno'`,https://github.com/pytorch/pytorch/issues/28308,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-02-28T18:54:51Z
issue,27479,[JIT] Figure out how to easily investigate memory usage issues issues,https://github.com/pytorch/pytorch/issues/27479,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-02-28T18:54:51Z
issue,31953,JIT tests are linked directly into libtorch and register operators even when unused,https://github.com/pytorch/pytorch/issues/31953,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-02-28T18:52:51Z
issue,31882,Example cmakelists for custom cuda operator?,https://github.com/pytorch/pytorch/issues/31882,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-02-28T18:52:51Z
issue,31574,attribute and register_buffer are not the same on gpu,https://github.com/pytorch/pytorch/issues/31574,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-02-28T18:52:50Z
issue,31551,TorchBind broken on rocm,https://github.com/pytorch/pytorch/issues/31551,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-02-28T18:52:50Z
issue,31430,[jit] `del` with slices doesn't work,https://github.com/pytorch/pytorch/issues/31430,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-02-28T18:52:49Z
issue,31386,[jit] Python type hints in TorchScript classes don't work,https://github.com/pytorch/pytorch/issues/31386,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-02-28T18:52:49Z
issue,31202,Request for Lint Pass to Detect Modification on Parameters/Attributes during TorchScript Inference,https://github.com/pytorch/pytorch/issues/31202,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-02-28T18:52:49Z
issue,30226,Wrong substitution of aten::to,https://github.com/pytorch/pytorch/issues/30226,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: oncall: jit, oncall: mobile",scan when time allows,2020-02-28T18:52:48Z
issue,30049,CPU Memory Leak for JIT ScriptModule on DataParallel,https://github.com/pytorch/pytorch/issues/30049,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-02-28T18:52:47Z
issue,29987,Move the attributes of a module to the given device,https://github.com/pytorch/pytorch/issues/29987,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-02-28T18:52:47Z
issue,29822,[jit] Traced `cat` on GPU doesn't support negative indexing,https://github.com/pytorch/pytorch/issues/29822,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-02-28T18:52:46Z
issue,29750,[jit] Printing the graph doesn't include function calls,https://github.com/pytorch/pytorch/issues/29750,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-02-28T18:52:46Z
issue,29642,error: ‘struct torch::jit::RegisterOperators’ has no member named ‘op’,https://github.com/pytorch/pytorch/issues/29642,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-02-28T18:52:45Z
issue,29597,[jit] Script class attributes aren't automatically added,https://github.com/pytorch/pytorch/issues/29597,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-02-28T18:52:18Z
issue,29063,Trace of torch.tensor is impressively convoluted,https://github.com/pytorch/pytorch/issues/29063,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-02-28T18:52:17Z
issue,28549,Expose DifferentiableGraphBackward to python,https://github.com/pytorch/pytorch/issues/28549,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, oncall: jit",scan when time allows,2020-02-28T18:52:17Z
issue,28519,[jit] C++ Documentation,https://github.com/pytorch/pytorch/issues/28519,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-02-28T18:52:16Z
issue,28517,[JIT] __repr__ support for ScriptModules,https://github.com/pytorch/pytorch/issues/28517,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-02-28T18:52:16Z
issue,28307,[jit] scripted module and user defined methods,https://github.com/pytorch/pytorch/issues/28307,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-02-28T18:52:16Z
issue,27505,[JIT] ndimension not supported,https://github.com/pytorch/pytorch/issues/27505,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-02-28T18:52:14Z
issue,27497,TSAN failure related to mkldnn,https://github.com/pytorch/pytorch/issues/27497,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: mkldnn, oncall: jit",scan when time allows,2020-02-28T18:52:13Z
issue,27475,[JIT] script::Module API parity with nnmodule,https://github.com/pytorch/pytorch/issues/27475,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-02-28T18:52:13Z
issue,27406,[jit] String frontend doesn't support default arg values,https://github.com/pytorch/pytorch/issues/27406,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-02-28T18:52:13Z
issue,27003,[JIT] Script doesn't preserve builtin torch named tuples upon python return,https://github.com/pytorch/pytorch/issues/27003,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-02-28T18:52:12Z
issue,26086,[Feature Request] Trace / Script C++ models,https://github.com/pytorch/pytorch/issues/26086,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, oncall: jit",scan when time allows,2020-02-28T18:51:43Z
issue,25803,[jit] `random` module support,https://github.com/pytorch/pytorch/issues/25803,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-02-28T18:51:43Z
issue,25047,ScriptModule and nn.Module parameter ordering difference,https://github.com/pytorch/pytorch/issues/25047,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-02-28T18:51:41Z
issue,24248,Loading custom Torchscript C++ operators in python segfaults due to ABI compatibility issue between pytorch and libtorch,https://github.com/pytorch/pytorch/issues/24248,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-02-28T18:51:41Z
issue,24229,Confusing error message for Custom Class type mismatch,https://github.com/pytorch/pytorch/issues/24229,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-02-28T18:51:40Z
issue,24225,Allow forward method to be defined with .define() in new TorchScript API,https://github.com/pytorch/pytorch/issues/24225,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-02-28T18:51:40Z
issue,23958,[jit] Python @property's not supported in TorchScript,https://github.com/pytorch/pytorch/issues/23958,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-02-28T18:51:40Z
issue,23890,[JIT] Can't use ndim in script,https://github.com/pytorch/pytorch/issues/23890,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-02-28T18:51:39Z
issue,24176,torch.fft crash when used with nn.DataParallel,https://github.com/pytorch/pytorch/issues/24176,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: data parallel",scan when time allows,2020-02-28T16:24:11Z
issue,33367,Expose `internal::GRAIN_SIZE` through Python API.,https://github.com/pytorch/pytorch/issues/33367,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: multithreading,scan when time allows,2020-02-27T02:30:13Z
issue,20343,torch.distributions.Binomial.sample() uses a massive amount of memory,https://github.com/pytorch/pytorch/issues/20343,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: distributions, module: memory usage",scan when time allows,2020-02-26T23:16:24Z
issue,29548,Dispatch key reorganization,https://github.com/pytorch/pytorch/issues/29548,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: internals,scan when time allows,2020-02-26T21:28:27Z
issue,2512,detach_() variant that affects all past uses too,https://github.com/pytorch/pytorch/issues/2512,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2020-02-26T18:18:19Z
issue,26447,parallel_for may hang when called in main process and then on daemon process,https://github.com/pytorch/pytorch/issues/26447,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: deadlock, module: multiprocessing",scan when time allows,2020-02-25T21:09:07Z
issue,26691,Support FPGA Xilinx,https://github.com/pytorch/pytorch/issues/26691,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: backend,scan when time allows,2020-02-25T20:32:16Z
issue,33482,Distributed Data Parallel for computation graphs that make RPCs in forward(),https://github.com/pytorch/pytorch/issues/33482,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: rpc, oncall: distributed",scan when time allows,2020-02-25T18:19:11Z
issue,33738,Support pipelining the backward pass and optimizer.step() for distributed autograd.,https://github.com/pytorch/pytorch/issues/33738,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: rpc, oncall: distributed",scan when time allows,2020-02-25T04:48:16Z
issue,33725,Tensor.random_ is not implemented for bfloat16 on CPU(but implemented on CUDA),https://github.com/pytorch/pytorch/issues/33725,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: random",scan when time allows,2020-02-25T00:55:48Z
issue,32510,Tensor.random_ is not implemented for bool on CUDA(but implemented on CPU),https://github.com/pytorch/pytorch/issues/32510,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: random",scan when time allows,2020-02-24T22:13:19Z
issue,24130,[RPC] Add type annotations for RPC-related Python files,https://github.com/pytorch/pytorch/issues/24130,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rpc,scan when time allows,2020-02-24T19:02:18Z
issue,28055,Traced resnet101 leaks memory during `forward`,https://github.com/pytorch/pytorch/issues/28055,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: memory usage,scan when time allows,2020-02-24T18:24:20Z
issue,33691,Some module has incorrect scope in complex naming situations in tensorboard graph export,https://github.com/pytorch/pytorch/issues/33691,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tensorboard,scan when time allows,2020-02-24T15:52:06Z
issue,33684,Decouple Lifetime of Local RRef and RPC,https://github.com/pytorch/pytorch/issues/33684,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2020-02-24T04:07:58Z
issue,33670,Some module info is missing in nested graph for tensorboard,https://github.com/pytorch/pytorch/issues/33670,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tensorboard,scan when time allows,2020-02-23T19:35:03Z
issue,33664,torch.rand() not having same values on using torch.manual_seed(0),https://github.com/pytorch/pytorch/issues/33664,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: random,scan when time allows,2020-02-23T19:29:12Z
issue,33653,master build error,https://github.com/pytorch/pytorch/issues/33653,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-02-23T19:21:39Z
issue,33436,It seems nn.Sequential.add_module() could take duplicate names,https://github.com/pytorch/pytorch/issues/33436,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2020-02-23T10:58:32Z
issue,33360,Dropout of attention weights in function F.multi_head_attention_forward() breaks sum-to-1 constraint,https://github.com/pytorch/pytorch/issues/33360,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2020-02-21T21:17:43Z
issue,33611,Resnet 32x32 in caffe2 generates ResNet with wrong number of layers,https://github.com/pytorch/pytorch/issues/33611,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-02-21T15:35:48Z
issue,32576,Document different gradient wrt to jax when nesting,https://github.com/pytorch/pytorch/issues/32576,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: docs",scan when time allows,2020-02-20T20:49:29Z
issue,6613,[caffe2] benchmark performance for different operators,https://github.com/pytorch/pytorch/issues/6613,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-02-20T14:42:49Z
issue,33476,Make ArrayRef::size() return int64_t rather than size_t,https://github.com/pytorch/pytorch/issues/33476,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: internals,scan when time allows,2020-02-20T03:23:36Z
issue,33477,"Add 32-bit CI (e.g., Raspberry PI CI)",https://github.com/pytorch/pytorch/issues/33477,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2020-02-20T03:22:53Z
issue,33514,"Make setter non-optional, e.g., TensorOptions::device(optional<Device>) -> device(Device), and add a device_opt setter",https://github.com/pytorch/pytorch/issues/33514,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp,scan when time allows,2020-02-20T03:18:39Z
issue,4825,[feature request]Support AVX512F intrinstics to vectorize operations,https://github.com/pytorch/pytorch/issues/4825,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpu,scan when time allows,2020-02-20T00:07:25Z
issue,2407,Hard-negative mining using __getitem__ directive in Dataset class,https://github.com/pytorch/pytorch/issues/2407,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2020-02-19T21:24:04Z
issue,2466,Autograd test failure on ppc64le,https://github.com/pytorch/pytorch/issues/2466,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: power,scan when time allows,2020-02-19T21:22:39Z
issue,28093,Move options logic in _like functions from parsing layer to implementations,https://github.com/pytorch/pytorch/issues/28093,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-02-19T18:40:14Z
issue,33396,PyTorch 1.4.0 does not support using `Module` or `ModuleList` in attribute annotations in ScriptModule,https://github.com/pytorch/pytorch/issues/33396,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-02-18T18:30:43Z
issue,33413,IterableDataset with num_workers > 0 and drop_last=True drops more instances than expected,https://github.com/pytorch/pytorch/issues/33413,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2020-02-18T18:13:31Z
issue,33363,Make it easier to add new messages in RPC layer,https://github.com/pytorch/pytorch/issues/33363,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rpc,scan when time allows,2020-02-18T16:55:55Z
issue,33390,It is not good to separate the steps of modules making and forward computation,https://github.com/pytorch/pytorch/issues/33390,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-02-18T15:58:12Z
issue,21016,[utils.bottleneck] throws initialization error for cuda profiling,https://github.com/pytorch/pytorch/issues/21016,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2020-02-18T00:13:21Z
issue,32924,Connect timeout feature do not work in DDP with TCPStore,https://github.com/pytorch/pytorch/issues/32924,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2020-02-17T18:35:49Z
issue,33114,Move the custom pass execution back to the beginning of runNondiffOptimization,https://github.com/pytorch/pytorch/issues/33114,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-02-16T19:26:23Z
issue,32490,"RuntimeError: has_marked_unused_parameters_ INTERNAL ASSERT FAILED at /pytorch/torch/csrc/distributed/c10d/reducer.cpp:290, please report a bug to PyTorch.",https://github.com/pytorch/pytorch/issues/32490,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2020-02-16T00:09:59Z
issue,32390,[JIT] Compile group of functions/modules,https://github.com/pytorch/pytorch/issues/32390,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-02-15T16:05:57Z
issue,33342,malloc(): memory corruption (fast),https://github.com/pytorch/pytorch/issues/33342,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: data parallel,scan when time allows,2020-02-14T18:11:46Z
issue,33341,how-to-adjust-learning-rate using libtorch,https://github.com/pytorch/pytorch/issues/33341,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-02-14T17:57:33Z
issue,31038,Pytorch 1.3.0 on RTX cards: CUDA error: an illegal memory access was encountered,https://github.com/pytorch/pytorch/issues/31038,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: memory usage",scan when time allows,2020-02-13T21:06:50Z
issue,33303,torch::var_out and dimnames,https://github.com/pytorch/pytorch/issues/33303,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: named tensor",scan when time allows,2020-02-13T19:50:28Z
issue,33301,Don't take TensorOptions by reference,https://github.com/pytorch/pytorch/issues/33301,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp,scan when time allows,2020-02-13T18:59:39Z
issue,33249,Core dumps being created when running test_c10d.py and test_multiprocessing_spawn.py,https://github.com/pytorch/pytorch/issues/33249,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tests,scan when time allows,2020-02-13T16:43:26Z
issue,33241,"[feature request] Add ""groups"" argument to nn.Fold and nn.Unfold",https://github.com/pytorch/pytorch/issues/33241,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2020-02-13T08:28:41Z
issue,33180,Loading pretrained model,https://github.com/pytorch/pytorch/issues/33180,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-02-12T02:17:00Z
issue,31708,Computing dot product of columns sliced from large matrix causes illegal memory access in CUDA,https://github.com/pytorch/pytorch/issues/31708,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: 64-bit, module: cublas, module: cuda, module: dependency bug",scan when time allows,2020-02-12T01:17:20Z
issue,2312,DataParallel is not compatible with pack_padded_sequence,https://github.com/pytorch/pytorch/issues/2312,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: data parallel,scan when time allows,2020-02-11T18:58:23Z
issue,30291,[feature request] A way to restore/assign tensor _version,https://github.com/pytorch/pytorch/issues/30291,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2020-02-11T17:11:03Z
issue,33051,Recover from CUDA runtime error,https://github.com/pytorch/pytorch/issues/33051,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: third_party",scan when time allows,2020-02-11T02:14:42Z
issue,5900,"Build Fails on Gentoo with CUDA 9.1, GCC 6.4, Python 3.5",https://github.com/pytorch/pytorch/issues/5900,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2020-02-11T01:53:14Z
issue,26109,Support the AVX512 runtime dispatch,https://github.com/pytorch/pytorch/issues/26109,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-02-10T23:57:56Z
issue,31866,Negative indices in chunk could cause Out of Range access on loss.backward in JIT,https://github.com/pytorch/pytorch/issues/31866,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-02-10T18:39:00Z
issue,33115,[RFC] Add ability to get all remote parameters when constructing DistributedOptimizer.,https://github.com/pytorch/pytorch/issues/33115,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rpc,scan when time allows,2020-02-10T12:01:25Z
issue,33081,DataParallel gives different gradients when using LSTMs,https://github.com/pytorch/pytorch/issues/33081,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: data parallel, module: determinism",scan when time allows,2020-02-10T11:43:43Z
issue,33049,Warning when link libtorch and opencv4.2.0 together,https://github.com/pytorch/pytorch/issues/33049,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2020-02-10T11:32:08Z
issue,32406,KL divergence for diagonal Gaussian distributions,https://github.com/pytorch/pytorch/issues/32406,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distributions,scan when time allows,2020-02-10T10:14:15Z
issue,27610,[FR][RFC] Build a serving framework to host and serve trained PyTorch models,https://github.com/pytorch/pytorch/issues/27610,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-02-08T12:34:18Z
issue,22778,In-place updating the original value tensor should also update version counter of sparse tensor's values_ tensor,https://github.com/pytorch/pytorch/issues/22778,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: sparse",scan when time allows,2020-02-07T23:03:09Z
issue,32884,[RFC] Nested scopes in autograd profiler should support RPC calls properly.,https://github.com/pytorch/pytorch/issues/32884,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: rpc",scan when time allows,2020-02-07T20:55:14Z
issue,25941,TestAutograd.test_deep_reentrant fails with SIGBUS on macOS,https://github.com/pytorch/pytorch/issues/25941,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: macos, module: tests",scan when time allows,2020-02-07T20:17:54Z
issue,31185,Retain Subgraph or Save Intermediate Grad support?,https://github.com/pytorch/pytorch/issues/31185,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2020-02-07T19:27:15Z
issue,33007,Upper/Lower attributes for named dimensions for proper Ricci notation and to generalize matrix operations,https://github.com/pytorch/pytorch/issues/33007,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: named tensor,scan when time allows,2020-02-07T19:06:29Z
issue,33006,matmul: no warning when contracting differently named dimensions,https://github.com/pytorch/pytorch/issues/33006,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: named tensor,scan when time allows,2020-02-07T19:06:03Z
issue,29804,Parallel data loader performance degradation for IterableDataset with num_workers > 1 (but not for Dataset).,https://github.com/pytorch/pytorch/issues/29804,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2020-02-07T09:46:41Z
issue,33034,Scripting fails to preserve attribute aliasing,https://github.com/pytorch/pytorch/issues/33034,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-02-06T23:22:53Z
issue,3790,"Add SGDR, SGDW, AdamW and AdamWR",https://github.com/pytorch/pytorch/issues/3790,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2020-02-06T21:52:44Z
issue,30380,Add Google Colab as an option on the 'Get Started' page.,https://github.com/pytorch/pytorch/issues/30380,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2020-02-05T11:44:26Z
issue,32972,"which pytorch do i install for running this project in my windows.link for this is :: https://github.com/xiaojunxu/SQLNet .IN this they have used python 2.7 ,but i am unable to install pytorch on python 2.7 environment .help me with this",https://github.com/pytorch/pytorch/issues/32972,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: binaries,scan when time allows,2020-02-05T07:55:31Z
issue,32754,JIT performance discrepancies,https://github.com/pytorch/pytorch/issues/32754,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: performance, oncall: jit",scan when time allows,2020-02-04T19:37:45Z
issue,32840,Can we add support for Enum in scripted models?,https://github.com/pytorch/pytorch/issues/32840,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-02-03T22:03:55Z
issue,32698,Add Julia Bindings to Torch backend,https://github.com/pytorch/pytorch/issues/32698,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: language binding,scan when time allows,2020-02-03T18:25:19Z
issue,32687,Building PyTorch ignores my current Python version,https://github.com/pytorch/pytorch/issues/32687,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2020-02-03T18:17:32Z
issue,31528,cuCtxGetDevice error and seg fault with DDP and OpenMPI,https://github.com/pytorch/pytorch/issues/31528,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, oncall: distributed",scan when time allows,2020-01-31T01:50:55Z
issue,24835,NetworkX's Version,https://github.com/pytorch/pytorch/issues/24835,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-01-30T15:23:37Z
issue,32784,Implement Backend-Agnostic RPC functionality in RpcAgent,https://github.com/pytorch/pytorch/issues/32784,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rpc,scan when time allows,2020-01-29T20:23:30Z
issue,32783,Comments Separating Class Methods from Different Classes in C++ Files,https://github.com/pytorch/pytorch/issues/32783,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-01-29T19:54:10Z
issue,32767,Documentation is not loaded by IDEs,https://github.com/pytorch/pytorch/issues/32767,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2020-01-29T19:38:04Z
issue,32752,RPC mock mode for unit tests.,https://github.com/pytorch/pytorch/issues/32752,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rpc,scan when time allows,2020-01-29T18:56:58Z
issue,31752,Need a launch utility for Distributed RPC framework.,https://github.com/pytorch/pytorch/issues/31752,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rpc,scan when time allows,2020-01-29T06:53:00Z
issue,30405,Codegen refactoring master task,https://github.com/pytorch/pytorch/issues/30405,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-01-28T21:12:09Z
issue,27517,Allow explicit gradients in torch.distributed.autograd.backward() API,https://github.com/pytorch/pytorch/issues/27517,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rpc,scan when time allows,2020-01-28T02:05:34Z
issue,32452,Numba Enhancement Proposal (NBEP) 7: External Memory Management Plugins,https://github.com/pytorch/pytorch/issues/32452,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-01-27T22:46:21Z
issue,32098,torch.gt ge lt le,https://github.com/pytorch/pytorch/issues/32098,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: type promotion,scan when time allows,2020-01-27T22:46:09Z
issue,22577,Double backward 3 times slower for conv2d with padding = 1,https://github.com/pytorch/pytorch/issues/22577,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: dependency bug, module: double backwards, module: nn",scan when time allows,2020-01-27T17:48:59Z
issue,18643,Memory not being deallocated in backward(),https://github.com/pytorch/pytorch/issues/18643,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: memory usage",scan when time allows,2020-01-27T17:47:42Z
issue,32580,SGD optimizer with deprecation warning,https://github.com/pytorch/pytorch/issues/32580,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: deprecation, module: optimizer",scan when time allows,2020-01-24T17:17:33Z
issue,32504,LibTorch operates very slowly on data blobs from GPU,https://github.com/pytorch/pytorch/issues/32504,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: performance",scan when time allows,2020-01-23T18:34:55Z
issue,32488,TracerWarning When Using Tensor Size in Torchscript Trace,https://github.com/pytorch/pytorch/issues/32488,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-01-22T22:20:29Z
issue,32511,TorchScript C++ API Tracking Issue,https://github.com/pytorch/pytorch/issues/32511,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-01-22T21:41:40Z
issue,32419,MAGMA libraries,https://github.com/pytorch/pytorch/issues/32419,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: build",scan when time allows,2020-01-22T13:42:16Z
issue,32227,"0 INTERNAL ASSERT FAILED at /pytorch/c10/util/intrusive_ptr.h:348, please report a bug to PyTorch.",https://github.com/pytorch/pytorch/issues/32227,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: assert failure,scan when time allows,2020-01-22T10:51:57Z
issue,32403,How to accelerate the compiling of pytorch,https://github.com/pytorch/pytorch/issues/32403,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2020-01-21T23:25:36Z
issue,32416,Is there a way to use SYSTEM_INSTALLED 3rdparty libraries?,https://github.com/pytorch/pytorch/issues/32416,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2020-01-21T23:07:03Z
issue,32427,[C++] Don't use DeprecatedTypeProperties in torch::utils::reorder_tensors_like,https://github.com/pytorch/pytorch/issues/32427,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: internals,scan when time allows,2020-01-21T19:19:47Z
issue,32415,DistributedStreamSampler: support stream sampler in distributed setting,https://github.com/pytorch/pytorch/issues/32415,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, oncall: distributed",scan when time allows,2020-01-21T19:16:32Z
issue,32402,torchvision,https://github.com/pytorch/pytorch/issues/32402,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: vision,scan when time allows,2020-01-21T19:14:54Z
issue,32372,Migrate processGroup::async_work to use Future,https://github.com/pytorch/pytorch/issues/32372,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2020-01-21T19:12:25Z
issue,19125,Multi-gpu via torch::nn::parallel::data_parallel,https://github.com/pytorch/pytorch/issues/19125,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, oncall: distributed",scan when time allows,2020-01-21T16:10:03Z
issue,32036,Internal assert failed using multiple GPUs with DataParallel,https://github.com/pytorch/pytorch/issues/32036,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: assert failure, module: cuda, module: data parallel",scan when time allows,2020-01-20T18:43:38Z
issue,31937,RPC and dist_autograd should respect no_grad mode,https://github.com/pytorch/pytorch/issues/31937,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rpc,scan when time allows,2020-01-20T02:23:44Z
issue,32357,Pytorch JIT Compilation Does Not Finish (Infinite Loop?) for Deeper Models,https://github.com/pytorch/pytorch/issues/32357,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-01-18T18:31:27Z
issue,32375,Make _rpc_sync_torchscript and _rpc_async_torchscript work with autograd profiler,https://github.com/pytorch/pytorch/issues/32375,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-01-18T07:36:19Z
issue,32363,Possible ProcessGroup::Work::abort correctness issue,https://github.com/pytorch/pytorch/issues/32363,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: rpc, oncall: distributed",scan when time allows,2020-01-17T19:34:07Z
issue,32269,Support strategy to train large model that exceeds GPU mem and DRAM mem,https://github.com/pytorch/pytorch/issues/32269,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rpc,scan when time allows,2020-01-17T03:13:32Z
issue,32264,torch.nn.functional.threshold not work with LongTensor,https://github.com/pytorch/pytorch/issues/32264,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2020-01-16T23:07:08Z
issue,11612,[feature request] Triangular Matrix Representation,https://github.com/pytorch/pytorch/issues/11612,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-01-16T17:01:41Z
issue,32237,PyTorch C++ API docs only tracks master branch,https://github.com/pytorch/pytorch/issues/32237,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: doc infra",scan when time allows,2020-01-15T20:13:35Z
issue,32132,Slighty out of tolerance for `test_mv` and `test_cholesky_solve_batched_cuda_float64`,https://github.com/pytorch/pytorch/issues/32132,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tests,scan when time allows,2020-01-15T18:43:11Z
issue,32088,Batch convolutional layer with 5d weight tensor that is not contiguous,https://github.com/pytorch/pytorch/issues/32088,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-01-15T17:25:36Z
issue,32087,Out-of-date link to pytorch ci dockerfiles in jenkins readme,https://github.com/pytorch/pytorch/issues/32087,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2020-01-15T17:13:34Z
issue,32219,[Feature proposal] improved algorithms for checkpointing,https://github.com/pytorch/pytorch/issues/32219,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-01-15T16:48:59Z
issue,32005,gradients inside gradient checkpoint,https://github.com/pytorch/pytorch/issues/32005,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: activation checkpointing, module: autograd",scan when time allows,2020-01-15T16:41:56Z
issue,32105,TensorRT: CheckDims() need adjustment for EXPLICIT_BATCH?,https://github.com/pytorch/pytorch/issues/32105,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-01-14T20:08:57Z
issue,32141,We should change DeprecationWarnings to UserWarnings in 27361,https://github.com/pytorch/pytorch/issues/32141,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2020-01-14T19:47:04Z
issue,32150,pin_memory may change the type of instance returned by collate_fn.,https://github.com/pytorch/pytorch/issues/32150,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2020-01-14T19:42:51Z
issue,32165,Serialization inconsistency with pickling tensors breaks caching,https://github.com/pytorch/pytorch/issues/32165,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: serialization,scan when time allows,2020-01-14T19:31:54Z
issue,32023,Inaccurate batched GRU results on CPU,https://github.com/pytorch/pytorch/issues/32023,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: nn, module: numerical-reproducibility, module: rnn",scan when time allows,2020-01-14T17:17:28Z
issue,32033,run detectron test test_spatial_narrow_as_op.py fails,https://github.com/pytorch/pytorch/issues/32033,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-01-14T17:16:12Z
issue,30569,"Assertion `index >= -sizes[i] && index < sizes[i] && ""index out of bounds""` failed.",https://github.com/pytorch/pytorch/issues/30569,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: error checking",scan when time allows,2020-01-13T23:02:21Z
issue,30387,[docs] Missing docs for Storage.from_buffer,https://github.com/pytorch/pytorch/issues/30387,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2020-01-11T07:05:58Z
issue,20967,Track running stats regardless of track_running_stats=False,https://github.com/pytorch/pytorch/issues/20967,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2020-01-10T14:43:30Z
issue,31907,Compile libtorch by source code failed.,https://github.com/pytorch/pytorch/issues/31907,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: cpp",scan when time allows,2020-01-09T22:52:25Z
issue,31779,torch.poisson returns floating point tensor,https://github.com/pytorch/pytorch/issues/31779,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distributions,scan when time allows,2020-01-09T18:54:31Z
issue,31979,Reuse spawned subprocesses in RPC tests,https://github.com/pytorch/pytorch/issues/31979,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rpc,scan when time allows,2020-01-09T15:50:30Z
issue,16901,Issue with dataloader using pin_memory = True,https://github.com/pytorch/pytorch/issues/16901,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2020-01-08T19:18:56Z
issue,31867,The model training time is increasing between runs if the same DataLoader reused to train multiple models.,https://github.com/pytorch/pytorch/issues/31867,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2020-01-08T19:04:39Z
issue,31842,Support rpc/remote torch script call with script class/module name and class/module method name,https://github.com/pytorch/pytorch/issues/31842,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: rpc, oncall: jit",scan when time allows,2020-01-08T18:54:03Z
issue,31487,Disable PSIMD? Why Pytorch is trying to download PSimd when PSIMD_SOURCE_DIR is defined?,https://github.com/pytorch/pytorch/issues/31487,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2020-01-06T19:51:23Z
issue,31688,Parallelization: more balanced work distribution among workers,https://github.com/pytorch/pytorch/issues/31688,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: data parallel",scan when time allows,2020-01-06T18:26:15Z
issue,31107,Add torch.version.nccl,https://github.com/pytorch/pytorch/issues/31107,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, module: nccl, oncall: distributed",scan when time allows,2020-01-06T10:36:29Z
issue,31651,"update embedding at indices, other than those passed as input, in the case of sparse tensors",https://github.com/pytorch/pytorch/issues/31651,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: sparse",scan when time allows,2020-01-04T12:30:36Z
issue,30404,affine_grid CUDA / cuDNN support for Half removed in 1.3.x,https://github.com/pytorch/pytorch/issues/30404,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: cudnn",scan when time allows,2020-01-04T00:15:20Z
issue,31422,Support DataParallel with PackedSequence,https://github.com/pytorch/pytorch/issues/31422,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, module: data parallel, oncall: distributed",scan when time allows,2020-01-02T21:54:50Z
issue,31356,Optimizing DLRM for CPU,https://github.com/pytorch/pytorch/issues/31356,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: performance",scan when time allows,2020-01-02T14:56:09Z
issue,31425,cuDNN convolution does not handle empty input tensor,https://github.com/pytorch/pytorch/issues/31425,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: cudnn",scan when time allows,2020-01-02T11:41:51Z
issue,31697,the cmake problem with build from source ?,https://github.com/pytorch/pytorch/issues/31697,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: binaries,scan when time allows,2020-01-02T10:56:33Z
issue,6442,[feature request] dropout1d,https://github.com/pytorch/pytorch/issues/6442,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2020-01-02T10:18:42Z
issue,31259,Script method can't call a scripted function when it is decorated with `@torch.no_grad`,https://github.com/pytorch/pytorch/issues/31259,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-01-02T07:48:56Z
issue,31553,allow setting different batch size splits for data_parallel.py and distributed.py,https://github.com/pytorch/pytorch/issues/31553,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: data parallel, oncall: distributed",scan when time allows,2020-01-01T04:34:03Z
issue,26802,Google Summer of Code,https://github.com/pytorch/pytorch/issues/26802,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-12-31T19:57:08Z
issue,31685,What is the significance of torchvision._is_tracing()?,https://github.com/pytorch/pytorch/issues/31685,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: vision,scan when time allows,2019-12-30T21:50:08Z
issue,31657,Use of Sequence collections for abstract classes in Dataset,https://github.com/pytorch/pytorch/issues/31657,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2019-12-27T18:05:48Z
issue,31321,Ability to download docs HTML for offline use,https://github.com/pytorch/pytorch/issues/31321,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2019-12-27T09:42:49Z
issue,20006,"RuntimeError: invalid argument 10: ldb should be at least max(1, 0), but have 0 at ../aten/src/TH/generic/THBlas.cpp:36",https://github.com/pytorch/pytorch/issues/20006,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: internals,scan when time allows,2019-12-26T15:45:39Z
issue,31606,No auto-suggest capacity for Transformer,https://github.com/pytorch/pytorch/issues/31606,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-12-26T04:51:26Z
issue,31557,[docs] F.ctc_loss docs to warn clearly about invalid inf-causing inputs; zero_infinity to become enabled by default,https://github.com/pytorch/pytorch/issues/31557,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2019-12-25T13:23:23Z
issue,31596,pytorch forward hangs in multiprocess environment,https://github.com/pytorch/pytorch/issues/31596,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: multiprocessing,scan when time allows,2019-12-25T00:21:28Z
issue,29697,MultiStepLR does not return good lr after load_state_dict,https://github.com/pytorch/pytorch/issues/29697,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2019-12-23T17:07:01Z
issue,24963,torch.cuda.synchronize blocks CUDA execution on other threads using other devices.,https://github.com/pytorch/pytorch/issues/24963,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2019-12-23T13:18:23Z
issue,31548,"c++ PReLUFuncOptions declared, not used or valid",https://github.com/pytorch/pytorch/issues/31548,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-12-23T07:52:12Z
issue,1861,Factorized Output Layer,https://github.com/pytorch/pytorch/issues/1861,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-12-23T07:35:56Z
issue,19668,Importing matlab.engine after torch causes bad_alloc,https://github.com/pytorch/pytorch/issues/19668,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: assert failure, module: internals",scan when time allows,2019-12-20T07:59:22Z
issue,30803,Pytorch openmp thread number tuning option for CPU trainning,https://github.com/pytorch/pytorch/issues/30803,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: multithreading, module: performance",scan when time allows,2019-12-19T17:50:10Z
issue,31471,Mnasnet0_5 first layer shape incorrect,https://github.com/pytorch/pytorch/issues/31471,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: vision,scan when time allows,2019-12-19T17:05:18Z
issue,31300,torch runtime error when manual link libmkldnn.so,https://github.com/pytorch/pytorch/issues/31300,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: mkldnn",scan when time allows,2019-12-19T15:48:12Z
issue,31289,Intel OMP multiprocessing assertion failure: Assertion failure at z_Linux_util.cpp(2338),https://github.com/pytorch/pytorch/issues/31289,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dependency bug, module: multiprocessing",scan when time allows,2019-12-19T09:17:29Z
issue,31228,[RPC] Support nn.Module pickling with share memory,https://github.com/pytorch/pytorch/issues/31228,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rpc,scan when time allows,2019-12-19T09:10:53Z
issue,19163,LayerNorm is very slow (almost frozen) in CPU of multiprocessing,https://github.com/pytorch/pytorch/issues/19163,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpu,scan when time allows,2019-12-19T03:09:57Z
issue,31391,"Error ""builtin cannot be used as a value"" when add Python snippets in C++",https://github.com/pytorch/pytorch/issues/31391,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2019-12-18T20:49:44Z
issue,31360,DDP/MP not yielding nontrivial speedup,https://github.com/pytorch/pytorch/issues/31360,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: data parallel, module: multiprocessing, module: performance",scan when time allows,2019-12-17T21:32:39Z
issue,31367,`clip_grad_norm` allows negative `max_norm` values,https://github.com/pytorch/pytorch/issues/31367,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2019-12-17T18:25:28Z
issue,31353,Memory management is inefficient which limits performance,https://github.com/pytorch/pytorch/issues/31353,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: memory usage, module: performance",scan when time allows,2019-12-17T16:18:53Z
issue,22954,Support serializing IValue to bytes (and deserialize from bytes),https://github.com/pytorch/pytorch/issues/22954,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2019-12-17T05:07:12Z
issue,4959,Speed up data loading for `TensorDataset` if the underlying dataset supports index by a list of indices,https://github.com/pytorch/pytorch/issues/4959,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dataloader, module: performance",scan when time allows,2019-12-16T22:37:22Z
issue,31308,[mac] Failure to import torch,https://github.com/pytorch/pytorch/issues/31308,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: macos,scan when time allows,2019-12-16T17:52:34Z
issue,28194,New Stochastic Optimization Algorithms in Pytorch,https://github.com/pytorch/pytorch/issues/28194,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2019-12-16T00:06:52Z
issue,21144,Data Parallel Implementation Improvements,https://github.com/pytorch/pytorch/issues/21144,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2019-12-15T22:04:46Z
issue,27144,torch::jit::script::Module has no zero_grad(),https://github.com/pytorch/pytorch/issues/27144,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2019-12-13T19:01:12Z
issue,31110,torch.nn.Softplus threshold argument bug?,https://github.com/pytorch/pytorch/issues/31110,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2019-12-12T23:40:17Z
issue,17666,Building pytorch on ARM failed,https://github.com/pytorch/pytorch/issues/17666,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2019-12-12T19:59:24Z
issue,31191,[JIT] Slice with optional not supported,https://github.com/pytorch/pytorch/issues/31191,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2019-12-12T18:27:33Z
issue,27409,Installer not setting rpath for MAGMA (OS X w/ GPU),https://github.com/pytorch/pytorch/issues/27409,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cuda, module: macos",scan when time allows,2019-12-12T16:51:46Z
issue,30780,Remove Ops bound in Python Layer for Legacy Reasons,https://github.com/pytorch/pytorch/issues/30780,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: internals, module: pybind",scan when time allows,2019-12-12T15:59:02Z
issue,23655,TorchScript GPU Fuser Doesn't Handle In-Place Operations,https://github.com/pytorch/pytorch/issues/23655,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: performance, oncall: jit",scan when time allows,2019-12-12T15:22:29Z
issue,31129,[jit] Python objects as arguments are not mutated,https://github.com/pytorch/pytorch/issues/31129,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2019-12-11T21:47:51Z
issue,19581,[jit] Do aten::values dispatch at build time instead of runtime,https://github.com/pytorch/pytorch/issues/19581,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2019-12-11T21:03:37Z
issue,31048,MathJax too small in Firefox,https://github.com/pytorch/pytorch/issues/31048,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2019-12-11T17:14:30Z
issue,31085,Common lookup of generic types across full script and mobile parsers,https://github.com/pytorch/pytorch/issues/31085,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-12-11T17:06:26Z
issue,30718,Method to broadcast parameters/buffers of DDP model,https://github.com/pytorch/pytorch/issues/30718,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2019-12-11T09:08:12Z
issue,30962,How can I add masks to parameters,https://github.com/pytorch/pytorch/issues/30962,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2019-12-11T07:37:43Z
issue,25783,Detaching a distribution's `log_prob` to block gradients only w.r.t its parameters,https://github.com/pytorch/pytorch/issues/25783,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distributions,scan when time allows,2019-12-11T03:24:29Z
issue,30965,JIT breaks with postponed annotations,https://github.com/pytorch/pytorch/issues/30965,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2019-12-10T22:10:12Z
issue,31007,"GPU version of minimal example for libtorch fails with ""no kernel image is available...""",https://github.com/pytorch/pytorch/issues/31007,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2019-12-10T08:09:47Z
issue,30744,Issues in linking libtorch c++,https://github.com/pytorch/pytorch/issues/30744,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2019-12-10T06:43:01Z
issue,19248,add stable distribution in torch.distributions,https://github.com/pytorch/pytorch/issues/19248,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distributions,scan when time allows,2019-12-10T00:08:19Z
issue,30953,false CHECK FAILED at ../aten/src/ATen/core/function_schema_inl.h,https://github.com/pytorch/pytorch/issues/30953,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: internals,scan when time allows,2019-12-09T18:26:37Z
issue,30947,Wrong initialization with kaiming_uniform_,https://github.com/pytorch/pytorch/issues/30947,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2019-12-09T18:25:45Z
issue,30903,Don't ship protoc in wheels,https://github.com/pytorch/pytorch/issues/30903,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: binaries,scan when time allows,2019-12-09T18:11:20Z
issue,30899,Adding max_norm constraint to an Embedding layer leads to an error,https://github.com/pytorch/pytorch/issues/30899,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2019-12-09T18:09:43Z
issue,30762,Page for `torch.__config__` 404s,https://github.com/pytorch/pytorch/issues/30762,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2019-12-06T17:15:28Z
issue,28577,`Tensor.__reversed__` breaks protocol for reversible objects,https://github.com/pytorch/pytorch/issues/28577,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: internals, module: pybind",scan when time allows,2019-12-05T04:00:56Z
issue,24037,Tests do not pass with the latest protobuf,https://github.com/pytorch/pytorch/issues/24037,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: protobuf,scan when time allows,2019-12-05T00:00:51Z
issue,23032,Proposal: Optional AutogradMeta for Variable,https://github.com/pytorch/pytorch/issues/23032,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2019-12-04T20:46:20Z
issue,30596,Overlapping strides not supported by cublas,https://github.com/pytorch/pytorch/issues/30596,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cublas,scan when time allows,2019-12-04T15:26:24Z
issue,23355,[jit] add named tuple as output type to the tracer,https://github.com/pytorch/pytorch/issues/23355,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2019-12-04T00:59:17Z
issue,30589,weight norm missing p=,https://github.com/pytorch/pytorch/issues/30589,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2019-12-03T15:40:00Z
issue,30611,Refactor/consolidate code for generating test tensors,https://github.com/pytorch/pytorch/issues/30611,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: tests, module: type promotion",scan when time allows,2019-12-03T15:09:27Z
issue,30633,Remove RPC internal helper that overrides the default pickler,https://github.com/pytorch/pytorch/issues/30633,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rpc,scan when time allows,2019-12-03T08:37:18Z
issue,30489,Torchscript Precision Issue with PyTorch/vision pretrained model inception v3,https://github.com/pytorch/pytorch/issues/30489,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2019-12-03T02:48:16Z
issue,24834,PyTorch 1.2 'module' object has no attribute 'BFloat16StorageBase',https://github.com/pytorch/pytorch/issues/24834,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: undefined reference, module: vision",scan when time allows,2019-12-02T09:52:30Z
issue,18212,[JIT] bitwise NOT does not handle tensor shapes correctly under JIT,https://github.com/pytorch/pytorch/issues/18212,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2019-12-02T09:50:03Z
issue,18005,Support 'bytes' type in torchscript,https://github.com/pytorch/pytorch/issues/18005,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2019-12-02T09:41:50Z
issue,17970,Binary not operator causes crash when Jit module is executed on different device,https://github.com/pytorch/pytorch/issues/17970,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2019-12-02T09:38:59Z
issue,17797,Non-coherent result for C++ with multithreading and GPU,https://github.com/pytorch/pytorch/issues/17797,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, oncall: jit",scan when time allows,2019-12-02T09:33:29Z
issue,30533,Document memory characteristics of in-place ops,https://github.com/pytorch/pytorch/issues/30533,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: sparse, module: type promotion",scan when time allows,2019-11-27T21:50:11Z
issue,30535,[FR] nn.init.* accept None as input tensor,https://github.com/pytorch/pytorch/issues/30535,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2019-11-27T21:37:46Z
issue,1736,Expose optimizer options as attributes when there's a single param group,https://github.com/pytorch/pytorch/issues/1736,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, module: optimizer",scan when time allows,2019-11-27T18:47:03Z
issue,30505,Adding a function to change the process_group in DistributedDataParallel,https://github.com/pytorch/pytorch/issues/30505,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2019-11-27T16:11:46Z
issue,30469,Add single input/output tensor scatter/gather to ProcessGroup base class,https://github.com/pytorch/pytorch/issues/30469,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2019-11-27T15:38:53Z
issue,30463,cuda support,https://github.com/pytorch/pytorch/issues/30463,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2019-11-27T15:38:23Z
issue,30421,[jit] Module references aren't preserved,https://github.com/pytorch/pytorch/issues/30421,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2019-11-26T23:32:06Z
issue,30418,More detailed information about TensorType in error messages,https://github.com/pytorch/pytorch/issues/30418,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-11-26T21:08:56Z
issue,30440,Redundant counter in batchnorm impl,https://github.com/pytorch/pytorch/issues/30440,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2019-11-26T19:50:35Z
issue,29545,Tensorboard GPU Problems,https://github.com/pytorch/pytorch/issues/29545,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: tensorboard",scan when time allows,2019-11-26T13:22:51Z
issue,30461,`torch.multiprocessing.spawn` fails when `join=False`,https://github.com/pytorch/pytorch/issues/30461,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: multiprocessing,scan when time allows,2019-11-26T13:12:19Z
issue,28275,nn.parallel.replicate in v1.1+ is much slower than v1.0,https://github.com/pytorch/pytorch/issues/28275,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: data parallel, module: performance",scan when time allows,2019-11-26T12:39:55Z
issue,15291,linked error of Pytorch 1.0 release,https://github.com/pytorch/pytorch/issues/15291,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: macos",scan when time allows,2019-11-26T12:25:37Z
issue,30413,nn.functional should maintain API parity with nn where possible,https://github.com/pytorch/pytorch/issues/30413,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2019-11-26T09:30:45Z
issue,30399,Conv2d: Inconsistent results on Raspberry Pi 3B,https://github.com/pytorch/pytorch/issues/30399,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: convolution,scan when time allows,2019-11-26T07:39:12Z
issue,30443,Expose bhistc to python,https://github.com/pytorch/pytorch/issues/30443,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-11-26T04:37:24Z
issue,30403,last updated timestamp,https://github.com/pytorch/pytorch/issues/30403,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2019-11-26T00:47:28Z
issue,30384,Docs missing link to torch.__config__,https://github.com/pytorch/pytorch/issues/30384,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2019-11-26T00:38:14Z
issue,10006,RNN gradients in eval mode in pytorch 0.4,https://github.com/pytorch/pytorch/issues/10006,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: rnn",scan when time allows,2019-11-25T20:56:13Z
issue,26119,Make GloballyUniqueId a common type for both rpc and dist autograd,https://github.com/pytorch/pytorch/issues/26119,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rpc,scan when time allows,2019-11-25T19:04:52Z
issue,27671,Avoid RTTI in DistEngine,https://github.com/pytorch/pytorch/issues/27671,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rpc,scan when time allows,2019-11-25T19:01:15Z
issue,22022,nn.modules.functional.h does not support optional arguments,https://github.com/pytorch/pytorch/issues/22022,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: nn",scan when time allows,2019-11-25T16:59:20Z
issue,24273,Check PyTorch version when initializing process groups,https://github.com/pytorch/pytorch/issues/24273,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, oncall: distributed",scan when time allows,2019-11-25T09:10:38Z
issue,27843,Be able to build torch.distributed documentation easier,https://github.com/pytorch/pytorch/issues/27843,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: doc infra, oncall: distributed",scan when time allows,2019-11-25T08:56:57Z
issue,30225,c10:Error: could not unlink the shared memory file,https://github.com/pytorch/pytorch/issues/30225,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dataloader, module: multiprocessing",scan when time allows,2019-11-22T20:39:12Z
issue,30150,I have reconstructed LSTM model and tested by mnist data but the loss is not changed (loss=2.3),https://github.com/pytorch/pytorch/issues/30150,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rnn,scan when time allows,2019-11-22T09:19:53Z
issue,30182,[quantization][graph mode] SubgraphRewriter discards SourceRanges,https://github.com/pytorch/pytorch/issues/30182,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2019-11-22T07:36:25Z
issue,1591,Pad PackedSequences to original batch length,https://github.com/pytorch/pytorch/issues/1591,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nestedtensor,scan when time allows,2019-11-21T21:49:43Z
issue,30155,Transform caffe2 to trt failed,https://github.com/pytorch/pytorch/issues/30155,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-11-21T21:30:02Z
issue,30248,Add scripts for comprehensive benchmark TensorIterator,https://github.com/pytorch/pytorch/issues/30248,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: performance,scan when time allows,2019-11-21T21:07:03Z
issue,30162,[docs] Newline / whitespace / comma missing in formula for PoissonNLLLoss,https://github.com/pytorch/pytorch/issues/30162,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2019-11-21T20:43:13Z
issue,29722,Slow (20-50x) RNN tutorial/example when torch is installed using pip comp. to conda installation,https://github.com/pytorch/pytorch/issues/29722,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: mkl, module: performance, module: rnn",scan when time allows,2019-11-21T12:41:53Z
issue,9208,[Caffe2] compiling error with gcc-6,https://github.com/pytorch/pytorch/issues/9208,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-11-21T07:28:13Z
issue,22911,JIT trace parameter sharing error if Module attributes happen to be the same,https://github.com/pytorch/pytorch/issues/22911,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2019-11-20T22:29:58Z
issue,30161,Support struct that is initializable / mutable in CPP,https://github.com/pytorch/pytorch/issues/30161,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2019-11-20T21:42:00Z
issue,27955,Expand Pytorch C10D backend to dynamic load third party communication library,https://github.com/pytorch/pytorch/issues/27955,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2019-11-20T10:44:50Z
issue,26759,[RFC] RRef Protocol,https://github.com/pytorch/pytorch/issues/26759,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rpc,scan when time allows,2019-11-19T16:22:34Z
issue,26790,Importing tensorboard jams CUDA device selection,https://github.com/pytorch/pytorch/issues/26790,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tensorboard,scan when time allows,2019-11-18T15:43:47Z
issue,29860,torch.save/load shows raw path on the pickle_module arg,https://github.com/pytorch/pytorch/issues/29860,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: serialization",scan when time allows,2019-11-15T00:16:23Z
issue,29800,unsqueeze has 'out=' option documented but not implemented(?),https://github.com/pytorch/pytorch/issues/29800,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2019-11-14T22:17:37Z
issue,29720,Provide a mechanism to set global state per test in thread-safe manner,https://github.com/pytorch/pytorch/issues/29720,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tests,scan when time allows,2019-11-14T19:12:42Z
issue,29717,CPU and CUDA error messages are divergent in type promotion,https://github.com/pytorch/pytorch/issues/29717,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: tests, module: type promotion",scan when time allows,2019-11-14T18:33:31Z
issue,29739,"[build] gcc 7.4 needs CMAKE_CXX_FLAGS=""-std=gnu++11""",https://github.com/pytorch/pytorch/issues/29739,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: dependency bug",scan when time allows,2019-11-14T18:22:00Z
issue,29713,[FR] F.pad support syntax sugars for specifying the padding amount,https://github.com/pytorch/pytorch/issues/29713,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: convolution,scan when time allows,2019-11-14T18:14:02Z
issue,29754,[FR] general nll_loss and cross_entropy along arbitrary dimension,https://github.com/pytorch/pytorch/issues/29754,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: loss,scan when time allows,2019-11-14T18:00:09Z
issue,11793,DataParallel: Parallel_apply assert len(modules) == len(inputs) AssertionError,https://github.com/pytorch/pytorch/issues/11793,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2019-11-14T08:55:22Z
issue,26344,How to get rid of zombie processes using torch.multiprocessing.Pool?,https://github.com/pytorch/pytorch/issues/26344,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dependency bug, module: multiprocessing, oncall: distributed",scan when time allows,2019-11-14T00:08:11Z
issue,8028,Build error : mpi/mpi_gpu_test.cc.o: undefined reference to symbol '_ZN3MPI8Datatype4FreeEv',https://github.com/pytorch/pytorch/issues/8028,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-11-13T09:27:26Z
issue,29692,[FR] add generator= kwarg support for torch.randn and torch.rand,https://github.com/pytorch/pytorch/issues/29692,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: random,scan when time allows,2019-11-13T08:23:31Z
issue,29662,[jit] Dict construction fails at runtime,https://github.com/pytorch/pytorch/issues/29662,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2019-11-12T22:16:59Z
issue,29235,Redo our library structure,https://github.com/pytorch/pytorch/issues/29235,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2019-11-12T22:07:10Z
issue,20151,RuntimeError: Given input size: (2048x1x1). Calculated output size: (2048x-5x-5). Output size is too small at /pytorch/aten/src/THNN/generic/SpatialAveragePooling.c:48,https://github.com/pytorch/pytorch/issues/20151,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: tensorboard",scan when time allows,2019-11-12T12:38:46Z
issue,20204,"SyncBatchNorm should support 2D input (B, C)",https://github.com/pytorch/pytorch/issues/20204,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, oncall: distributed",scan when time allows,2019-11-12T00:29:41Z
issue,29402,[RFC] RPC timeout,https://github.com/pytorch/pytorch/issues/29402,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rpc,scan when time allows,2019-11-11T19:08:02Z
issue,29553,`print` uses lots of GPU memory,https://github.com/pytorch/pytorch/issues/29553,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: printing",scan when time allows,2019-11-11T18:44:20Z
issue,5737,[feature request] Callback on learning rate drop in torch.optim.lr_scheduler.ReduceLROnPlateau,https://github.com/pytorch/pytorch/issues/5737,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2019-11-11T03:47:00Z
issue,25297,torch.as_tensor(bytearray(...)) seems to leak memory,https://github.com/pytorch/pytorch/issues/25297,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: dataloader, module: memory usage",scan when time allows,2019-11-10T22:08:36Z
issue,29429,NLLLoss reduce=True returning nan in float16,https://github.com/pytorch/pytorch/issues/29429,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2019-11-08T15:54:00Z
issue,29419,"IValue can't be constructed from one of `int`, `long`, or `long long`.",https://github.com/pytorch/pytorch/issues/29419,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2019-11-08T14:19:48Z
issue,29377,[jit] Saving a `ScriptFunction` to a buffer doesn't work,https://github.com/pytorch/pytorch/issues/29377,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2019-11-07T19:09:56Z
issue,29172,Modules without copying in multiprocess,https://github.com/pytorch/pytorch/issues/29172,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: multiprocessing,scan when time allows,2019-11-07T17:03:13Z
issue,29254,Add support for ivalue float scalars,https://github.com/pytorch/pytorch/issues/29254,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2019-11-07T16:59:23Z
issue,29369,nn.Conv(n)d constructor doesn't check for the number of kernel dimensions,https://github.com/pytorch/pytorch/issues/29369,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: nn",scan when time allows,2019-11-07T16:37:44Z
issue,1487,[feature request/proposal] Relax scale_factor for nearest neighbor upsampling,https://github.com/pytorch/pytorch/issues/1487,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-11-07T16:32:45Z
issue,11978,[Caffe2] Attempting to install Caffe2 in Google Colab,https://github.com/pytorch/pytorch/issues/11978,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-11-07T13:31:48Z
issue,28745,Problem when installing Pytorch from source on CentOS 7.4,https://github.com/pytorch/pytorch/issues/28745,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2019-11-07T09:25:32Z
issue,27740,Scripting torchvision.models.detection.maskrcnn_resnet50_fpn,https://github.com/pytorch/pytorch/issues/27740,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: vision,scan when time allows,2019-11-07T02:43:34Z
issue,29198,RuntimeError: CUDA out of memory with available GPU memory,https://github.com/pytorch/pytorch/issues/29198,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: memory usage",scan when time allows,2019-11-06T09:17:59Z
issue,28757,[feature request] Docs for fuse_conv_bn_eval,https://github.com/pytorch/pytorch/issues/28757,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2019-11-05T23:14:10Z
issue,29177,[FR] trace_module traces both eval and train graph,https://github.com/pytorch/pytorch/issues/29177,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2019-11-05T22:17:18Z
issue,24946,[RPC] Make ProcessGroupAgent send task non-blocking,https://github.com/pytorch/pytorch/issues/24946,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rpc,scan when time allows,2019-11-05T17:43:04Z
issue,26446,Avoid sending zero grads over the wire in distributed autograd backward pass,https://github.com/pytorch/pytorch/issues/26446,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rpc,scan when time allows,2019-11-05T17:40:15Z
issue,27642,Test re-entrant backward works with torch.distributed.autograd.backward(),https://github.com/pytorch/pytorch/issues/27642,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: rpc",scan when time allows,2019-11-05T17:38:55Z
issue,27641,torch.distributed.autograd.backward() should populate .grad field on Tensors by default.,https://github.com/pytorch/pytorch/issues/27641,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: rpc",scan when time allows,2019-11-05T17:38:55Z
issue,27579,Support Anomaly detection for distributed autograd.,https://github.com/pytorch/pytorch/issues/27579,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: rpc",scan when time allows,2019-11-05T17:38:54Z
issue,27099,Support implicit RRef type conversion,https://github.com/pytorch/pytorch/issues/27099,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rpc,scan when time allows,2019-11-05T17:38:52Z
issue,28925,"Provide rpc, remote and dist autograd C++ APIs and register them as Prim::ops",https://github.com/pytorch/pytorch/issues/28925,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: cpp, module: rpc",scan when time allows,2019-11-05T17:37:07Z
issue,28882,Support RRef[T].__call__(*args) which invokes T.__call__(*args) on owner,https://github.com/pytorch/pytorch/issues/28882,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rpc,scan when time allows,2019-11-05T17:37:06Z
issue,29202,Add instructions for building torch.distributed on macOS,https://github.com/pytorch/pytorch/issues/29202,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, oncall: distributed",scan when time allows,2019-11-05T16:04:44Z
issue,29057,Docker issue for Pytorch 1.3,https://github.com/pytorch/pytorch/issues/29057,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: build, module: docker",scan when time allows,2019-11-05T15:14:31Z
issue,14370,[Caffe2] Check failed: output->size() == values_.size() output size: 1 given size: 1563551,https://github.com/pytorch/pytorch/issues/14370,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-11-05T03:17:39Z
issue,28938,Cannot select version in the tutorials page,https://github.com/pytorch/pytorch/issues/28938,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2019-11-04T21:33:12Z
issue,28515,Get rid of libc10.so,https://github.com/pytorch/pytorch/issues/28515,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2019-11-04T18:18:17Z
issue,28791,[CUDA-MEMCHECK] TestTorchDeviceTypeCUDA.test_pin_memory_from_constructor_cuda fails,https://github.com/pytorch/pytorch/issues/28791,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2019-11-01T23:40:05Z
issue,29028,Inconsistent documentation for in-place functions at https://pytorch.org/docs/stable/torch.html,https://github.com/pytorch/pytorch/issues/29028,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2019-11-01T22:28:47Z
issue,1359,"Conjugate Gradient mentioned in docs, but not implemented",https://github.com/pytorch/pytorch/issues/1359,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2019-11-01T18:25:21Z
issue,29033,Move allgather_coalesced functionality to comm.cpp.,https://github.com/pytorch/pytorch/issues/29033,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2019-11-01T17:31:54Z
issue,29021,Named Tensors: size of a named dimension,https://github.com/pytorch/pytorch/issues/29021,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: named tensor",scan when time allows,2019-11-01T17:26:29Z
issue,12535,[CMake] Linking against Intel OpenMP,https://github.com/pytorch/pytorch/issues/12535,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: mkldnn, module: openmp",scan when time allows,2019-10-31T17:56:15Z
issue,28962,[feature request] Some activations modules missing inplace argument,https://github.com/pytorch/pytorch/issues/28962,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2019-10-31T17:06:36Z
issue,28845,Implement operator<< for bfloat16,https://github.com/pytorch/pytorch/issues/28845,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: bfloat16,scan when time allows,2019-10-31T00:53:20Z
issue,28859,Remove bce_with_logits in derivatives.yaml,https://github.com/pytorch/pytorch/issues/28859,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-10-31T00:36:21Z
issue,28930,Add support for multidimensional input to `at::tensor`,https://github.com/pytorch/pytorch/issues/28930,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp,scan when time allows,2019-10-31T00:30:22Z
issue,16405,error on cmake_version from tools/build_pytorch_libs.py,https://github.com/pytorch/pytorch/issues/16405,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2019-10-29T19:03:56Z
issue,28780,'import torch` fails with Illegal instruction,https://github.com/pytorch/pytorch/issues/28780,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2019-10-28T18:20:35Z
issue,28756,cmake allows both MKL and MKLDNN to be OFF; aten/src/ATen/CMakeLists.txt then ignores c++ sources,https://github.com/pytorch/pytorch/issues/28756,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2019-10-28T17:40:39Z
issue,28721,[docs] Improve docs of nn.MultiheadAttention,https://github.com/pytorch/pytorch/issues/28721,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2019-10-28T17:28:47Z
issue,28417,Patch: Fix for using `clang` to compile CUDA,https://github.com/pytorch/pytorch/issues/28417,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cuda",scan when time allows,2019-10-28T17:10:17Z
issue,20864,Import warning when using the wrong version of CUDA,https://github.com/pytorch/pytorch/issues/20864,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2019-10-27T08:04:45Z
issue,28552,Doing rendezvous twice can cause hangs,https://github.com/pytorch/pytorch/issues/28552,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2019-10-25T16:42:41Z
issue,28655,Cmake fails due to bad python call,https://github.com/pytorch/pytorch/issues/28655,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2019-10-25T15:46:21Z
issue,11451,[caffe2] caffe2 openmp linking error with xcode 9.0(AppleClang 9.0) on mac,https://github.com/pytorch/pytorch/issues/11451,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-10-25T11:29:25Z
issue,26409,Generated file not getting cleaned up by clean,https://github.com/pytorch/pytorch/issues/26409,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2019-10-24T21:56:32Z
issue,15070,In-place operations on `.data` or `.detach()` of sparse tensor doesn't update the original tensor,https://github.com/pytorch/pytorch/issues/15070,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2019-10-24T19:23:01Z
issue,28558,We should suggest using as_strided_ instead of set_,https://github.com/pytorch/pytorch/issues/28558,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2019-10-24T15:30:02Z
issue,28441,gradient of Dirichlet.log_prob gives nan,https://github.com/pytorch/pytorch/issues/28441,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distributions,scan when time allows,2019-10-24T15:23:44Z
issue,28532,TestTorch.test_c10_layer_norm fails if you run it on a build of PyTorch with BUILD_CAFFE2_OPS=0,https://github.com/pytorch/pytorch/issues/28532,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dispatch,scan when time allows,2019-10-24T09:16:53Z
issue,14277,Provide Protobuf library if libtorch was built with included version,https://github.com/pytorch/pytorch/issues/14277,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cpp, module: protobuf",scan when time allows,2019-10-23T22:13:22Z
issue,21061,Misleading Error when doing Large Batch Matrix Multiplication,https://github.com/pytorch/pytorch/issues/21061,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2019-10-23T21:40:07Z
issue,28390,torch.multinominal ignores elements from cumulative distribution,https://github.com/pytorch/pytorch/issues/28390,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: random",scan when time allows,2019-10-23T12:40:10Z
issue,26990,torch.quantized.modules.floatFunctional -- add a little color commentary to the doc,https://github.com/pytorch/pytorch/issues/26990,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2019-10-22T22:46:55Z
issue,16710,culibos linker errors on binary_linux_conda_3.6_cu90_build,https://github.com/pytorch/pytorch/issues/16710,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2019-10-22T07:19:01Z
issue,27687,`nn.Sequential.__setattr__` appends to the execution list,https://github.com/pytorch/pytorch/issues/27687,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2019-10-21T18:35:49Z
issue,28318,`torch.nn.Module._load_state_dict` catch-all error message can be misleading,https://github.com/pytorch/pytorch/issues/28318,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-10-19T17:52:30Z
issue,23299,Add gatherv/allgatherv primitives to support non-equal contribution,https://github.com/pytorch/pytorch/issues/23299,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, oncall: distributed",scan when time allows,2019-10-18T20:48:17Z
issue,26379,DistAutogradContext should be cleaned up in case of node failures.,https://github.com/pytorch/pytorch/issues/26379,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: memory usage, module: rpc",scan when time allows,2019-10-18T20:43:03Z
issue,28249,Add scopes to autograd profiler,https://github.com/pytorch/pytorch/issues/28249,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-10-17T23:21:29Z
issue,28142,Named Tensor: Support -1 in `unflatten`.,https://github.com/pytorch/pytorch/issues/28142,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: named tensor,scan when time allows,2019-10-17T23:14:29Z
issue,28110,Stable docs show dead __config__ section link,https://github.com/pytorch/pytorch/issues/28110,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2019-10-17T22:14:11Z
issue,24354,Default warning handler in C++ doesn't seem to unique warnings,https://github.com/pytorch/pytorch/issues/24354,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: error checking,scan when time allows,2019-10-17T19:44:50Z
issue,28146,Named tensor: align_to align_as error messages,https://github.com/pytorch/pytorch/issues/28146,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: named tensor,scan when time allows,2019-10-17T19:20:16Z
issue,28221,TorchScript doesn't support torch.channels_last or any other memory format constants,https://github.com/pytorch/pytorch/issues/28221,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: memory format, oncall: jit",scan when time allows,2019-10-17T15:05:26Z
issue,28220,TorchScript custom ops like `cuda` `byte` etc. doesn't support memory_format argument,https://github.com/pytorch/pytorch/issues/28220,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2019-10-17T15:04:28Z
issue,9699,[Build error] libcudnn.so: error adding symbols: File in wrong format,https://github.com/pytorch/pytorch/issues/9699,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-10-17T07:19:16Z
issue,20315,libtorch+opencv Mat result error: different from the python ones,https://github.com/pytorch/pytorch/issues/20315,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp,scan when time allows,2019-10-16T07:05:02Z
issue,27708,Documentation makefile should include torchvision,https://github.com/pytorch/pytorch/issues/27708,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: doc infra, module: docs",scan when time allows,2019-10-15T23:56:51Z
issue,27926,`num_batches_tracked` update in `_BatchNorm` forward should be a single scalar update on host regardless of the residence of the layer,https://github.com/pytorch/pytorch/issues/27926,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: nn, module: performance",scan when time allows,2019-10-15T18:44:05Z
issue,21738,symbol lookup error: libmkl_intel_lp64.so: undefined symbol: mkl_blas_dsyrk (binaries built with static linking -DBUILD_SHARED_LIBS=OFF fail due to dynamic linker problem),https://github.com/pytorch/pytorch/issues/21738,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: static linking, module: third_party",scan when time allows,2019-10-15T14:31:07Z
issue,17796,[Caffe2] install error,https://github.com/pytorch/pytorch/issues/17796,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-10-15T08:45:12Z
issue,25314,ConstQuantizerPtr is misnamed,https://github.com/pytorch/pytorch/issues/25314,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2019-10-14T23:10:18Z
issue,25821,Conv2D 2x~20x slower than Tensorflow when channel count is small,https://github.com/pytorch/pytorch/issues/25821,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: cuda, module: cudnn, module: performance",scan when time allows,2019-10-11T16:48:19Z
issue,27175,Easier way to create tensors with names,https://github.com/pytorch/pytorch/issues/27175,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: named tensor,scan when time allows,2019-10-11T13:04:44Z
issue,27015,Missing bin and include when building with torchvision on CentOS,https://github.com/pytorch/pytorch/issues/27015,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cpp, module: vision",scan when time allows,2019-10-10T20:49:37Z
issue,27608,Unify warning logging mechanism,https://github.com/pytorch/pytorch/issues/27608,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: internals,scan when time allows,2019-10-10T00:10:16Z
issue,27540,CTCLoss cuda backend large batch handling takes up to 1.8x more memory,https://github.com/pytorch/pytorch/issues/27540,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: loss, module: memory usage",scan when time allows,2019-10-08T15:05:13Z
issue,26534,MKLDNN+AMD BLIS path for PyTorch,https://github.com/pytorch/pytorch/issues/26534,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: mkldnn",scan when time allows,2019-10-08T14:12:43Z
issue,25478,Delete TensorOptions::operator==,https://github.com/pytorch/pytorch/issues/25478,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: internals,scan when time allows,2019-10-08T10:14:24Z
issue,18850,Allow tracing of models which output `None`,https://github.com/pytorch/pytorch/issues/18850,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2019-10-07T22:51:42Z
issue,25892,load_state_dict on CPU first,https://github.com/pytorch/pytorch/issues/25892,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: serialization,scan when time allows,2019-10-07T20:33:57Z
issue,15864,Deformable Convolution,https://github.com/pytorch/pytorch/issues/15864,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: vision,scan when time allows,2019-10-05T16:17:11Z
issue,27297,CUDA error: device-side assert triggered(insert_events at /pytorch/c10/cuda/CUDACachingAllocator.cpp:569),https://github.com/pytorch/pytorch/issues/27297,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2019-10-04T17:26:27Z
issue,27218,pytorch data_parallel oom on gpu:0,https://github.com/pytorch/pytorch/issues/27218,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: data parallel, module: memory usage",scan when time allows,2019-10-04T08:10:56Z
issue,26730,[jit] Bad error when instantiating TorchScript class with incorrect types,https://github.com/pytorch/pytorch/issues/26730,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2019-10-03T21:27:12Z
issue,25410,[jit] NamedTuples don't respect `__new__`,https://github.com/pytorch/pytorch/issues/25410,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2019-10-03T21:24:47Z
issue,25329,[jit] Bad error for incorrect container type,https://github.com/pytorch/pytorch/issues/25329,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2019-10-03T21:24:47Z
issue,24786,[jit] Dict iterator invalidation doesn't match Python,https://github.com/pytorch/pytorch/issues/24786,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2019-10-03T21:24:45Z
issue,21544,"[jit] In pickler, don't memoize if not necessary",https://github.com/pytorch/pytorch/issues/21544,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2019-10-03T21:16:31Z
issue,21442,[JIT] kwarg with default doesn't work for class instantiation,https://github.com/pytorch/pytorch/issues/21442,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2019-10-03T21:16:31Z
issue,20119,[JIT] Source highlighting doesn't line up when tabs are used for indentation,https://github.com/pytorch/pytorch/issues/20119,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2019-10-03T21:15:42Z
issue,27138,CUDAPytorchToCaffe2.MutualResizes is flaky,https://github.com/pytorch/pytorch/issues/27138,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: flaky-tests, module: tests",scan when time allows,2019-10-03T17:22:41Z
issue,27046,The Gather problem in DataParallel: dimension are not matched.,https://github.com/pytorch/pytorch/issues/27046,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: data parallel,scan when time allows,2019-10-02T08:41:13Z
issue,24234,torch.utils.tensorboard.SummaryWriter fails to flush at program exit,https://github.com/pytorch/pytorch/issues/24234,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tensorboard,scan when time allows,2019-10-01T21:25:21Z
issue,24102,[JIT] script doesn't convert dtypes back to torch.dtype from long,https://github.com/pytorch/pytorch/issues/24102,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2019-09-30T21:23:42Z
issue,27055,torch::NoGradGuard no_grad get wrong when I use batchsize!=1,https://github.com/pytorch/pytorch/issues/27055,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp,scan when time allows,2019-09-30T18:18:29Z
issue,27021,Cmake warnings during build,https://github.com/pytorch/pytorch/issues/27021,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build warnings,scan when time allows,2019-09-30T18:15:04Z
issue,27034,RuntimeError:[enforce fail at context.h:48] option.device_type() ==PROTO_CPU. 1vs0,https://github.com/pytorch/pytorch/issues/27034,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-09-30T18:11:38Z
issue,27024,Is it an incompleted dst tensor synchronization in CUDA device to device copy ?,https://github.com/pytorch/pytorch/issues/27024,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2019-09-30T18:10:29Z
issue,26950,conv2d Memory usage is too large； pytorch 1.1.0,https://github.com/pytorch/pytorch/issues/26950,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: cudnn, module: dependency bug, module: memory usage",scan when time allows,2019-09-30T17:11:46Z
issue,25691,[dataloader] Hang because of too many open files (and probably some process dead),https://github.com/pytorch/pytorch/issues/25691,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2019-09-29T23:03:31Z
issue,26967,Lint rule to test for creation of tensor in native/ without options(),https://github.com/pytorch/pytorch/issues/26967,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: lint",scan when time allows,2019-09-27T14:58:51Z
issue,26964,AnyValueTest.CorrectlyAccessesIntWhenCorrectType UBSAN failure: owncast of address 0x60300105d750 which does not point to an object of type 'Holder<const int>' Sep 27 00:01:03 0x60300105d750: note: object is of type 'torch::nn::AnyModule::Value::Holder<int>',https://github.com/pytorch/pytorch/issues/26964,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cpp",scan when time allows,2019-09-27T13:43:47Z
issue,26863,[BUG Report]Integrate libtorch to ffmpeg but memory leak happened!,https://github.com/pytorch/pytorch/issues/26863,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: vision",scan when time allows,2019-09-26T21:24:14Z
issue,9029,DataWorkersTest::testRNNInput timeout,https://github.com/pytorch/pytorch/issues/9029,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-09-26T14:31:20Z
issue,26796,setup.py install error,https://github.com/pytorch/pytorch/issues/26796,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2019-09-25T22:12:27Z
issue,14487,TestAdagrad.test_row_wise_sparse_adagrad intermittently fails health check,https://github.com/pytorch/pytorch/issues/14487,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-09-25T19:47:09Z
issue,26659,torch::nn::Sequential not compatible with torch::nn::RNN,https://github.com/pytorch/pytorch/issues/26659,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: nn",scan when time allows,2019-09-25T13:47:24Z
issue,15149,Use std::variant to represent C++ side enumerations (with binding support),https://github.com/pytorch/pytorch/issues/15149,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp,scan when time allows,2019-09-24T17:00:54Z
issue,26396,NNPACK condition should be changed (ARM processors),https://github.com/pytorch/pytorch/issues/26396,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: nnpack",scan when time allows,2019-09-24T14:06:11Z
issue,26567,Make `torch.save` serialize a zip file,https://github.com/pytorch/pytorch/issues/26567,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: serialization,scan when time allows,2019-09-24T14:05:53Z
issue,26258,"""git describe"" shows incorrect version 1.0 instead of 1.2",https://github.com/pytorch/pytorch/issues/26258,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2019-09-23T17:05:46Z
issue,22682,CMAKE_PARSE_IMPLICIT_LINK_INFO Function invoked with incorrect arguments,https://github.com/pytorch/pytorch/issues/22682,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: flaky-tests",scan when time allows,2019-09-23T13:54:25Z
issue,12869,caffe2: Unsupported type of tensor: nullptr (uninitialized),https://github.com/pytorch/pytorch/issues/12869,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-09-23T12:56:25Z
issue,12983,Fail to Run Caffe2 with Successful Build: This caffe2 python run does not have GPU support,https://github.com/pytorch/pytorch/issues/12983,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-09-22T21:02:05Z
issue,17458,Overflow in fbgemm,https://github.com/pytorch/pytorch/issues/17458,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-09-22T07:41:46Z
issue,21745,torch.save also saves docstrings into pickle for some reason,https://github.com/pytorch/pytorch/issues/21745,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: serialization,scan when time allows,2019-09-20T21:22:54Z
issue,26551,No way to disable mse_loss broadcasting warning,https://github.com/pytorch/pytorch/issues/26551,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2019-09-20T18:48:05Z
issue,24478,Port CPU_tensor_apply functions to TensorIterator (umbrella issue),https://github.com/pytorch/pytorch/issues/24478,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpu_tensor_apply,scan when time allows,2019-09-20T18:09:28Z
issue,12117,"Error: Internal Compiler Error (codegen): ""there was an error in verifying the lgenfe output!""",https://github.com/pytorch/pytorch/issues/12117,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cuda",scan when time allows,2019-09-17T19:33:52Z
issue,26228,Higher dimension support for `MultiLableSoftMarginLoss`,https://github.com/pytorch/pytorch/issues/26228,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: loss, module: nn",scan when time allows,2019-09-17T19:31:29Z
issue,26338,Behavior of F.dropout in eval mode,https://github.com/pytorch/pytorch/issues/26338,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2019-09-17T19:05:52Z
issue,26110,Access data_ptr in RNN.cpp,https://github.com/pytorch/pytorch/issues/26110,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rnn,scan when time allows,2019-09-17T15:39:48Z
issue,24016,Multiplying a very large CUDA tensor with another tensor yields unexpected result,https://github.com/pytorch/pytorch/issues/24016,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: dependency bug",scan when time allows,2019-09-17T14:02:43Z
issue,17354,testConvnetBenchmarks intermittently segfaults,https://github.com/pytorch/pytorch/issues/17354,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: flaky-tests,scan when time allows,2019-09-13T15:21:26Z
issue,26136,There should be gating around BFloat16,https://github.com/pytorch/pytorch/issues/26136,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bfloat16, module: performance",scan when time allows,2019-09-13T15:20:38Z
issue,25991,Why doc building isn't failing us for referring to a non-existent method?,https://github.com/pytorch/pytorch/issues/25991,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2019-09-13T10:54:45Z
issue,26058,sccache stats can cause whole build to fail,https://github.com/pytorch/pytorch/issues/26058,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2019-09-13T10:48:23Z
issue,25548,The inference speed of the torch c++ dynamic library compiled manually is slower than the torch library officially provided,https://github.com/pytorch/pytorch/issues/25548,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: cuda, module: performance",scan when time allows,2019-09-12T16:55:35Z
issue,25880,Incorrect lable read with ImageInput Op of Caffe2,https://github.com/pytorch/pytorch/issues/25880,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-09-11T01:53:36Z
issue,25752,torch.cuda.empty_cache() write data to gpu0,https://github.com/pytorch/pytorch/issues/25752,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2019-09-10T21:39:08Z
issue,23657,Accelerate PyTorch just-in-time compilation using MKL-DNN,https://github.com/pytorch/pytorch/issues/23657,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: mkldnn, oncall: jit",scan when time allows,2019-09-10T19:19:14Z
issue,25820,Pytorch master can not build with computer capability 3.0 under Mac OS X 10.13.16 with Nvidia GT 750m,https://github.com/pytorch/pytorch/issues/25820,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: macos",scan when time allows,2019-09-10T01:52:29Z
issue,25743,Custom sampler for Seq2Seq models to avoid padding,https://github.com/pytorch/pytorch/issues/25743,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2019-09-09T21:20:11Z
issue,25747,Int32 overflow in bincount indexing,https://github.com/pytorch/pytorch/issues/25747,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: 64-bit, module: cuda",scan when time allows,2019-09-09T21:09:06Z
issue,25417,Add a mode to check input tensor sizes in allreduce_coalesced,https://github.com/pytorch/pytorch/issues/25417,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2019-09-09T09:50:48Z
issue,14318,No code example for AdaptiveLogSoftmaxWithLoss,https://github.com/pytorch/pytorch/issues/14318,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2019-09-06T18:18:41Z
issue,25460,Build link not right,https://github.com/pytorch/pytorch/issues/25460,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cpp",scan when time allows,2019-09-06T05:34:05Z
issue,25648,"Torch.jit.trace unexpected error with `torch.cat(…, dim=-1)`",https://github.com/pytorch/pytorch/issues/25648,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2019-09-05T21:07:46Z
issue,19172,[Feature Request] Common constants in the torch.* namespace,https://github.com/pytorch/pytorch/issues/19172,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: numpy,scan when time allows,2019-09-05T18:50:03Z
issue,25635,[C++] Support negative index in `torch::TensorAccessor::size()`,https://github.com/pytorch/pytorch/issues/25635,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp,scan when time allows,2019-09-05T11:47:10Z
issue,25514,Multithreaded backpropagation with custom autograd.Functions,https://github.com/pytorch/pytorch/issues/25514,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: data parallel,scan when time allows,2019-09-04T22:02:56Z
issue,25552,Usage of DDP on a module that doesn't require gradients,https://github.com/pytorch/pytorch/issues/25552,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2019-09-04T20:19:20Z
issue,21700,(LLD 8.0.0) ld: error: can't create dynamic relocation R_X86_64_DTPOFF32 against symbol: ideep::utils::computation_cache,https://github.com/pytorch/pytorch/issues/21700,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: mkldnn, module: static linking",scan when time allows,2019-09-04T01:02:53Z
issue,25601,Avoid non-POD data in thread_local,https://github.com/pytorch/pytorch/issues/25601,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: multithreading, module: performance",scan when time allows,2019-09-04T00:27:23Z
issue,25520,pytorch c++ api cannot call operator() on torch::nn::Sequential,https://github.com/pytorch/pytorch/issues/25520,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: nn",scan when time allows,2019-09-03T18:19:17Z
issue,25522,[dataloader] Problem in exception reraise mechanism,https://github.com/pytorch/pytorch/issues/25522,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2019-09-03T15:41:02Z
issue,24831,subprocess.CalledProcessError: Compile source in NVIDIA TX2,https://github.com/pytorch/pytorch/issues/24831,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cuda",scan when time allows,2019-09-03T13:23:23Z
issue,16268,Complete dtype support for torch.norm,https://github.com/pytorch/pytorch/issues/16268,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2019-08-30T21:51:38Z
issue,12795,C++ frontend: how to debug nan gradients,https://github.com/pytorch/pytorch/issues/12795,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp,scan when time allows,2019-08-30T21:12:31Z
issue,25480,Vectorize bool operations,https://github.com/pytorch/pytorch/issues/25480,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: boolean tensor,scan when time allows,2019-08-30T20:57:03Z
issue,25381,regarding builtin_function_or_method,https://github.com/pytorch/pytorch/issues/25381,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-08-29T14:50:59Z
issue,23940,Improve the performance of linear algebra operations in CUDA for small problem sizes,https://github.com/pytorch/pytorch/issues/23940,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: linear algebra, module: performance",scan when time allows,2019-08-28T22:02:29Z
issue,6402,[caffe2] How to freeze a layer?,https://github.com/pytorch/pytorch/issues/6402,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-08-28T21:08:12Z
issue,9701,[request] speed-up multidim slicing backward,https://github.com/pytorch/pytorch/issues/9701,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: performance",scan when time allows,2019-08-28T15:46:29Z
issue,24090,`suggest_memory_format` has ambiguity & cannot represent intended layout format for corner cases,https://github.com/pytorch/pytorch/issues/24090,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: internals,scan when time allows,2019-08-28T09:03:18Z
issue,15994,More data type support for gather_map,https://github.com/pytorch/pytorch/issues/15994,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2019-08-27T16:10:25Z
issue,17041,Errors running distributed example,https://github.com/pytorch/pytorch/issues/17041,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2019-08-26T02:32:44Z
issue,13224,Caffe2: Causes error when using flag remove_legacy_pad while converting from caffe to caffe2,https://github.com/pytorch/pytorch/issues/13224,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-08-24T08:51:22Z
issue,9310,/usr/bin/ld: cannot find -lpthreads,https://github.com/pytorch/pytorch/issues/9310,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2019-08-23T20:29:26Z
issue,25057,ctc_loss computes different losses and gradients on batched utterances vs. individual utterances,https://github.com/pytorch/pytorch/issues/25057,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: loss,scan when time allows,2019-08-23T19:40:04Z
issue,25070,[FR] torch.(Generator|random).seed allows specifying the seed value,https://github.com/pytorch/pytorch/issues/25070,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: random,scan when time allows,2019-08-23T19:33:25Z
issue,25071,Generator objects should not always use the same seed,https://github.com/pytorch/pytorch/issues/25071,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: random,scan when time allows,2019-08-23T19:32:43Z
issue,25112,[RFC] NestedTensor - 0.0.2,https://github.com/pytorch/pytorch/issues/25112,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-08-23T19:12:14Z
issue,24985,Data worker should fetch a sample instead of a batch.,https://github.com/pytorch/pytorch/issues/24985,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2019-08-23T19:04:26Z
issue,25034,"make add_module accept tuples as well or change containers(ModuleList, Sequential, etc) to allow this",https://github.com/pytorch/pytorch/issues/25034,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2019-08-23T18:36:16Z
issue,23720,segmentation faults when using multiprocessing_context='spawn' with large number of processes,https://github.com/pytorch/pytorch/issues/23720,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dataloader, module: multiprocessing, oncall: distributed",scan when time allows,2019-08-23T04:03:40Z
issue,25014,Benchmark cuDNN affine_grid_generator vs native,https://github.com/pytorch/pytorch/issues/25014,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: cudnn, module: nn, module: performance",scan when time allows,2019-08-23T03:55:30Z
issue,18805,Running custom operator tests manually is too difficult,https://github.com/pytorch/pytorch/issues/18805,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp-extensions, module: tests",scan when time allows,2019-08-22T20:30:04Z
issue,24956,[RPC] Fix logging initialization warning in ProcessGroupAgent,https://github.com/pytorch/pytorch/issues/24956,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: internals,scan when time allows,2019-08-21T15:28:55Z
issue,22128,Automatic rank selection when using file:// initialization method,https://github.com/pytorch/pytorch/issues/22128,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2019-08-21T06:20:18Z
issue,24836,Gloo scatter gives wrong result for stride != 1,https://github.com/pytorch/pytorch/issues/24836,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, oncall: distributed",scan when time allows,2019-08-20T15:20:11Z
issue,21318,"Undefined symbols for architecture x86_64: ""testing::internal::UntypedFunctionMockerBase::UntypedInvokeWith(void const*)"" on Mac OS X",https://github.com/pytorch/pytorch/issues/21318,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: internals, module: macos",scan when time allows,2019-08-20T15:18:31Z
issue,24353,Visual Studio Code not providing autosuggestions for submodules,https://github.com/pytorch/pytorch/issues/24353,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-08-20T15:13:12Z
issue,24336,Auto tuner takes too much time in serialized model,https://github.com/pytorch/pytorch/issues/24336,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2019-08-19T18:29:01Z
issue,24419,Building Python bits separate from C++ bits and making one play well with the other,https://github.com/pytorch/pytorch/issues/24419,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2019-08-19T16:24:39Z
issue,24798,Installs empty directories under Python's sitelibdir,https://github.com/pytorch/pytorch/issues/24798,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2019-08-19T16:23:34Z
issue,24500,Failed to compile PyTorch on IBM Power 9 architecture with CUDA 10,https://github.com/pytorch/pytorch/issues/24500,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2019-08-19T16:21:41Z
issue,24456,Build PyTorch 1.2.0 occur `recipe for target bin/test_parallel' failed,https://github.com/pytorch/pytorch/issues/24456,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2019-08-19T16:11:41Z
issue,24432,Use a ScriptModule on GPU that was saved from CPU,https://github.com/pytorch/pytorch/issues/24432,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: serialization, oncall: jit",scan when time allows,2019-08-19T16:10:42Z
issue,24343,Unified representation for enum types,https://github.com/pytorch/pytorch/issues/24343,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2019-08-19T16:10:07Z
issue,24417,1.0rc0-6216 installs empty directories under include and duplicates under /,https://github.com/pytorch/pytorch/issues/24417,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2019-08-19T16:08:36Z
issue,24498,Migrate CPU_tensor_apply to TensorIterator in aten/src/ATen/native/TensorCompare.cpp:30,https://github.com/pytorch/pytorch/issues/24498,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpu_tensor_apply,scan when time allows,2019-08-16T17:27:42Z
issue,24167,Failed to build pytorch with NanoPi M4,https://github.com/pytorch/pytorch/issues/24167,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2019-08-16T15:56:18Z
issue,24079,[Caffe2] build android in v1.1.0 with headfile error,https://github.com/pytorch/pytorch/issues/24079,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-08-14T22:46:23Z
issue,24264,"TensorIterator ""builder"" options should be documented.",https://github.com/pytorch/pytorch/issues/24264,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: internals,scan when time allows,2019-08-14T18:31:16Z
issue,8098,Please provide wheel package for windows on PyPI,https://github.com/pytorch/pytorch/issues/8098,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: binaries,scan when time allows,2019-08-13T23:00:11Z
issue,24193,TensorIterator stubs are designed for merge conflicts.,https://github.com/pytorch/pytorch/issues/24193,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: internals,scan when time allows,2019-08-13T08:41:39Z
issue,23749,Wrong device in graph - Tensorboard SummaryWriter,https://github.com/pytorch/pytorch/issues/23749,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tensorboard,scan when time allows,2019-08-12T19:02:50Z
issue,23771,SummaryWriter doesn't read comment if log_dir precised,https://github.com/pytorch/pytorch/issues/23771,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tensorboard,scan when time allows,2019-08-12T17:58:21Z
issue,23787,Better version of chrome://tracing,https://github.com/pytorch/pytorch/issues/23787,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2019-08-12T17:20:04Z
issue,23768,Better documentation about PyTorch's dependencies,https://github.com/pytorch/pytorch/issues/23768,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: docs, module: third_party",scan when time allows,2019-08-12T17:08:54Z
issue,5857,Change THCudaCheck to suggest that device-side asserts likely mean that you have out of bound indices,https://github.com/pytorch/pytorch/issues/5857,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: assert failure, module: cuda, module: error checking",scan when time allows,2019-08-12T13:54:34Z
issue,23732,nn.Module.forward signature with **kwargs,https://github.com/pytorch/pytorch/issues/23732,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: activation checkpointing, module: nn",scan when time allows,2019-08-12T12:49:08Z
issue,23865,model->to(device) costs over a millisecond when doing nothing,https://github.com/pytorch/pytorch/issues/23865,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: performance,scan when time allows,2019-08-07T06:56:53Z
issue,23896,Remove USE_C10D flag,https://github.com/pytorch/pytorch/issues/23896,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, oncall: distributed",scan when time allows,2019-08-06T20:48:04Z
issue,21554,Pytorch hangs when dataloader multiprocessing workers are killed,https://github.com/pytorch/pytorch/issues/21554,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2019-08-06T01:04:00Z
issue,23721,Using PyTorch on AWS EFA network,https://github.com/pytorch/pytorch/issues/23721,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dependency bug, oncall: distributed",scan when time allows,2019-08-05T17:13:48Z
issue,23490,upcoming PEP 554: how much effort we need to support sub-interpreter,https://github.com/pytorch/pytorch/issues/23490,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-08-02T16:16:00Z
issue,19682,View in Sequential - reasoned case for it,https://github.com/pytorch/pytorch/issues/19682,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2019-08-01T21:51:57Z
issue,23676,"Bogus ""Your compiler (clang++) is not compatible"" message",https://github.com/pytorch/pytorch/issues/23676,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: macos",scan when time allows,2019-08-01T19:59:21Z
issue,23072,Inconsistent axis argument names in torch.diagonal and torch.transpose,https://github.com/pytorch/pytorch/issues/23072,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2019-08-01T02:55:31Z
issue,23540,Error from PyTorch when finalizing Python embedded in C++,https://github.com/pytorch/pytorch/issues/23540,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: pybind,scan when time allows,2019-07-31T17:42:49Z
issue,21454,[JIT] Memory Leak during tracing?,https://github.com/pytorch/pytorch/issues/21454,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2019-07-30T15:36:37Z
issue,23512,Build reconfiguration should consistently honor env variables,https://github.com/pytorch/pytorch/issues/23512,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2019-07-29T18:51:05Z
issue,23094,Versioning for libtorch nightlies,https://github.com/pytorch/pytorch/issues/23094,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: build",scan when time allows,2019-07-29T18:25:01Z
issue,11633,DataLoader: Could not wrapper a exception in threads,https://github.com/pytorch/pytorch/issues/11633,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dataloader, module: error checking",scan when time allows,2019-07-29T17:19:48Z
issue,6135,Test suite should test implementations,https://github.com/pytorch/pytorch/issues/6135,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tests,scan when time allows,2019-07-29T15:30:36Z
issue,18849,More efficient STFT on CUDA,https://github.com/pytorch/pytorch/issues/18849,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: performance",scan when time allows,2019-07-29T04:20:32Z
issue,12855,Model with Caffe2 runs much slower than it with pytorch in GPU mode !!!!,https://github.com/pytorch/pytorch/issues/12855,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-07-29T00:35:30Z
issue,23429,Unreachable code in tanh,https://github.com/pytorch/pytorch/issues/23429,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-07-26T22:32:39Z
issue,21015,How to use Infiniband for cpu-cluster with backend gloo?,https://github.com/pytorch/pytorch/issues/21015,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2019-07-26T02:26:22Z
issue,23364,dyndep function not working,https://github.com/pytorch/pytorch/issues/23364,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-07-25T04:20:47Z
issue,23301,Error while using Libtorch + OpenCV + Qt Creator,https://github.com/pytorch/pytorch/issues/23301,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp,scan when time allows,2019-07-24T21:43:25Z
issue,23240,UBSAN failure in test_simple_model (__main__.TestTensorBoardNumpy): runtime error: call to function pybind11::class_<caffe2::GradientWrapper>::dealloc(pybind11::detail::value_and_holder&) through pointer to incorrect function type 'void (*)(pybind11::detail::value_and_holder &)',https://github.com/pytorch/pytorch/issues/23240,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tests,scan when time allows,2019-07-23T18:44:59Z
issue,23103,Test utility for non-contiguous tensors,https://github.com/pytorch/pytorch/issues/23103,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tests,scan when time allows,2019-07-23T15:38:33Z
issue,23230,Consolidate multiprocessing helpers in distributed tests,https://github.com/pytorch/pytorch/issues/23230,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2019-07-23T15:05:54Z
issue,23079,update docs that sorting is not needed in,https://github.com/pytorch/pytorch/issues/23079,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: rnn",scan when time allows,2019-07-22T17:36:36Z
issue,23068,[c++] torch::conv2d() expected output_padding to be a single integer value or a list of 3 values,https://github.com/pytorch/pytorch/issues/23068,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: cpp, module: docs, module: nn",scan when time allows,2019-07-19T19:48:21Z
issue,23054,ConcatDataset returns different error messages setting out of range plus index and minus index.,https://github.com/pytorch/pytorch/issues/23054,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2019-07-19T18:53:16Z
issue,6794,[Caffe2] TensorProtosDBInput AttributeError,https://github.com/pytorch/pytorch/issues/6794,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-07-19T00:36:08Z
issue,23026,BatchNorm1d fails on first run through GPU,https://github.com/pytorch/pytorch/issues/23026,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: nn",scan when time allows,2019-07-18T21:21:56Z
issue,21818,C++ ABI - Coupling different libraries issue,https://github.com/pytorch/pytorch/issues/21818,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2019-07-18T06:14:15Z
issue,22687,DispatchStub should report what operator it failed to find kernel for,https://github.com/pytorch/pytorch/issues/22687,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: internals",scan when time allows,2019-07-17T15:27:09Z
issue,22868,Make it easier to bisect on PyTorch,https://github.com/pytorch/pytorch/issues/22868,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: ci",scan when time allows,2019-07-17T11:53:05Z
issue,22859,CRITICAL:root:Cannot load caffe2.python. Error: DLL load failed: The specified module could not be found.,https://github.com/pytorch/pytorch/issues/22859,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-07-17T06:22:46Z
issue,22879,Difference between dropout2d and dropout3d,https://github.com/pytorch/pytorch/issues/22879,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2019-07-17T01:04:31Z
issue,22785,Getting cuda runtime error (48) with Jetson TX2 when running simple program,https://github.com/pytorch/pytorch/issues/22785,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2019-07-13T00:22:57Z
issue,22766,assert_no_internal_overlap should pass const char*,https://github.com/pytorch/pytorch/issues/22766,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: internals,scan when time allows,2019-07-13T00:17:56Z
issue,14478,Flaky download from files.pythonhosted.org when installing botocore,https://github.com/pytorch/pytorch/issues/14478,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: flaky-tests,scan when time allows,2019-07-12T19:10:05Z
issue,20488,RoiAlignTest.CheckCPUGPUEqual is still flaky,https://github.com/pytorch/pytorch/issues/20488,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: flaky-tests,scan when time allows,2019-07-12T18:21:29Z
issue,22495,`binary_linux_libtorch_2.7m_cu100_devtoolset3_build` times out after running for 5 hours,https://github.com/pytorch/pytorch/issues/22495,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2019-07-11T22:17:33Z
issue,22604,"""CrossEntropyLoss"" should mention in its name that it takes softmax for target",https://github.com/pytorch/pytorch/issues/22604,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: loss, module: nn",scan when time allows,2019-07-11T00:12:56Z
issue,22414,CPU random number generator is slow,https://github.com/pytorch/pytorch/issues/22414,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: performance, module: random",scan when time allows,2019-07-10T19:34:05Z
issue,22669,BFloat16 numeric limits should contain more info,https://github.com/pytorch/pytorch/issues/22669,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: bfloat16,scan when time allows,2019-07-10T16:29:29Z
issue,21780,cumsum cuda numerical instability,https://github.com/pytorch/pytorch/issues/21780,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: numerical-stability",scan when time allows,2019-07-10T13:35:07Z
issue,22609,caffe_translator TranslateCrop fails when more than one dimensions is cropped,https://github.com/pytorch/pytorch/issues/22609,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-07-10T08:01:21Z
issue,5280,BatchNorm1d raises RuntimeError (CUDNN_STATUS_BAD_PARAM) on 3D input.,https://github.com/pytorch/pytorch/issues/5280,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cudnn,scan when time allows,2019-07-10T07:59:44Z
issue,13304,ASSERT FAILED at /opt/conda/conda-bld/pytorch-nightly_1539602533843/work/aten/src/ATen/core/blob.h:79,https://github.com/pytorch/pytorch/issues/13304,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-07-10T06:09:18Z
issue,22580,Pytorch compilation error on Mac OS,https://github.com/pytorch/pytorch/issues/22580,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2019-07-09T21:01:49Z
issue,20682,[distribution] Support for various domain for AffineTransform,https://github.com/pytorch/pytorch/issues/20682,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distributions,scan when time allows,2019-07-08T17:39:28Z
issue,22585,[dataloader] Mysterious error when using spawn start_method,https://github.com/pytorch/pytorch/issues/22585,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2019-07-08T16:44:13Z
issue,22578,torch.gels runs 100 time slower on gpu than on cpu,https://github.com/pytorch/pytorch/issues/22578,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: performance,scan when time allows,2019-07-08T14:58:39Z
issue,22557,CPU torch.exponential_ function may generate 0 which can cause downstream NaN,https://github.com/pytorch/pytorch/issues/22557,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: random,scan when time allows,2019-07-08T14:36:40Z
issue,22494,Error in equation,https://github.com/pytorch/pytorch/issues/22494,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2019-07-07T10:46:33Z
issue,22536,libtorch new op,https://github.com/pytorch/pytorch/issues/22536,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: custom-operators, module: docs",scan when time allows,2019-07-06T21:12:05Z
issue,22532,Know which function is used by conv and force to use a function,https://github.com/pytorch/pytorch/issues/22532,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: nn",scan when time allows,2019-07-05T10:38:02Z
issue,22274,second derivatives of unfold,https://github.com/pytorch/pytorch/issues/22274,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: derivatives,scan when time allows,2019-07-04T16:37:43Z
issue,18011,caffe2 Segmentation fault (core dumped),https://github.com/pytorch/pytorch/issues/18011,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-07-04T07:17:07Z
issue,22040,SigAbort while running the Caffe2 unit test - thread_init_test - built on Clang7 +glibc 2.23,https://github.com/pytorch/pytorch/issues/22040,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-07-04T04:39:52Z
issue,22277,Handle all IntArrayRef expansions in ATen,https://github.com/pytorch/pytorch/issues/22277,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2019-07-03T09:01:41Z
issue,22406,Build failure with setup.py,https://github.com/pytorch/pytorch/issues/22406,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2019-07-03T06:16:06Z
issue,22455,"Move csrc/distributed/c10d/{comm,reducer} to libtorch.so",https://github.com/pytorch/pytorch/issues/22455,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2019-07-02T17:39:28Z
issue,22383,Storage operation failing on second GPU,https://github.com/pytorch/pytorch/issues/22383,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2019-07-01T21:38:37Z
issue,22375,Label.dim Enforcement Check in AccuracyOp,https://github.com/pytorch/pytorch/issues/22375,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-07-01T21:17:00Z
issue,22370,returned non-zero exit status 2.,https://github.com/pytorch/pytorch/issues/22370,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2019-07-01T20:56:44Z
issue,22358,"When I run python setup.py install to install Caffe2, I have an error: ""No such file or directory: 'nvcc': 'nvcc'""",https://github.com/pytorch/pytorch/issues/22358,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-07-01T20:45:17Z
issue,12308,nn.functional.linear() for sparse tensor,https://github.com/pytorch/pytorch/issues/12308,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: sparse",scan when time allows,2019-07-01T17:25:42Z
issue,22340,Reducer bucketing based on autograd profile,https://github.com/pytorch/pytorch/issues/22340,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2019-06-28T14:03:57Z
issue,17249,Proposal: Add __tensor_wrap__ method similar to numpy __array_wrap__,https://github.com/pytorch/pytorch/issues/17249,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: internals, module: numpy",scan when time allows,2019-06-27T18:25:00Z
issue,7471,Make c10d/FileStore cache file descriptor,https://github.com/pytorch/pytorch/issues/7471,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2019-06-27T14:33:09Z
issue,22252,[libtorch] header warning suppression,https://github.com/pytorch/pytorch/issues/22252,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build warnings,scan when time allows,2019-06-27T03:53:38Z
issue,22220,[doc] nn.Module.forward documentation unclear,https://github.com/pytorch/pytorch/issues/22220,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: nn",scan when time allows,2019-06-26T14:40:33Z
issue,12819,Initialization error when moving data to the GPU,https://github.com/pytorch/pytorch/issues/12819,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2019-06-26T14:36:59Z
issue,15309,libtorch without cmake,https://github.com/pytorch/pytorch/issues/15309,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cpp",scan when time allows,2019-06-26T02:56:09Z
issue,21899,Crash when using tensor.set_data() function in libtorch on windows,https://github.com/pytorch/pytorch/issues/21899,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp,scan when time allows,2019-06-26T02:33:03Z
issue,21902,contradictory output values,https://github.com/pytorch/pytorch/issues/21902,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp,scan when time allows,2019-06-26T02:28:14Z
issue,22232,[FYI] Introducing Quantized Tensor,https://github.com/pytorch/pytorch/issues/22232,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2019-06-25T21:30:01Z
issue,21777,typo and missing return statements,https://github.com/pytorch/pytorch/issues/21777,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: third_party",scan when time allows,2019-06-25T18:22:30Z
issue,8044,[PyTorch] Windows CI CUDA mem leak check on BN tests are flaky,https://github.com/pytorch/pytorch/issues/8044,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: flaky-tests, module: memory usage",scan when time allows,2019-06-24T19:04:07Z
issue,9563,batch_sampler/test_worker_seed intermittently fails with address already in use on OS X,https://github.com/pytorch/pytorch/issues/9563,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: flaky-tests, module: serialization",scan when time allows,2019-06-24T19:03:02Z
issue,22136,[dataloader] SIGCHLD handler should poll the queue for exception first,https://github.com/pytorch/pytorch/issues/22136,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2019-06-24T18:30:33Z
issue,22035,Have a different way to check if gradient was computed in the optimizer (not checking for None),https://github.com/pytorch/pytorch/issues/22035,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: optimizer",scan when time allows,2019-06-24T18:21:50Z
issue,21847,Logging mode for saying when tensor broadcast occurs,https://github.com/pytorch/pytorch/issues/21847,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: internals, module: molly-guard",scan when time allows,2019-06-24T18:17:29Z
issue,3931,Suppress hidden state output of RNNs?,https://github.com/pytorch/pytorch/issues/3931,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: memory usage,scan when time allows,2019-06-24T16:18:43Z
issue,21682,C++ module API footgun: assigning to parameter doesn't update `parameters()` list,https://github.com/pytorch/pytorch/issues/21682,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2019-06-23T14:36:11Z
issue,21673,[FR] Diagonal Transform for Distributions,https://github.com/pytorch/pytorch/issues/21673,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distributions,scan when time allows,2019-06-23T12:07:17Z
issue,21731,Improve multithreaded random number generation (RNG),https://github.com/pytorch/pytorch/issues/21731,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: multithreading, module: random",scan when time allows,2019-06-20T07:00:24Z
issue,19016,Clean up and consolidate DDP tests,https://github.com/pytorch/pytorch/issues/19016,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: tests, oncall: distributed",scan when time allows,2019-06-18T22:13:21Z
issue,21828,torch::zeros is slow for small tensors (C++),https://github.com/pytorch/pytorch/issues/21828,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: performance",scan when time allows,2019-06-18T21:02:47Z
issue,20273,Linker errors when linking statically (avx perfkernels) [Caffe2],https://github.com/pytorch/pytorch/issues/20273,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-06-16T19:29:54Z
issue,21824,How about add torch::end for slicing in c++ frontend,https://github.com/pytorch/pytorch/issues/21824,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: internals,scan when time allows,2019-06-15T22:32:04Z
issue,21766,Missing header files for prebuilt libtorch,https://github.com/pytorch/pytorch/issues/21766,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-06-14T20:56:59Z
issue,21277,[caffe2] check Range Operator inputs with bug,https://github.com/pytorch/pytorch/issues/21277,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-06-14T06:53:35Z
issue,8430,batchnorm2d track_running_stats,https://github.com/pytorch/pytorch/issues/8430,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: serialization,scan when time allows,2019-06-13T23:08:18Z
issue,21684,The cuda problem in caffe2,https://github.com/pytorch/pytorch/issues/21684,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-06-13T21:55:12Z
issue,21602,Strange latency overhead of F.conv2d,https://github.com/pytorch/pytorch/issues/21602,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: performance",scan when time allows,2019-06-12T15:26:06Z
issue,21477,Not obvious how to install torchvision with PyTorch source build,https://github.com/pytorch/pytorch/issues/21477,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: vision,scan when time allows,2019-06-11T22:10:40Z
issue,21587,RuntimeError: cublas runtime error,https://github.com/pytorch/pytorch/issues/21587,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cublas,scan when time allows,2019-06-10T22:14:56Z
issue,3468,improve performance of common CPU clone / contiguous calls with HPTT,https://github.com/pytorch/pytorch/issues/3468,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpu,scan when time allows,2019-06-10T19:33:16Z
issue,21459,RuntimeError: cublas runtime error,https://github.com/pytorch/pytorch/issues/21459,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cublas,scan when time allows,2019-06-10T17:41:01Z
issue,18206,Rename ignore_index to ignore_target in CrossEntropyLoss,https://github.com/pytorch/pytorch/issues/18206,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2019-06-09T00:35:50Z
issue,16655,"CUDA cache doubles on the second batch and causes OOM, `empty_cache` doesn't empty it",https://github.com/pytorch/pytorch/issues/16655,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cudnn, module: dependency bug",scan when time allows,2019-06-06T18:02:14Z
issue,21405,how libtorch can work with tensor data as same as pytorch,https://github.com/pytorch/pytorch/issues/21405,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: docs",scan when time allows,2019-06-06T15:25:06Z
issue,21399,collect_env ignores conda environment,https://github.com/pytorch/pytorch/issues/21399,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: collect_env.py, module: docs",scan when time allows,2019-06-05T14:58:53Z
issue,16257,why check ArgumentInfo is_pod? suffer bugs,https://github.com/pytorch/pytorch/issues/16257,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2019-06-05T14:32:32Z
issue,21290,"Title of docs page includes ""PyTorch master documentation"" even for non-master branches.",https://github.com/pytorch/pytorch/issues/21290,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2019-06-04T18:31:15Z
issue,8913,Multiprocessing Self Test Error,https://github.com/pytorch/pytorch/issues/8913,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: multiprocessing,scan when time allows,2019-06-04T18:22:45Z
issue,21165,Getting Access to Blob/Tensor reference in jit::script::Module,https://github.com/pytorch/pytorch/issues/21165,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2019-06-04T18:22:06Z
issue,14131,running caffe2 float16 tensors results in aten runtime error,https://github.com/pytorch/pytorch/issues/14131,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-06-02T20:19:36Z
issue,21135,[cmake build] can't build pytorch with install mkl library,https://github.com/pytorch/pytorch/issues/21135,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2019-05-31T07:59:26Z
issue,9204,/usr/bin/ld: cannot find -lpthreads,https://github.com/pytorch/pytorch/issues/9204,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-05-30T10:00:51Z
issue,21050,output values not same and much slower than Python API,https://github.com/pytorch/pytorch/issues/21050,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp,scan when time allows,2019-05-30T00:10:32Z
issue,20975,"build caffe2 operators failed, 'sorry, unimplemented: non-trivial designated initializers not supported'",https://github.com/pytorch/pytorch/issues/20975,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-05-29T23:32:20Z
issue,16843,Allow to build pytorch for a *specific* architecture,https://github.com/pytorch/pytorch/issues/16843,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2019-05-29T23:29:33Z
issue,3990,Raise an error when using magma built against wrong version of cuda,https://github.com/pytorch/pytorch/issues/3990,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: build",scan when time allows,2019-05-29T23:24:27Z
issue,21012,"Error when creating new caffe2::Predictor(_initNet, _predictNet)",https://github.com/pytorch/pytorch/issues/21012,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-05-29T23:18:06Z
issue,8741,[Feature Request] Add to() method for optimizers/schedulers,https://github.com/pytorch/pytorch/issues/8741,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2019-05-29T23:00:35Z
issue,18677,How to compile/install caffe2 with cuda 9.0?,https://github.com/pytorch/pytorch/issues/18677,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-05-27T12:43:18Z
issue,20416,Class based Sampler for Class Incremental/Continual Learning research,https://github.com/pytorch/pytorch/issues/20416,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2019-05-27T05:19:48Z
issue,20531,c++ torch::nn::Sequential increments count on name errors,https://github.com/pytorch/pytorch/issues/20531,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp,scan when time allows,2019-05-26T19:42:49Z
issue,13836,"Pytorch-Caffe2 export: ""Arrays are not almost equal to 3 decimals""",https://github.com/pytorch/pytorch/issues/13836,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-05-26T06:37:24Z
issue,13664,Caffe2 install,https://github.com/pytorch/pytorch/issues/13664,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-05-24T03:05:05Z
issue,20753,cmake for Torch unusable in archlinux,https://github.com/pytorch/pytorch/issues/20753,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: dependency bug",scan when time allows,2019-05-22T02:15:11Z
issue,20776,Caffe2: executor_test.py::ExecutorGPUResNetTest fails with an AssertionError,https://github.com/pytorch/pytorch/issues/20776,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-05-21T22:08:19Z
issue,20777,"Caffe2: operator_test/instance_norm_test.py::TestInstanceNorm::test_instance_norm_gradients test case fails with ""RuntimeError: dim() called on undefined TensorError from operator""",https://github.com/pytorch/pytorch/issues/20777,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-05-21T22:07:47Z
issue,20562,Performance issue when accessing an extremely large (10GB) longtensor,https://github.com/pytorch/pytorch/issues/20562,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: cudnn",scan when time allows,2019-05-21T00:07:12Z
issue,20734,[jit] set up views for Autodiff and autograd hooks,https://github.com/pytorch/pytorch/issues/20734,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2019-05-20T23:34:38Z
issue,20641,Main page broadcasting (?) example image bug,https://github.com/pytorch/pytorch/issues/20641,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2019-05-20T19:25:48Z
issue,10232,[Caffe2] AVX not enabled for pytorch/caffe2,https://github.com/pytorch/pytorch/issues/10232,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-05-17T17:50:20Z
issue,20576,NVRTC_ERROR unknown when using self-built libtorch,https://github.com/pytorch/pytorch/issues/20576,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2019-05-17T09:39:18Z
issue,7035,[Compilation] how to disable caffe2?,https://github.com/pytorch/pytorch/issues/7035,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2019-05-16T23:18:58Z
issue,20272,Redundantly saving sizes of SavedVariables in autograd Function,https://github.com/pytorch/pytorch/issues/20272,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: bootcamp",scan when time allows,2019-05-15T19:50:00Z
issue,20481,[docs] Automatically detect docs missing in rst,https://github.com/pytorch/pytorch/issues/20481,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: tests",scan when time allows,2019-05-14T21:20:33Z
issue,20466,Lint rule to prevent direct use of #pragma omp,https://github.com/pytorch/pytorch/issues/20466,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: lint,scan when time allows,2019-05-14T21:05:22Z
issue,20412,String in tensor,https://github.com/pytorch/pytorch/issues/20412,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: internals,scan when time allows,2019-05-13T21:34:39Z
issue,10577,[BUG]: unstable happend in saving model.,https://github.com/pytorch/pytorch/issues/10577,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: serialization,scan when time allows,2019-05-13T21:30:52Z
issue,20405,CosineAnnealingLR has unexpected behavior with large step,https://github.com/pytorch/pytorch/issues/20405,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2019-05-13T18:26:43Z
issue,20375,Better documentation / molly-guards around use of multiprocessing with spawn in Jupyter/ipython notebooks,https://github.com/pytorch/pytorch/issues/20375,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2019-05-13T17:59:13Z
issue,20363,"Define portable M_PI replacement, use it instead of non-standard M_PI in math.h",https://github.com/pytorch/pytorch/issues/20363,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: internals,scan when time allows,2019-05-10T18:35:34Z
issue,20342,TensorIterator resizes output to a scalar if there are no inputs,https://github.com/pytorch/pytorch/issues/20342,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: internals,scan when time allows,2019-05-10T17:40:14Z
issue,5388,Perf regression: indexing 1-d tensor,https://github.com/pytorch/pytorch/issues/5388,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: performance,scan when time allows,2019-05-10T17:29:40Z
issue,13118,"warning: attribute namespace ""clang"" is unrecognized; High Sierra / Fedora compilation with clang results in spurious clang errors in nvcc",https://github.com/pytorch/pytorch/issues/13118,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: build warnings, module: cuda",scan when time allows,2019-05-09T15:18:20Z
issue,20183,Latex Errors when Compiling documentation to latexpdf,https://github.com/pytorch/pytorch/issues/20183,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: doc infra, module: docs",scan when time allows,2019-05-08T15:12:51Z
issue,20268,[Caffe2] Convert caffe to caffe2 with solverstate,https://github.com/pytorch/pytorch/issues/20268,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-05-08T13:48:20Z
issue,12401,AttributeError: Method VideoInput is not a registered operator,https://github.com/pytorch/pytorch/issues/12401,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-05-08T06:38:37Z
issue,19335,CUDA optimization: using `__restrict__` whenever possible,https://github.com/pytorch/pytorch/issues/19335,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2019-05-08T02:02:17Z
issue,20245,ProcessGroupMPI tests in test_c10d.py,https://github.com/pytorch/pytorch/issues/20245,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: tests, oncall: distributed",scan when time allows,2019-05-07T21:14:13Z
issue,20111,Split libtorch binary build CI job into separate variants,https://github.com/pytorch/pytorch/issues/20111,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2019-05-07T17:40:48Z
issue,20189,Unify Caffe2 and PyTorch OpenMP initialization,https://github.com/pytorch/pytorch/issues/20189,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: multithreading,scan when time allows,2019-05-07T16:48:27Z
issue,18436,UnpicklingError when trying to load multiple objects from a file,https://github.com/pytorch/pytorch/issues/18436,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: pickle, module: serialization",scan when time allows,2019-05-07T10:32:59Z
issue,20165,torch.nn.threshold cannot accept tensor as a threshold,https://github.com/pytorch/pytorch/issues/20165,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2019-05-06T21:14:34Z
issue,17495,RuntimeError: storage_.IsType<T>() ASSERT FAILED,https://github.com/pytorch/pytorch/issues/17495,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-05-06T19:31:08Z
issue,17350,"torch.nn.CrossEntropyLoss with ""reduction"" sum/mean is not deterministic on segmentation outputs / labels",https://github.com/pytorch/pytorch/issues/17350,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2019-05-06T19:30:18Z
issue,20138,Inconsistant values of lr_scheduler.get_lr and lr in optimizer.param_groups,https://github.com/pytorch/pytorch/issues/20138,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2019-05-06T16:00:07Z
issue,20086,CosineAnnealingLR giving unexpected learning rates on PyTorch 1.1.,https://github.com/pytorch/pytorch/issues/20086,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2019-05-03T21:28:24Z
issue,18000,Unsupported type of tensor c10::half when running resnet50_trainer.py in Caffe2,https://github.com/pytorch/pytorch/issues/18000,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-05-03T05:07:52Z
issue,19504,[jit] Bad error when calling `ScriptModule`s with attributes/parameters,https://github.com/pytorch/pytorch/issues/19504,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2019-05-03T00:06:39Z
issue,20022,Precision of sparse float embeddings differs from dense embeddings on CPU,https://github.com/pytorch/pytorch/issues/20022,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: numerical-reproducibility",scan when time allows,2019-05-02T16:31:38Z
issue,19825,Performance issue master (a25b79531),https://github.com/pytorch/pytorch/issues/19825,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: multithreading, module: performance",scan when time allows,2019-05-01T23:31:58Z
issue,4632,CUDNN_STATUS_INTERNAL_ERROR when training with conv3d,https://github.com/pytorch/pytorch/issues/4632,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: cudnn",scan when time allows,2019-05-01T03:30:15Z
issue,19960,"On first construction, CUDAContext changes default CPU allocator behavior",https://github.com/pytorch/pytorch/issues/19960,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-05-01T00:42:59Z
issue,19826,vectorized convert_to_int_of_same_size <int64_t> can't handle nan,https://github.com/pytorch/pytorch/issues/19826,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: error checking, module: nans and infs",scan when time allows,2019-05-01T00:18:08Z
issue,19930,caffe2/resnet50 assert in fetch_blob() when base_learning_rate = 0,https://github.com/pytorch/pytorch/issues/19930,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-04-30T16:19:34Z
issue,19408,Massive memory overhead over NumPy,https://github.com/pytorch/pytorch/issues/19408,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: internals, module: memory usage",scan when time allows,2019-04-30T11:13:53Z
issue,19263,GCP Base Image Wrong CUDA Version,https://github.com/pytorch/pytorch/issues/19263,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: doc infra,scan when time allows,2019-04-29T23:28:27Z
issue,19708,Test failure for depthwise3x3_conv,https://github.com/pytorch/pytorch/issues/19708,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-04-26T08:14:32Z
issue,19652,[docs] torch.set_flush_denormal(...) to mention default mode,https://github.com/pytorch/pytorch/issues/19652,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2019-04-24T22:12:14Z
issue,19637,"[jit] Traced {zeros,empty}()/{zeros,empty}_like() calls do not respect device args.",https://github.com/pytorch/pytorch/issues/19637,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2019-04-24T21:59:18Z
issue,8153,GRU is implementation of GRU v1 draft rather than final GRU paper algo,https://github.com/pytorch/pytorch/issues/8153,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: rnn",scan when time allows,2019-04-24T20:52:40Z
issue,19672,torch.set_flush_denormal not working on some (old) OSX machines,https://github.com/pytorch/pytorch/issues/19672,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: macos,scan when time allows,2019-04-24T20:46:08Z
issue,12498,Support calculating grad for dense in sparse @ dense,https://github.com/pytorch/pytorch/issues/12498,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2019-04-23T19:37:01Z
issue,19599,cudnn conv doesn't check batch_size > 0,https://github.com/pytorch/pytorch/issues/19599,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: cudnn, module: error checking",scan when time allows,2019-04-23T19:27:43Z
issue,19244,[Caffe2] Retraining saved model,https://github.com/pytorch/pytorch/issues/19244,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-04-22T14:00:15Z
issue,19416,[caffe2]Reloading model gives segmentation fault,https://github.com/pytorch/pytorch/issues/19416,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-04-19T21:15:20Z
issue,19499,C++ API 'nn::Sequential' has inconsistent behavior with python conterpart,https://github.com/pytorch/pytorch/issues/19499,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp,scan when time allows,2019-04-19T20:21:51Z
issue,8290,[feature request] MPI init_method for torch.distributed,https://github.com/pytorch/pytorch/issues/8290,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2019-04-19T18:58:04Z
issue,19437,The `unary_kernel` call should be pushed into CopyKernel.cpp and it should completely replace the current `copy_kernel`.,https://github.com/pytorch/pytorch/issues/19437,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpu,scan when time allows,2019-04-18T21:54:21Z
issue,19330,Multiprocessing on distributed Multi-nodes shutdown error: ‘spawn’ on slave node leads to semaphore_tracker leaked,https://github.com/pytorch/pytorch/issues/19330,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2019-04-18T19:45:58Z
issue,6083,[feature request] DC-ASGD (Delay Compensated Asynchronous Stochastic Gradient Descent),https://github.com/pytorch/pytorch/issues/6083,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: optimizer, oncall: distributed",scan when time allows,2019-04-15T17:56:57Z
issue,19229,Different behavior of torch.nn.MultiMarginLoss on CPU/GPU Tensors,https://github.com/pytorch/pytorch/issues/19229,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: error checking",scan when time allows,2019-04-14T04:14:10Z
issue,16804,Wrong description of positive class weight in BCEWithLogitsLoss,https://github.com/pytorch/pytorch/issues/16804,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2019-04-12T22:00:47Z
issue,12509,"The text design (color, type) makes it hard to read",https://github.com/pytorch/pytorch/issues/12509,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-04-11T16:27:17Z
issue,14632,[c10d] Check that allgather/gather output tensors point to different storage,https://github.com/pytorch/pytorch/issues/14632,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2019-04-11T16:20:27Z
issue,15491,LibTorch: include cmake files for all distributed headers,https://github.com/pytorch/pytorch/issues/15491,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp,scan when time allows,2019-04-11T16:18:02Z
issue,19107,pytorch/caffe2/onnx/backend.cc:1668:57: error: invalid conversion from ‘google::protobuf::int32 {aka int}’ to ‘onnx_torch::TensorProto::DataType {aka onnx_torch::TensorProto_DataType}’ [-fpermissive],https://github.com/pytorch/pytorch/issues/19107,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-04-10T22:45:53Z
issue,19120,Embedding layer does not check input range,https://github.com/pytorch/pytorch/issues/19120,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: error checking",scan when time allows,2019-04-10T20:17:33Z
issue,18933,Completion of error handling,https://github.com/pytorch/pytorch/issues/18933,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: error checking, module: internals",scan when time allows,2019-04-09T00:32:00Z
issue,6408,[caffe2] Prefetching blobs for memory optimization,https://github.com/pytorch/pytorch/issues/6408,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-04-08T21:12:47Z
issue,18920,[c10d] CUDA tests for C++ reducer,https://github.com/pytorch/pytorch/issues/18920,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2019-04-08T16:45:51Z
issue,18998,torch.from_PIL() Request ?,https://github.com/pytorch/pytorch/issues/18998,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: vision,scan when time allows,2019-04-08T16:26:10Z
issue,18944,Caffe2 on google colab:,https://github.com/pytorch/pytorch/issues/18944,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-04-06T16:33:02Z
issue,16291,Confusing documentation with distributions.Categorical about logits,https://github.com/pytorch/pytorch/issues/16291,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: distributions, module: docs",scan when time allows,2019-04-05T14:11:02Z
issue,17897,CUDA large matrix-vector product (torch.mv) causes illegal memory access,https://github.com/pytorch/pytorch/issues/17897,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: 64-bit, module: cublas, module: cuda, module: dependency bug",scan when time allows,2019-04-04T16:59:15Z
issue,15120,Different implementations of upsampleBilinear between pytorch and caffe2,https://github.com/pytorch/pytorch/issues/15120,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-04-03T23:47:30Z
issue,1369,"""Sparsified"" mathematical operations",https://github.com/pytorch/pytorch/issues/1369,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2019-04-03T20:06:33Z
issue,18658,[FR] add CPU information in collect_env.py,https://github.com/pytorch/pytorch/issues/18658,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: collect_env.py,scan when time allows,2019-04-02T22:25:43Z
issue,18692,Add build tests for feature environment vars,https://github.com/pytorch/pytorch/issues/18692,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2019-04-02T22:23:04Z
issue,6357,[feature request] Unpooling layer in Caffe2,https://github.com/pytorch/pytorch/issues/6357,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-04-01T19:23:09Z
issue,16299,caffe2::CudnnConvOp::RunOnDevice() fails on Squeezenet,https://github.com/pytorch/pytorch/issues/16299,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-04-01T01:42:09Z
issue,8886,[Caffe2] Dose caffe2 support the function similar to itersize of caffe,https://github.com/pytorch/pytorch/issues/8886,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-03-28T01:29:16Z
issue,18273,[caffe2] Broken Operators Catalog,https://github.com/pytorch/pytorch/issues/18273,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-03-25T13:06:58Z
issue,14731,Caffe2 C++ runs single threaded,https://github.com/pytorch/pytorch/issues/14731,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-03-22T00:44:48Z
issue,17936,[caffe2] resize_op_test.py::TestResize::test_nearest FAILED,https://github.com/pytorch/pytorch/issues/17936,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-03-18T19:44:13Z
issue,17800,[Caffe2] cudnn mismatch,https://github.com/pytorch/pytorch/issues/17800,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-03-15T17:20:47Z
issue,14486,"building bundled nccl fails in (caffe2, cuda 8, cudnn 7) CI environment",https://github.com/pytorch/pytorch/issues/14486,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-03-13T18:13:20Z
issue,17907,"[caffe2] resnet_trainer example fails with MNIST dataset, when parameter num_channels=1 is provided",https://github.com/pytorch/pytorch/issues/17907,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-03-12T13:23:21Z
issue,17887,Caffe2 building failure,https://github.com/pytorch/pytorch/issues/17887,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-03-11T21:32:55Z
issue,17872,Linking Caffe2 with sequential MKL but still calling multi-threaded mkl,https://github.com/pytorch/pytorch/issues/17872,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-03-11T16:48:22Z
issue,17632,Caffe2 C++ script for classification/object_detection with CMakeLists.txt,https://github.com/pytorch/pytorch/issues/17632,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-03-10T21:41:36Z
issue,17850,Build fails for caffe2/CMakeFiles/caffe2.dir/__/aten/src/ATen/native/mkldnn/Conv.cpp.o; possibly MKL-DNN issue,https://github.com/pytorch/pytorch/issues/17850,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-03-10T16:50:53Z
issue,17712,Can we to build Caffe2 custom ops independently from a Caffe2 build ?,https://github.com/pytorch/pytorch/issues/17712,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-03-08T04:26:20Z
issue,15630,[Caffe2] Internal compiler error for CUDA.,https://github.com/pytorch/pytorch/issues/15630,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-03-07T09:12:37Z
issue,17634,[Caffe2] android app crashed while linked with libcaffe2_detectron_ops.so,https://github.com/pytorch/pytorch/issues/17634,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-03-02T12:41:38Z
issue,15290,[caffe2] Installation problem on OSX,https://github.com/pytorch/pytorch/issues/15290,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-02-24T17:03:37Z
issue,17437,Automatic aggregation of a mix of sparse and dense gradients is not supported yet,https://github.com/pytorch/pytorch/issues/17437,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-02-24T04:16:24Z
issue,16666,numa_test.py fails with AssertionError,https://github.com/pytorch/pytorch/issues/16666,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-02-04T16:43:27Z
issue,16590,Benchmarking pre-trained model in Caffe2,https://github.com/pytorch/pytorch/issues/16590,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-01-31T02:52:52Z
issue,16280,caffe2: cudaHostRegister() and mbind()-related test failures on ppc64le,https://github.com/pytorch/pytorch/issues/16280,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-01-29T23:52:29Z
issue,16494,caffe2 softmaxwithloss problem,https://github.com/pytorch/pytorch/issues/16494,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-01-29T08:36:44Z
issue,16436,How to retrain the modelzoo model in Caffe2?,https://github.com/pytorch/pytorch/issues/16436,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-01-28T09:02:59Z
issue,16385,Adding new module to caffe2,https://github.com/pytorch/pytorch/issues/16385,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-01-25T22:37:57Z
issue,11354,caffe2 failed to build from source,https://github.com/pytorch/pytorch/issues/11354,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-01-24T22:07:29Z
issue,16254,I hope Caffe2's python interface add DataLoader Module Similar to pytorch's DataSet and DataLoader,https://github.com/pytorch/pytorch/issues/16254,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-01-23T09:22:26Z
issue,9704,"A newer protobuf should be used, so that cmake install will work when building for mobile",https://github.com/pytorch/pytorch/issues/9704,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-01-23T05:25:08Z
issue,6868,[Caffe2] Android NNApi integration bugs.,https://github.com/pytorch/pytorch/issues/6868,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-01-22T00:32:29Z
issue,16195,how to implement crf predict for segmentation task in Caffe2?,https://github.com/pytorch/pytorch/issues/16195,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-01-20T03:02:21Z
issue,16138,[Caffe2] resize5d stride/dim issue,https://github.com/pytorch/pytorch/issues/16138,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-01-18T04:45:17Z
issue,13598,caffe2: RuntimeError: [enforce fail at reshape_op.h:110] with Alexnet onnx test with cuda,https://github.com/pytorch/pytorch/issues/13598,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-01-16T15:51:43Z
issue,14939,caffe2/predictor missing from libtorch,https://github.com/pytorch/pytorch/issues/14939,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-01-16T10:03:35Z
issue,14989,[Caffe2] CNN Training on CPU is faster than GPU,https://github.com/pytorch/pytorch/issues/14989,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-01-14T21:11:29Z
issue,15869,RuntimeError: [enforce fail at pybind_state.cc:1111] success. Error running net train,https://github.com/pytorch/pytorch/issues/15869,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-01-14T14:08:18Z
issue,9676,"[caffe2]running problem, No module named caffe2_pybind11_state_hip",https://github.com/pytorch/pytorch/issues/9676,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-01-13T09:53:07Z
issue,15432,caffe2: multi-thread(or multi-instance) predict is much slower than the single thread or single instance,https://github.com/pytorch/pytorch/issues/15432,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-01-12T01:09:35Z
issue,14459,[Caffe2] Error protos.protos_size() == OutputSize() when loading dataset created by regular Caffe (datum),https://github.com/pytorch/pytorch/issues/14459,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-01-11T08:02:18Z
issue,14940,caffe2::Predictor: Cannot find operator schema for GivenTensorFill. Will skip schema checking.,https://github.com/pytorch/pytorch/issues/14940,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-01-11T02:02:26Z
issue,15431,IDEEP error:could not initialize a memory descriptor,https://github.com/pytorch/pytorch/issues/15431,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-01-11T00:38:28Z
issue,15771,Implicit conversion error in caffe2,https://github.com/pytorch/pytorch/issues/15771,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-01-10T17:18:35Z
issue,15465,Transplant caffe2 on my DNN-accelerator,https://github.com/pytorch/pytorch/issues/15465,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-01-10T17:16:21Z
issue,13772,Error in building Caffe2 on Windows (experimental operators),https://github.com/pytorch/pytorch/issues/13772,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-01-09T20:03:58Z
issue,11865,[Caffe2] Errors occured when running Cpp Predictor,https://github.com/pytorch/pytorch/issues/11865,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-01-08T03:03:26Z
issue,12322,[Caffe2] Segmentation fault (core dumped) while import caffe2.python.core,https://github.com/pytorch/pytorch/issues/12322,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-01-07T16:37:22Z
issue,15004,[Caffe2] How to link Caffe2 in cmake file for C++ compilation?,https://github.com/pytorch/pytorch/issues/15004,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-01-07T01:14:48Z
issue,14963,PyTorch 1.0 source build fails (error in caffe2/utils/fatal_signal_asan_no_sig_test.cc),https://github.com/pytorch/pytorch/issues/14963,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-01-02T08:45:14Z
issue,15608,[Caffe2] How to switch to test phase?,https://github.com/pytorch/pytorch/issues/15608,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-12-30T16:50:06Z
issue,9531,[caffe2]train on GPU and test on cpu failed,https://github.com/pytorch/pytorch/issues/9531,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-12-30T16:22:10Z
issue,15249,[caffe2] controlled forward pass,https://github.com/pytorch/pytorch/issues/15249,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-12-20T16:58:34Z
issue,10471,[Caffe2] Multithreading in Caffe2,https://github.com/pytorch/pytorch/issues/10471,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-12-20T06:31:54Z
issue,12530,[caffe2] Memory usage,https://github.com/pytorch/pytorch/issues/12530,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-12-20T04:28:34Z
issue,15313,AttributeError: module 'caffe2.python._import_c_extension' has no attribute 'get_cudnn_version' when Caffe2 is not built with CuDNN,https://github.com/pytorch/pytorch/issues/15313,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-12-17T22:10:25Z
issue,14925,Am I able to run caffe2 and gloo example: resnet50_trainer.py using pytorch v1.0rc1?,https://github.com/pytorch/pytorch/issues/14925,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-12-17T07:24:34Z
issue,11735,Caffe2 installation: libcaffe2_gpu.so: undefined reference to `caffe2::ClipTransformRGB,https://github.com/pytorch/pytorch/issues/11735,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-12-14T07:51:22Z
issue,12007,caffe2's installation is still problematic...,https://github.com/pytorch/pytorch/issues/12007,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-12-11T09:58:38Z
issue,15009,[Caffe2] Caffe2Config.cmake,https://github.com/pytorch/pytorch/issues/15009,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-12-10T20:06:00Z
issue,14726,[caffe2] Corresponding C++ API for prepare_prediction_net,https://github.com/pytorch/pytorch/issues/14726,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-12-10T16:23:36Z
issue,14788,[caffe2] [NNPACK] how to set thread pool for NNPACK?,https://github.com/pytorch/pytorch/issues/14788,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-12-10T16:21:18Z
issue,14799,which operator of caffe2 have the same function as torch.nn.Parameter,https://github.com/pytorch/pytorch/issues/14799,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-12-10T16:20:27Z
issue,14970,"[Caffe2] ""Same"" padding",https://github.com/pytorch/pytorch/issues/14970,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-12-09T23:01:32Z
issue,14796,Cannot build Caffe2 with TensorRT,https://github.com/pytorch/pytorch/issues/14796,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-12-05T17:51:05Z
issue,7078,Fail to import Caffe2Backend,https://github.com/pytorch/pytorch/issues/7078,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-12-04T20:45:12Z
issue,11861,[Caffe2 installation problem],https://github.com/pytorch/pytorch/issues/11861,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-12-04T12:16:59Z
issue,14652,fc_without_bias / FCWithoutBias,https://github.com/pytorch/pytorch/issues/14652,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-12-03T18:27:42Z
issue,14685,[Caffe2] Exception when creating gradient for [Cast] SquaredL2Distance as output layer of CNN network,https://github.com/pytorch/pytorch/issues/14685,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-12-03T18:21:47Z
issue,13493,How to do inference with float16 (or half) in caffe2?,https://github.com/pytorch/pytorch/issues/13493,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-11-30T10:20:48Z
issue,12913,ubuntu16.04 build from source caffe2,https://github.com/pytorch/pytorch/issues/12913,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-11-29T19:43:56Z
issue,14508,[Caffe2] GPU test passed. Cannot see on nvidia-smi,https://github.com/pytorch/pytorch/issues/14508,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-11-29T15:40:20Z
issue,14538,[Caffe2] How to fetch trainable parameters?,https://github.com/pytorch/pytorch/issues/14538,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-11-29T15:33:35Z
issue,14408,DeviceOption::set_device_type() doesn't accept caffe2::CPU anymore,https://github.com/pytorch/pytorch/issues/14408,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-11-28T09:55:44Z
issue,14431,BUILD_CAFFE2_OPS=OFF is not tested in CI,https://github.com/pytorch/pytorch/issues/14431,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-11-27T22:28:22Z
issue,14380,[caffe2] Caffe2 GlobalInit should be run before any other API calls,https://github.com/pytorch/pytorch/issues/14380,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-11-26T19:15:09Z
issue,14366,How to use model.net.Clip?,https://github.com/pytorch/pytorch/issues/14366,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-11-26T18:23:09Z
issue,14312,[Caffe2] Cannot get repeated argument in custom operator in CUDA context,https://github.com/pytorch/pytorch/issues/14312,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-11-26T18:20:23Z
issue,14185,[Caffe2] Non-spatial batchnorm optimization,https://github.com/pytorch/pytorch/issues/14185,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-11-26T18:20:21Z
issue,14222,Caffe2->ONNX conversion issue due to spatial-bn,https://github.com/pytorch/pytorch/issues/14222,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-11-26T18:08:10Z
issue,14262,[Caffe2] How can I use detectron with pytorch?,https://github.com/pytorch/pytorch/issues/14262,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-11-26T18:07:58Z
issue,14320,[caffe2] Fails to build with fbgemm enabled,https://github.com/pytorch/pytorch/issues/14320,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-11-26T18:07:36Z
issue,14328,[Caffe2] How to use the euclidean loss (L2) as output of a CNN model?,https://github.com/pytorch/pytorch/issues/14328,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-11-26T18:07:29Z
issue,14337,Caffe2 C++ tutorial is not working,https://github.com/pytorch/pytorch/issues/14337,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-11-26T18:07:27Z
issue,14348,[Caffe2] Error when loading a leveldb dataset using brew.db_input (Error protos.protos_size() == OutputSize().),https://github.com/pytorch/pytorch/issues/14348,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-11-26T18:07:22Z
issue,10940,[Caffe2] Model work Alright in Python but Failed in C++,https://github.com/pytorch/pytorch/issues/10940,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-11-23T15:44:26Z
issue,13226,[Caffe2] cmake3 detection error?,https://github.com/pytorch/pytorch/issues/13226,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-11-22T04:09:42Z
issue,8471,Does the caffe2 have the depthwise separable convolution oprater ?,https://github.com/pytorch/pytorch/issues/8471,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-11-21T09:28:44Z
issue,12515,how to use mask-rcnn in caffe2 c++ gpu,https://github.com/pytorch/pytorch/issues/12515,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-11-20T10:09:40Z
issue,14030,Cannot import caffe2_pybind11_state_gpu,https://github.com/pytorch/pytorch/issues/14030,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-11-19T18:45:10Z
issue,13715,caffe2 c++ load pb model fail,https://github.com/pytorch/pytorch/issues/13715,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-11-14T15:07:24Z
issue,13655,[caffe2] How to export onnx model trained on Detectron in Caffe2?,https://github.com/pytorch/pytorch/issues/13655,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-11-12T18:41:55Z
issue,13671,Caffe2 Build Static,https://github.com/pytorch/pytorch/issues/13671,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-11-12T18:37:35Z
issue,13667,[caffe2] Modify models in model zoo,https://github.com/pytorch/pytorch/issues/13667,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-11-12T18:37:34Z
issue,13684,[caffe2] Adding CUDA operators for generate proposals and NMS layers,https://github.com/pytorch/pytorch/issues/13684,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-11-12T18:37:13Z
issue,11791,[caffe2]Can i build caffe2 library only for cpu inference purpose and reduce the binary size?,https://github.com/pytorch/pytorch/issues/11791,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-11-10T15:17:24Z
issue,11636,undefined reference to caffe2,https://github.com/pytorch/pytorch/issues/11636,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-11-07T08:10:09Z
issue,13208,[caffe2] How to use the operators that are not included in brew through the python API?,https://github.com/pytorch/pytorch/issues/13208,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-11-06T22:14:05Z
issue,13297,[Caffe2]How to accelerate group convolution in Caffe2?,https://github.com/pytorch/pytorch/issues/13297,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-11-06T17:39:30Z
issue,13498,Caffe2 missing headers in docker/conda,https://github.com/pytorch/pytorch/issues/13498,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-11-05T18:52:45Z
issue,13373,cannot run mobilenet_v2_quantized on pytorch/caffe2,https://github.com/pytorch/pytorch/issues/13373,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-11-02T00:38:43Z
issue,12931,USE_OPENMP=OFF is ignored [Caffe2],https://github.com/pytorch/pytorch/issues/12931,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-11-01T13:56:47Z
issue,6618,[Caffe2] CUDNN_STATUS_BAD_PARAM Error with the LRN layer while trying to run the code using CUDA. The training works fine on CPU,https://github.com/pytorch/pytorch/issues/6618,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-10-31T22:48:09Z
issue,13405,[caffe2] How to use Split operator?,https://github.com/pytorch/pytorch/issues/13405,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-10-31T18:56:09Z
issue,13041,[caffe2] How to handle multiple inputs and multiple outputs in the network architecture?,https://github.com/pytorch/pytorch/issues/13041,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-10-26T17:57:36Z
issue,13120,[caffe2] incompatible constructor arguments,https://github.com/pytorch/pytorch/issues/13120,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-10-25T14:38:02Z
issue,12808,tutorial_blob ERROR,https://github.com/pytorch/pytorch/issues/12808,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-10-25T00:09:02Z
issue,12576,[feature request] `ignore_label` argument in Caffe2 `SoftmaxWithLoss`,https://github.com/pytorch/pytorch/issues/12576,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-10-23T19:39:13Z
issue,12642,fail to visualize caffe2 model,https://github.com/pytorch/pytorch/issues/12642,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-10-23T16:28:27Z
issue,12879,Eigen in Caffe2 doesn't produce vectorized instructions,https://github.com/pytorch/pytorch/issues/12879,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-10-22T19:15:19Z
issue,12755,caffe2 python custom operator could not update parameters,https://github.com/pytorch/pytorch/issues/12755,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-10-22T17:59:18Z
issue,7912,[Caffe2] Operators of Detectron module not registered/compiled when built on windows,https://github.com/pytorch/pytorch/issues/7912,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-10-20T20:49:15Z
issue,12501,C++: Calling Workspace::RunNet for a prediction on a different thread each time causes a GPU memory leak,https://github.com/pytorch/pytorch/issues/12501,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-10-18T23:12:23Z
issue,12763,"Caffe2: Two entries of external_input ""data_0"" in mnist_predict_net.pbtxt file",https://github.com/pytorch/pytorch/issues/12763,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-10-17T23:19:58Z
issue,12646,Caffe2 Installation inside Pytorch,https://github.com/pytorch/pytorch/issues/12646,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-10-17T19:49:45Z
issue,12640,how to store a bounding box in Tensor?,https://github.com/pytorch/pytorch/issues/12640,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-10-15T17:54:12Z
issue,12658,Move BigTensorSerialization tests out of default caffe2_cpu_tests,https://github.com/pytorch/pytorch/issues/12658,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-10-15T17:30:09Z
issue,12280,`GLIBC_2.23' not found on Ubuntu14.04.,https://github.com/pytorch/pytorch/issues/12280,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-10-08T17:54:51Z
issue,7127,how can i use openMP for caffe2? why caffe2 not work in multi-threads mode?,https://github.com/pytorch/pytorch/issues/7127,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-10-08T05:42:14Z
issue,6660,[caffe2] GANs,https://github.com/pytorch/pytorch/issues/6660,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-10-05T22:33:02Z
issue,12297,[caffe2] Inference using multiple GPU,https://github.com/pytorch/pytorch/issues/12297,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-10-03T21:34:50Z
issue,12257,[Caffe2] Relink error after installing Caffe2 from conda,https://github.com/pytorch/pytorch/issues/12257,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-10-02T09:27:29Z
issue,12045,Create a custom function using the functions from math_cpu.cc source code in Caffe2?,https://github.com/pytorch/pytorch/issues/12045,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-10-01T17:47:52Z
issue,12086,Caffe2 issues with using Glog without GFlags,https://github.com/pytorch/pytorch/issues/12086,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-10-01T17:43:23Z
issue,12134,[caffe2] How to set different learning rates for different layers?,https://github.com/pytorch/pytorch/issues/12134,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-10-01T17:39:39Z
issue,12181,Network surgery for transfer fails,https://github.com/pytorch/pytorch/issues/12181,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-10-01T17:26:23Z
issue,8944,[caffe2] UnicodeDecodeError when running LeNet,https://github.com/pytorch/pytorch/issues/8944,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-09-29T06:09:53Z
issue,11850,[Caffe2/Bug] Cannot enable MKL-DNN,https://github.com/pytorch/pytorch/issues/11850,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-09-27T20:21:56Z
issue,12136,[Caffe2] GAN No Gradients in Generator,https://github.com/pytorch/pytorch/issues/12136,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-09-27T13:37:39Z
issue,10675,GenerateProposals CUDA implementation,https://github.com/pytorch/pytorch/issues/10675,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-09-26T10:25:15Z
issue,11885,[caffe2] Bug for softmaxwithloss operator,https://github.com/pytorch/pytorch/issues/11885,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-09-24T17:51:07Z
issue,11890,caffe2 argmax and argmin documentation incorrect for output type,https://github.com/pytorch/pytorch/issues/11890,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-09-24T17:49:35Z
issue,11996,Test the Caffe2 Installation with GPU erro,https://github.com/pytorch/pytorch/issues/11996,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-09-24T17:25:54Z
issue,12010,Caffe2 compiled with MKLDNN doesn't have device_type = MKLDNN,https://github.com/pytorch/pytorch/issues/12010,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-09-24T17:19:11Z
issue,9975,python caffe2/python/operator_test/activation_ops_test.py Segmentation fault (core dumped),https://github.com/pytorch/pytorch/issues/9975,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-09-24T09:59:39Z
issue,11976,[caffe2] Documentation of Optimizers (Python API),https://github.com/pytorch/pytorch/issues/11976,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-09-22T15:57:14Z
issue,10875,how can i run two model in a simple project?,https://github.com/pytorch/pytorch/issues/10875,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-09-18T12:59:21Z
issue,11678,[caffe2] adam_op implementation is incorrect.,https://github.com/pytorch/pytorch/issues/11678,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-09-14T00:32:48Z
issue,10254,"[Caffe2] I don't knw whether GPU support is right, is there something wrong with Hypothesis?",https://github.com/pytorch/pytorch/issues/10254,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-09-11T02:38:45Z
issue,11363,Hypothesis operator tests in Caffe2 generate too many warnings,https://github.com/pytorch/pytorch/issues/11363,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-09-10T22:13:18Z
issue,11433,Assorted issues in Caffe2's Metal ops,https://github.com/pytorch/pytorch/issues/11433,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-09-10T17:24:59Z
issue,9708,"[caffe2]mpirun multi-node multi GPU in Distributed mode ,run resnet50_trainer.py get RuntimeError",https://github.com/pytorch/pytorch/issues/9708,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-09-10T13:06:41Z
issue,9484,WARNING:root:This caffe2 python run does not have GPU support. Will run in CPU only mode. WARNING:root:Debug message: libcurand.so.9.0: cannot open shared object file: No such file or directory Segmentation fault (core dumped),https://github.com/pytorch/pytorch/issues/9484,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-09-09T15:57:08Z
issue,11267,How to simulate multi-node using single-node with 8 GPUs,https://github.com/pytorch/pytorch/issues/11267,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-09-07T03:24:05Z
issue,11268,How to re-shuffle lmdb per epoch in the caffe2 training process,https://github.com/pytorch/pytorch/issues/11268,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-09-07T03:23:51Z
issue,10984,[caffe2] Implement Gather for any value of `axis`,https://github.com/pytorch/pytorch/issues/10984,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-09-04T16:52:26Z
issue,10990,[Caffe2] [Question] How to disable the bias parameter?,https://github.com/pytorch/pytorch/issues/10990,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-09-04T16:52:22Z
issue,11058,A bug in roi_align.cc? [Caffe2],https://github.com/pytorch/pytorch/issues/11058,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-09-04T16:49:15Z
issue,11180,A Caffe2 Implementation of Pose Estimation,https://github.com/pytorch/pytorch/issues/11180,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-09-04T16:33:01Z
issue,11197,[Caffe2] How to correctly add Caffe2 libraries to Visual Studio to write C++ programs?,https://github.com/pytorch/pytorch/issues/11197,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-09-03T08:02:18Z
issue,11119,[Caffe2] Error C2492: data with thread storage duration may not have dll interface,https://github.com/pytorch/pytorch/issues/11119,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-08-31T10:27:43Z
issue,10991,Error when building,https://github.com/pytorch/pytorch/issues/10991,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-08-31T01:56:45Z
issue,10714,[feature request][caffe2] extend FC/FCTranspose op to handle 2d bias.,https://github.com/pytorch/pytorch/issues/10714,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-08-30T04:17:30Z
issue,10985,Where could I see all of the caffe2 operators and their arguments in python API?,https://github.com/pytorch/pytorch/issues/10985,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-08-29T07:48:50Z
issue,10902,LNK2019 error when linking with MSVC [Caffe2],https://github.com/pytorch/pytorch/issues/10902,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-08-28T18:28:02Z
issue,10808,[Feature Request] Crop op or ConvTranspose with output_shape,https://github.com/pytorch/pytorch/issues/10808,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-08-23T05:43:59Z
issue,7946,how to use Softmax when do segmentation,https://github.com/pytorch/pytorch/issues/7946,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-08-23T02:41:34Z
issue,10738,[caffe2] the caffe2 operators document is too old,https://github.com/pytorch/pytorch/issues/10738,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-08-22T17:25:01Z
issue,7047,"""Undefined symbols for architecture arm64"" when linking libcaffe2.a for iOS",https://github.com/pytorch/pytorch/issues/7047,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-08-21T22:13:24Z
issue,10582,[Caffe2] Unable to use MPI rendezvous in Caffe2,https://github.com/pytorch/pytorch/issues/10582,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-08-20T17:47:35Z
issue,7789,detectron net create error,https://github.com/pytorch/pytorch/issues/7789,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-08-20T01:54:53Z
issue,10341,[Caffe2] Which document should we follow to install caffe2?,https://github.com/pytorch/pytorch/issues/10341,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-08-19T17:22:27Z
issue,10515,[Caffe2] Segmentation faults in multithreading Caffe2,https://github.com/pytorch/pytorch/issues/10515,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-08-17T15:28:32Z
issue,10161,error when building caffe2 from source with gcc 7 & cuda gcc 5.5,https://github.com/pytorch/pytorch/issues/10161,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-08-16T06:57:09Z
issue,10314,[Caffe2] ConvPoolOpBase operator,https://github.com/pytorch/pytorch/issues/10314,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-08-13T17:58:36Z
issue,10342,[Caffe2] which kind of database can I read from caffe2,https://github.com/pytorch/pytorch/issues/10342,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-08-13T17:57:10Z
issue,10461,Key already registered. Offending key: caffe2_print_stacktraces.,https://github.com/pytorch/pytorch/issues/10461,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-08-13T17:29:49Z
issue,10460,Type name float registered twice. This should not happen. Do you have duplicated CAFFE_KNOWN_TYPE?,https://github.com/pytorch/pytorch/issues/10460,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-08-13T17:25:03Z
issue,10215,[Caffe2] Build failure on Ubuntu 16.04,https://github.com/pytorch/pytorch/issues/10215,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-08-13T16:21:53Z
issue,8982,parallelize_bmuf_distributed_test intermittently hangs,https://github.com/pytorch/pytorch/issues/8982,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-08-13T15:07:47Z
issue,10070,UnicodeDecodeError while loading caffe2 model,https://github.com/pytorch/pytorch/issues/10070,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-08-07T08:36:36Z
issue,9761,[caffe2] Questions about conv_transpose && ConvTranspose,https://github.com/pytorch/pytorch/issues/9761,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-08-06T18:01:48Z
issue,10176,[Caffe2] Modelling using Ops instead of Helper Functions,https://github.com/pytorch/pytorch/issues/10176,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-08-06T17:35:30Z
issue,10256,[Caffe 2] How to access/remove Blobs from Workspace?,https://github.com/pytorch/pytorch/issues/10256,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-08-06T17:20:48Z
issue,10007,Upgrade Caffe2 Hypothesis,https://github.com/pytorch/pytorch/issues/10007,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-08-06T12:33:43Z
issue,9016,SimpleMetaNetDefInitializer intermittently hangs,https://github.com/pytorch/pytorch/issues/9016,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-08-04T18:08:02Z
issue,10170,build error,https://github.com/pytorch/pytorch/issues/10170,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-08-03T18:28:23Z
issue,9921,A serious problem when installing caffe2. Can anyone help me?,https://github.com/pytorch/pytorch/issues/9921,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-08-03T09:37:47Z
issue,10088,[Caffe2] Implementing CoordConv layer,https://github.com/pytorch/pytorch/issues/10088,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-08-02T16:23:19Z
issue,9853,TestSequenceOps.test_gather_padding failing,https://github.com/pytorch/pytorch/issues/9853,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-07-30T17:59:04Z
issue,9854,TestConvolution.test_conv_separate_stride_pad_gradients failing,https://github.com/pytorch/pytorch/issues/9854,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-07-30T17:58:57Z
issue,9881,cannot open caffe2.lib in VS2015 debug mode,https://github.com/pytorch/pytorch/issues/9881,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-07-30T17:57:00Z
issue,9883,NetTest.OperatorWithExecutorHelper intermittently hangs,https://github.com/pytorch/pytorch/issues/9883,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-07-30T17:56:46Z
issue,9945,WERROR=1 doesn't work with FULL_CAFFE2,https://github.com/pytorch/pytorch/issues/9945,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-07-30T17:46:27Z
issue,9150,[caffe2] Drop connections,https://github.com/pytorch/pytorch/issues/9150,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-07-23T20:42:37Z
issue,9028,"[caffe2]build problem, can not find caffe2_pybind11_state_hip",https://github.com/pytorch/pytorch/issues/9028,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-07-21T06:57:38Z
issue,9449,PredictorTest.SimpleBatchSizedMapInput intermittently hangs,https://github.com/pytorch/pytorch/issues/9449,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-07-20T01:08:35Z
issue,9166,nvcc fatal : A single input file is required for a non-link phase when an outputfile is specified,https://github.com/pytorch/pytorch/issues/9166,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-07-16T17:10:03Z
issue,7571,"the latest version of caffe2, Which file is the function loadToNCHW()",https://github.com/pytorch/pytorch/issues/7571,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-07-16T14:03:40Z
issue,9218,[Caffe2] Cannot load caffe2.python. Error: libcaffe2.so: cannot open shared object file: No such file or directory,https://github.com/pytorch/pytorch/issues/9218,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-07-14T15:07:36Z
issue,9207,Where is the include and lib path for caffe2?,https://github.com/pytorch/pytorch/issues/9207,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-07-14T03:58:25Z
issue,9174,[Caffe2] build_ios.sh => 'is only available on iOS 11 or newer',https://github.com/pytorch/pytorch/issues/9174,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-07-11T17:57:43Z
issue,9024,cmake error:Could NOT find IDEEP (missing: /usr/local/include),https://github.com/pytorch/pytorch/issues/9024,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-07-09T16:56:34Z
issue,8984,[Caffe2] segmentation fault,https://github.com/pytorch/pytorch/issues/8984,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-07-03T19:09:05Z
issue,9003,cmake error,https://github.com/pytorch/pytorch/issues/9003,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-07-03T19:02:13Z
issue,8979,Check failed: error == cudaSuccess unspecified launch failure [caffe2],https://github.com/pytorch/pytorch/issues/8979,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-07-02T17:55:20Z
issue,9068,Deployment for ios for 1.0,https://github.com/pytorch/pytorch/issues/9068,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-07-02T17:34:49Z
issue,9098,[caffe2] how to substract the mean value ?,https://github.com/pytorch/pytorch/issues/9098,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-07-02T17:29:04Z
issue,8730,[caffe2] AffineChannelOp,https://github.com/pytorch/pytorch/issues/8730,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-07-01T07:26:15Z
issue,7781,[Caffe2] Fail to build after upgrading to cuda 9.2,https://github.com/pytorch/pytorch/issues/7781,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-06-28T17:36:52Z
issue,8012,[Caffe2] VideoInput/LMDB Reader to Standalone Predictor Question,https://github.com/pytorch/pytorch/issues/8012,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-06-27T03:03:10Z
issue,8902,[Caffe 2] Does caffe2 support Galaxy 5S with Android 6?,https://github.com/pytorch/pytorch/issues/8902,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-06-26T18:46:52Z
issue,8650,Cannot use IterOp at runtime CUDA&CPU,https://github.com/pytorch/pytorch/issues/8650,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-06-20T11:16:10Z
issue,8648,Cannot allocate memory Error from operator,https://github.com/pytorch/pytorch/issues/8648,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-06-19T16:04:18Z
issue,8479,An error occurred while creating a new notebook.,https://github.com/pytorch/pytorch/issues/8479,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-06-18T17:40:45Z
issue,8548,[caffe2] [feature request]Doese caffe2 support conv_nd with group?,https://github.com/pytorch/pytorch/issues/8548,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-06-18T17:35:50Z
issue,8607,[Caffe2] Wrong prediction with simple FF,https://github.com/pytorch/pytorch/issues/8607,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-06-18T17:17:58Z
issue,8546,Q: how to generate my own pb files to C++ Predictor,https://github.com/pytorch/pytorch/issues/8546,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-06-15T05:25:06Z
issue,8128,Cmake is getting permission denied when installed system wide,https://github.com/pytorch/pytorch/issues/8128,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-06-12T18:26:30Z
issue,8186,[Caffe2] Successive in-place operators cause RuntimeError of gradient operator versions,https://github.com/pytorch/pytorch/issues/8186,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-06-11T14:09:27Z
issue,8328,how can I fetch the result of output's Blobs_namet after I sum two numpy array via brew.sum,https://github.com/pytorch/pytorch/issues/8328,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-06-11T13:56:42Z
issue,7986,[Caffe2] place some nodes of the same network on GPU and others on CPU?,https://github.com/pytorch/pytorch/issues/7986,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-06-07T12:40:49Z
issue,6901,[Caffe2] [feature request] Dilations in grouped convolutons,https://github.com/pytorch/pytorch/issues/6901,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-06-01T13:25:42Z
issue,7945,[Caffe2]How to convert Caffe's mean file to Caffe2's ?,https://github.com/pytorch/pytorch/issues/7945,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-05-31T03:17:41Z
issue,7897,"[caffe2] build from source, cannot find my cudnn7.1",https://github.com/pytorch/pytorch/issues/7897,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-05-29T18:22:35Z
issue,6846,Jetson TX1 Caffe2 installation fails on latest JetPack release,https://github.com/pytorch/pytorch/issues/6846,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-05-22T08:40:50Z
issue,7729,[Caffe2][Caffe] Caffe to Caffe2 ParseError,https://github.com/pytorch/pytorch/issues/7729,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-05-21T13:57:17Z
issue,7493,Build from source with anaconda [caffe2],https://github.com/pytorch/pytorch/issues/7493,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-05-14T18:19:58Z
issue,7480,[caffe2] How to use multiple CPUs?,https://github.com/pytorch/pytorch/issues/7480,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-05-14T14:01:22Z
issue,7227,[Caffe2]Install problem,https://github.com/pytorch/pytorch/issues/7227,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-05-10T22:08:57Z
issue,6552,[Caffe2] - Run on different Cuda streams,https://github.com/pytorch/pytorch/issues/6552,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-05-07T17:30:51Z
issue,7133,[Caffe2 warpctc] How to use the offered warpctc?,https://github.com/pytorch/pytorch/issues/7133,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-05-03T03:51:39Z
issue,7179,[caffe2]when i do CMAKE,https://github.com/pytorch/pytorch/issues/7179,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-05-03T01:44:59Z
issue,6273,Multithreading Scaling Issue with MKL,https://github.com/pytorch/pytorch/issues/6273,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-05-01T02:28:44Z
issue,6902,[Caffe2] Flatten Layer in caffe2,https://github.com/pytorch/pytorch/issues/6902,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-04-26T17:30:44Z
issue,6875,[Bug] VS reports unresolved external symbol from caffe2 observer,https://github.com/pytorch/pytorch/issues/6875,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-04-26T16:25:52Z
issue,6783,[caffe2] IfOp,https://github.com/pytorch/pytorch/issues/6783,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-04-25T19:49:28Z
issue,6845,[caffe2] global average pool in caffe2,https://github.com/pytorch/pytorch/issues/6845,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-04-25T19:48:49Z
issue,6857,[caffe2] Understanding WorkFlow for Training and Testing,https://github.com/pytorch/pytorch/issues/6857,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-04-24T08:29:50Z
issue,6542,"【Train issue with caffe2 detectron】Aborted at 1523501813 (unix time) try ""date -d @1523501813"" if you are using GNU date *** PC: @ 0x7fa35103733a (unknown)",https://github.com/pytorch/pytorch/issues/6542,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-04-16T17:28:45Z